early - global journals · 2012. 9. 11. · iii. users of e-learning all users of ict-based tools...

The Early View of

“Global Journal of Computer Science and Technology”

In case of any minor updation/modification/correction, kindly inform within

3 working days after you have received this.

Kindly note, the Research papers may be removed, added, or altered according to the final status.

Early

View

Contents of the Volume

i. Copyright Notice ii. Editorial Board Members iii. Chief Author and Dean iv. Table of Contents v. From the Chief Editor’s Desk vi. Research and Review Papers

1. Users’ Training: The Predictor of Successful eLearning in HEIs 1-8 2. Performance Comparison of Radial Basis Function Networks and

Probabilistic Neural Networks for Telugu Character Recognition 9-16 3. Adaptive Routing Based on Delay Trusted Routing in Adhoc Network

17-22 4. Multiple Feasible Paths in Ant Colony Algorithm for mobile Ad-hoc

Networks with Minimum Overhead. 23-28 5. An Adaptive Fuzzy Switching Filter for Images Corrupted by Impulse

Noise. 29-34 6. An Effective Block Weightage Based Technique for Iris Recognition

Using Empirical Mode Decomposition. 35-46 7. Wavelets, its Application and Technique in signal and image

processing. 47-58 8. Dominating Set & Clustering based Network Coverage for Huge

Wireless Sensor Networks. 59-66 9. A Survey on Web Usage Mining 67-72

10. Spectrum of Effective Security Trust Architecture to Manage the Interception of Packet Transmission in Value Added Networks 73-77

vii. Auxiliary Memberships viii. Process of Submission of Research Paper ix. Preferred Author Guidelines x. Index

Early

View

©2011 Global Journals Inc. (US)

Users’ Training: The Predictor of Successful eLearning in HEIs

Allah Nawaz

Abstract- Research reveals over and over that the successful development and use of eLearning systems in higher education institutions (HEIs) are squarely anchored on the roles of users in the development and use phases of an eLearning project. At the development level, researchers suggest that all decisions and implementation must be user-centric by constantly scanning the diversity of ever-changing user-needs. While use level requires effective user-training and then sustained technical support that is available 24/7. However, user training is central issue for the project and organizational (university) management in terms of its contents, processes and follow-up. This paper aims at unfolding the nature and implications of user-training in the background of eLearning practices in HEIs particularly, in developing countries like Pakistan. Extensive literature have been surveyed to bring together diverse ideas, findings and comments of researchers about the nature, problems and solutions of user-training in the background of higher education thereby reducing it into a theoretical model of user-training. Keywords: eLearning, HEIs, ICTs, eProjects, eUsers, eTraining, eTeachers, eStudents, Net-Genres,

I. INTRODUCITON The development of innovative competencies in

eLearning is rapidly surfacing as the key issue for teacher training (Gray et al., 2003). Within universities, the implementation of eLearning is difficult for many reasons including the hesitance of faculty and staff members: decision makers and academics to change (Loing, 2005; Qureshi et al., 2009). Likewise, researchers have documented that many eLearning projects fail due to many reasons but particularly, the lack of adequate training to support the program (Wells, 2007; Nawaz et al., 2007; Nawaz & Kundi, 2010b).

Furthermore, technology means nothing if it is not used (Mujahid, 2002) but use depends on the users’ motivation towards eLearning (Lynch et al., 2005). For example, people need word processing not to `survive rather to command over the efficient ways of sharing information about livelihoods and employment. Information and Communication Technologies (ICTs) for human development are not about technology, but about people using the technology (Hameed, 2007). Similarly, teachers and students expect better support for lectures, a better access to databases, better support for research, better connectivity with the rest of the world but these high expectations are reported to be in contrast with reality (Vrana, 2007; Nawaz & Kundi, 2010c).

Depending on the theoretical model used by the developers and users, instrumental (ICTs as a tool) and/or substitutive (ICTs as a change-agent) roles of eLearning are available however; both models emphasize the role of eLearning-users (Young, 2003). Instrumentalists contend that technology is neutral and therefore its impacts and benefits entirely depend on how are they harnessed and used by teacher, student and administrators (Macleod, 2005). The substantive theorists accentuate that instrumental view is an underestimation and they can be used more intellectually and intuitively thereby changing the lifestyle of the society (Ezer, 2006). However, it is notable that no matter whether instrumental or substantive view is upheld, the success of eLearning squarely depends on the quality of “eTraining (Blázquez & Díaz, 2006)” available for teachers, students, and administrators.

Thus, the future of technology in higher education depends on the training of particularly, teachers because it is these teachers who prepare the students as well as administrators to use digital tools (Oh & French, 2004). The adoption of ICTs is a lifelong learning process however, for immediate uses particularly in organizations like universities, the users are supposed to quickly learn using new technologies. So, training is a narrow term than education that aims at preparing a learner for a particular job, function, or profession. Education refers to a long term learning process with high level objectives of developing moral, cultural, social and intellectual dimensions of an individual and society (Drinkwater et al., 2004; Kundi & Nawaz. 2010).

II. E-LEARNING IN HEIS & E-TRAINING OF E-USERS

Traditionally, transmissive modes of learning were popular, however, now there are shifts from content-centered to competency-based curricula as well as departures from teacher-centered to student-centered pedagogy in which students drive the learning process (Oliver, 2002). ICTs and particularly the educational technologies (ETS) provide complete support to the innovations of eLearning (Dinevski & Kokol, 2005) for example, its tools are usable in any learning situation including face-to-face, blended or hybrid courses, or virtual learning (Abrami et al., 2006). eLearning can be delivered either through self-managed

Global Journal of Computer Science and Technology Volume XI Issue IV Version I

1

March

201

1

Early

View

Users Training: The Predictor of Successful eLearning in HEIs


(asynchronous - offline) and teacher-led (synchronous - online). In asynchronous system, teacher and student are not required to be physically present at the time of communication rather programs are saved on the network, which is accessible at anytime from anywhere. Asynchronous learning is globally accessible, easily maintainable, platform-independent, quickly updatable and entertains a diversity of “learning styles” of the users (Manochehr, 2007).”

The concerns about eLearning practices in HEIs include debates over the best means of integrating technology into teacher-training and preparing them to replicate the same in the classrooms (Oh & French, 2004). A large body of literature supports the idea that technology training is the major factor that could help teachers develop positive attitudes toward technology and its integration into curriculum (Zhao & Bryant, 2006). Recent studies on educational technology confirm the necessity of educating teacher candidates in technology-integration into the curriculum as well as the inadequacy of existing education programs (Willis, 2006). Teachers must be kept fully abreast of the new perspectives on learning theories in general and particularly in their area of specialization (Haddad & Jurich, 2006).

III. USERS OF E-LEARNING All users of ICT-based tools use computers

however, their use varies from one group to another due to diversity of their functions and their personal attributes. Similarly, nature and extent of use is different under traditional computer-based learning, blended learning and virtual learning facilities (Sanyal, 2001; UNESCO, 2004). Teachers are pushed to adopt technology by media, government, educational institutions, professional associations, parents and society at large, but it can be counterproductive therefore, there is need to understand the teacher perceptions of ICTs and their integration into pedagogy and thereby develop training programs accordingly (Zhao & Bryant, 2006). Researchers have found that most of the educators prefer informal learning-methods than the formal courses of eTraining (Davey & Tatnall, 2007; Kundi & Nawaz, 2010).

The new technologies like Internet, web-based applications, and Web 2.0 products – all are reengineering the pedagogic and learning theories and practices. There are shifts from objectivism to constructivism in teaching and learning (Young, 2003), technocratic to reformist and holist paradigms in eLearning development and use (Aviram & Tami, 2004), and from instrumental uses of ETS to the substantive applications in the education (Mehra & Mital, 2007; Kundi & Nawaz, 2010; Nawaz & Kundi, 2010c).

1) Teachers

eLearning systems create challenges for the teachers and demands greater preparedness by possessing a wider repertoire of new teaching styles and techniques (UQA, 2001). An eTeacher has to play the roles of a mentor, coach/facilitator as well as perform the following functions:

1. Managerial: The teacher has to plan the teaching programs including objectives, timetable, rules and procedures, course-contents and deciding about the interactive activities.

2. Intellectual: This refers to the fact that teacher knows the syllabus and subject behind it.

3. Social: The teacher creates supportive learning environment, interacts with students and examines their feedback. To perform this function, the eTeacher should motivate, facilitate and encourage the students to use new digital tools (Blázquez & Díaz, 2006). In eLearning, five types of teacher-users have

been identified: builders of eLearning tools, tool-users, tool-adapters, tool-abiders and those who are indifferent to the use of computers (Johnson et al., 2006). They further suggest that universities must develop a large body of tool users. Then motivate some creative faculty members to perform as adapters and give them incentives and support from the highest levels of administration. The most important type of teacher users is the ‘tool adapters’, who are skilled users and can adapt it according to the teaching styles of the faculty. Tool adapters must be those who enjoy teaching and not intimidated by technology.

The research indicates that decisions made by teachers about the use of computers in their classrooms are influenced by multiple factors including the accessibility of hardware and relevant software, the nature of the curriculum, personal capabilities and teachers' beliefs in their capacity to work effectively with technology are a significant factor in determining patterns of classroom computer use (Albion, 1999). Furthermore, teachers’ fear of being replaced by technology or losing their authority in the classroom as the learning process becomes more learner-centered. These apprehensions can only be alleviated if teachers n understand and appreciate their changing roles in education (Tinio, 2002).

2) Students

Computers are regarded as beneficial to the students not because these machines can create a better form of learning but mainly because the knowledge and skills needed to operate the new tools are essential for working in new dot.com organizations. The ability to work with this new technology is perceived as an asset for the future success of their pupils


2

March

201

1

Early

View


(Sasseville, 2004). Even according to researchers, student manipulation of technology in achieving the goals of education is preferable to teacher manipulation of technology (Abrami et al., 2006). The challenge of evolving pedagogy to meet the needs of Net-savvy students is daunting, but educators are assisted by the fact that although these students learn in a different way than their predecessors did, but they do want to learn (Barnes et al., 2007).

Contemporary eStudents are denoted by several concepts to express their involvement with ICTs: Computer Geeks/Nerds (Thomas & Allen, 2006); Net-Generation, Net Geners, and Net-Savvy students (Barnes et al., 2007); as well as Millennials & Electronic Natives (Garcia & Qin, 2007). Instead of learning from computers, students can learn with computers in new constructivist environments (Young, 2003). Given that most students can access (almost anytime and from anywhere) various forms of information technology - MP3, cell phones, PDAs (Aaron et al., 2004), it is obvious that the Net Generation is different from the previous generations in terms of their technological abilities, teamwork abilities, and openness to participatory learning (Garcia & Qin, 2007).

3) Administrators/Staff

The actual ICT use fosters logistics and administrative processes, distribution of materials and communication about instructional issues (Valcke, 2004). ICT has had more impact on administrative services (e.g. admissions, registration, fee payment, purchasing) than on the pedagogic fundamentals of the classroom (Dalsgaard, 2006). Likewise, ICTs are also facilitating in organizational learning through improved forms of communication and sharing (Laffey & Musser, 2006). Usually, administration (or management) provides the original momentum to create an IT committee and will be responsible for charging the group with its mission. High-quality IT literacy teaching requires the administration to provide support for faculty by adequately funding the staffing of IT services personnel to levels that can accommodate the demands placed upon them (Ezziane, 2007).

Top management support defines the success or failure of any project. For ICT integration programs to be effective and sustainable, administrators must have a broad understanding of the technical, curricular, administrative, financial, and social dimensions of ICT in education (Tinio, 2002). The ‘yes’ from senior administrative level ensures the successful implementation of the strategic plan for educational technology (Stockley, 2004) however, university administrators and ICT-departments try to provide the resources for technology integration in isolation from the teachners (Juniu, 2005). Administrators must balance the needs of all stakeholders (Abrami et al., 2006).

IV. MODELLING THE E-TRAINING FOR HEIS

The design and development of eLearning is not simply a matter of selecting a technology and a team of content and instructional experts, it also includes choosing educationalists with pedagogical and ICT skills required to handle online learning (McPherson & Nunes, 2004). The technology-integration should not be based on technologically deterministic approach rather founded on broader social, cultural, political and economic factors (Macleod, 2005). In India, for example, most ICT education is ineffective because it is too technical and not at all concerned with local contexts and real world problems (Ezer, 2006). There is also increasing acknowledgement that it is not just technical skills needed by the eLearning developers rather soft skills’ are more critical (Jewels & Ford, 2006; Nawaz & Kundi, 2010b, 2010c).

Research tells that the ideal method for developing teachers' self-efficacy is effective training and support to work with computers in the classrooms (Albion, 1999). Educators are need resources, teaching techniques, greater cultural sensitivity, and ability to adjust with new teaching and learning structures (UQA, 2001). Likewise, effective teaching strategies & pedagogy, appropriate curriculum, faculty development and consistent updating are the most important considerations in teacher education (Oh & French, 2004). In the eLearning environments, eTeacher works as a mentor, coach or facilitator and is expected to perform managerial, intellectual and social functions with the help of modern technologies, which definitely demands continuous teacher-training (Blázquez & Díaz, 2006; Nawaz & Kundi, 2010c).

Similarly, the students with no computer-background, like those from natural sciences and social sciences need training in those tools which are needed in their own field of learning. This training is mostly conducted by the computer-personnel (Ezer, 2006). However, research shows that such trainers fall short of educating the students in how to use computers in a particular field of study except the general uses of the technology. Researchers have therefore suggested to use non-computer training personnel for the purpose of preparing non-computer students in practical use of computers in the real world (Gray et al., 2003; Blázquez & Díaz, 2006; Nawaz & Kundi, 2010c).

Thus, both the decision-making and implementation staff has to understand ICTs. Decision makers’ knowledge of computers and related technologies definitely help in making real-world decisions (Afghan, 2000). In most of the universities, administrators and administrative staff is given training in the use of computers for performing administrative functions like office automation tools particularly MS-Office (Marcella & Knox 2004) however, in the advanced



3

March

201

1

Early

View


countries, administrative staff is also trained in using EMIS, EDSS, LMS, CMS, and other eLearning software (UNESCO, 2006). In developing countries, there is still need to train administrators in the basic and preliminary use of computers in automating the routine administrative functions in an educational institution (Mehra & Mital, 2007). Administrative staff handles data about the university resources, operations, results, projects and correspondence with the external institutions (Wikipedia, 2009).

1) Continuous Users’ Need/Problems Analysis

Recent research shows that technology properly deployed in the classroom can make the learning process more interactive and enjoyable if curriculum is customized to learners' needs and personal interests (Radosevich & Kahn, 2006). The multiplicity of perceptions about the nature and role of ICTs in HEIs can be grouped into two broader views. Each of these views determines the contents for eTraining.

1. Instrumental View: It is the most popular belief, which views technology as a ‘tool’ without any inherent value rather its value lies in its use so a single digital model fits every situation (Macleod, 2005; Radosevich & Kahn, 2006). Instrumental education is based on the argument that education serves society therefore emphasis is on relevance and utility of education. The risk of this approach is that students simply meet some identified need, rather than think critically with the purpose of achieving broader intellectual advancement (Ezer, 2006).

2. Substantive Role: This is a determinist or autonomous approach which argues that technology is not neutral rather exerts positive or negative impacts. Technological determinism encourages the idea that: the mere presence of technology leads to familiar and standard applications, which in turn bring about social change (Macleod, 2005; Radosevich & Kahn, 2006). The substantive theory matches with the ‘liberal theory’ of education (Ezer, 2006), which views learning not as a mere recollection of facts rather an interconnected experience. Results show that promoters of technology view

ICTs as a way of transforming education (substantive-approach) whereas most of the teachers view it only as a means to an end (instrumental conception). The advocates of technology base their vision on broader social changes; the other group considers only the student-requirements and the practical ways to meet them (Sasseville, 2004) therefore, the developers must balance the needs of all stakeholders (Abrami et al., (2006) by getting academic computing staff, faculty, and

administrators together (Kopyc, 2007; Nawaz & Kundi, 2010c).

Figure 1 Schematic Diagram of the Theoretical Model for eTraining

The above figure gives a visual version of the essence analyzed in this publication. The numbers used in the model represent the following hypotheses. These hypotheses have mostly been empirically validated by the researcher while remaining hypotheses are under process.

1.

First arrow shows that the success of eLearning in higher education is dependant on the digital literacy (Nawaz & Kundi, 2010c) and personal attributes of the teachers, students and administrators (Nawaz & Kundi, 2010a).

2.

The second arrow hints that there are problems relating to ICTs, use and users, which interfere with the relationship of users and eLearning (Qureshi et al., 2009; Nawaz & Kundi, 2010b).

3.

An effective and powerful training program can help reducing impacts of the problems (arrow 2) however; it will work through changing the mindset of users by helping them in departing from objectivism to constructivism (Kundi & Nawaz, 2010) in the use of eLearning systems.

4.

Fourth arrow tells that eTraining will change the users psychologically, intellectually and thus, in practice as well.

5.

Researchers have identified problems with eProjects relating to the development, use (Qureshi et al., 2009) and user demographics (Nawaz & Kundi, 2010a). Thus users can add to the problems as well as get affected by the problems (arrow 5).

6.

The sixth arrow says that problems of eLearning do affect the successful operations of the system.

7.

Finally, eTraining aims at strengthening the relationship between users and eLearning (arrow 3) however it operates through the path of arrows 4, 5, & 6.

Users Training: The Predictor of Successful eLearning in HEIs Global Journal of Computer Science and Technology Volume XI Issue IV Version I

4

March

201

1

Early

View


V. DISCUSSIONS The research reveals that contemporary teacher

training does not match the educational needs partly because administrators and technologists disallow faculty in the decisions about the design and development of technology-integration (Juniu, 2005). For example, there is no prescribed national syllabus for ICTs for teacher training in UK however, Ghana has a standard curriculum for ICTs in initial teacher training (Cawson, 2005). Anyhow, teachers need that kind of eTraining, which can be reproduced in the classrooms and not a training which makes them expert in merely using one or another software application or digital gadget (Willis, 2006).

Besides, emotional and behavioral aspects of attitude, the ‘informational component’ is on the top in the sense that it creates the belief and perceptions of the person, therefore sets forth the foundation for practical attitude. Given this, attitudes can be changed by providing correct, complete and timely information to the users about ICTs, educational technologies, eLearning development and use practices and benefits for the user (Luthans, 2005:124). There is need to change the roles of both teachers and learners. The eTeacher is no more a ‘sage on the stage’ rather a ‘guide on the side’ in the new learning environments. Likewise, an eStudent is no more passive receiver of contents rather collaborating partners in the learning process (Kundi & Nawaz, 2010).

There is no denial that in the contemporary eLearning environments, a teacher’s role for students has changed from providing well-cooked teacher’s knowledge for passive students to self-cooked inputs by the students themselves. For this purpose, the students have to be self-disciplined, self-motivated and at the most mature in the field of ICTs and their applications (Hvorecký et al., 2005). However, it is notable that like teachers, the learners’ preferences for their learning path depends on their personal characteristics of age, gender, perceptions about ICTs, and familiarity with the computer applications (Mehra & Mital, 2007; Nawaz & Kundi, 2010a).

VI. CONCLUSIONS Given the indispensability of computers in the

educational environments, there is no option with the teachers, students and administrators except finding some way out for their digital literacy. They all have to understand their changing roles and responsibilities and make efforts to get knowledge and skills for play them effectively. The research tells that eLearning users mostly acquire their knowledge of computers either formally or informally from friends and fellows. However, there is need for a structured formal eTraining of users that is based on a thorough analysis of the requirements for technology, institution, individual users and society at large.

The training contents and the process must be user-centric meaning that eTraining has to be designed in accordance with the teaching styles of teachers and learning styles of the students and administrators. This is possible if a comprehensive research project is first initiated to collect data about different aspects of eLearning environments and then designing the systems, the results can be promising. However, implementation of such an ideal system should not be the immediate rather long term objective. Attitude management takes sometime but if consistent efforts are not make for eTraining, most of the institutions continue using ICTs for low level applications.

References Références Referencias 1.

Aaron, M., Dicks, D., Ives, C. & Montgomery, B.

(2004). Planning for Integrating Teaching Technologies. Canadian Journal of Learning and Technology, 30(2), Spring. Retrieved May 14, 2007, from http://www.cjlt.ca/.

2.

Abrami, P. C., Bernard , R. M., Wade, A., Schmid, R. F., Borokhovski, E., Tamim, R., Surkes, M. A., Lowerison, G., Zhang, D., Nicolaidou, I., Newman, S., Wozney, l., and Peretiatkowicz, A. (2006). A Review of e-Learning in Canada: A Rough Sketch of the Evidence, Gaps and Promising Directions. Canadian Journal of Learning and Technology, 32(3), Fall/Autumn. Retrieved May 14, 2007, from http://www.cjlt.ca/.

3.

Afghan, N. (2000). Human resources development in Pakistan. Pakistan Economist. Retrieved April 14, 2007, from http://pakistaneconomist.com/issue2000/issue36 /i8e3.htm.

4.

Albion, P. R (1999). Self-Efficacy Beliefs as an Indicator of Teachers' Preparedness for Teaching with Technology. Retrieved May 10, 2007, from http://www.usq.edu.au/users/albion/papers /site99/1345.html.

5.

Aviram, R. & Tami, D. (2004). The impact of ICT on education: the three opposed paradigms, the lacking discourse. Retrieved July 14, 2007, from http://www.informatik.uni-bremen.de/~mueller/ kr-004/ressources/ict_impact.pdf.

6.

Barnes, K., Marateo, R. C., & Ferris, S. P. (2007). Teaching and Learning with the Net Generation. Innovate Journal of Online Education, 3(4). Retrieved April 10, 2007, from http://Innovateonline.info.

7.

Bataineh, R. F. & Abdel-Rahman, A. A. (2006).

Jordanian EFL students' perceptions of their computer literacy: An exploratory case study.



5

March

201

1

Early

View

http://www.cjlt.ca/�http://www.cjlt.ca/�http://pakistaneconomist.com/issue2000/issue36%20/i8e3.htm�http://pakistaneconomist.com/issue2000/issue36%20/i8e3.htm�http://www.usq.edu.au/users/albion/papers%20/site99/1345.html�http://www.usq.edu.au/users/albion/papers%20/site99/1345.html�http://www.informatik.uni-bremen.de/~mueller/%20kr-004/ressources/ict_impact.pdf�http://www.informatik.uni-bremen.de/~mueller/%20kr-004/ressources/ict_impact.pdf�http://www.informatik.uni-bremen.de/~mueller/%20kr-004/ressources/ict_impact.pdf�http://innovateonline.info/�


International Journal of Education and Development using ICT, 2(2). Retrieved April 10, 2007, from http://ijedict.dec.uwi.edu//viewarticle.php?id=169&layout=html].

8. Blázquez, F. E. & Díaz, L. A. (2006). A Training Proposal for e-Learning Teachers. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

9. Cawson, P. A. (2005). ICTs in Teacher Education: What do teachers need to learn? International Conference on Teacher Education, “Quality Teacher Education: The Challenges of the 21st Century” University of Cape Coast, Ghana, 17-20 August 2005. Retrieved April 19, 2007, from http://www.fiankoma.org/pdf/ICTs%20in%20Teacher%20Education.pdf.

10. Dalsgaard, C. (2006). Social software: E-learning beyond learning management systems. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

11. Davey, B. & Tatnall, A. (2007). The Lifelong Learning Iceberg of Information Systems Academics – A Study of On-Going Formal and Informal Learning by Academics. Journal of Information Technology Education, 6. Retrieved July 14, 2007, from http://jite.org/documents/Vol6/.

12. Dinevski, D. & Kokol, D. P. (2005). ICT and Lifelong Learning. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

13. Drinkwater, P. M., Adeline, C. M., French, S., Papamichail, K. N. & Rickards, T. (2004). Adopting a Web-Based Collaborative Tool to Support The Manchester Method Approach to Learning. Electronic Journal on e-Learning, 2(1), 61-68. Retrieved April 19, 2007 from http://www.ejel.org/volume-2/vol2-issue1/issue1-art23-drinkwater.pdf.

14. Ehlers, U. (2005). Quality in e-learning from a learner's perspective. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

15. Ekstrom, J. J., Gorka, S., Kamali, R., Lawson, E., Lunt, B., Miller, J. & Reichgelt, H. (2006). The Information Technology Model Curriculum. Journal of Information Technology Education, 5. Retrieved July 14, 2007, from http://jite.org/documents/Vol5/.

16. Ezer, J. (2006). India and the USA: A Comparison through the Lens of Model IT Curricula. Journal of Information Technology

Education, 5. Retrieved July 14, 2007, from http://jite.org/documents/Vol5/.

17. Ezziane, Z. (2007) Information Technology Literacy: Implications on Teaching and Learning. Journal of Educational Technology & Society, 10 (3), 175-191. Retrieved April 10, 2007, from http://www.ask4research.info/.

18. Garcia, P. & Qin, J. (2007). Identifying the Generation Gap in Higher Education: Where Do the Differences Really Lie? Innovate Journal of Online Education, 3(4), April/May. Retrieved April 10, 2007, from http://Innovateonline.info.

19. Gay, G., Mahon, S., Devonish, D., Alleyne, P. & Alleyne, P. G. (2006). Perceptions of information and communication technology among undergraduate management students in Barbados. International Journal of Education and Development using ICT, 2(4). Retrieved May 11, 2007, from http://ijedict.dec.uwi.edu//

20. Graff, M., Davies, J. & McNorton, M. (2001). Cognitive Style and Cross Cultural Differences in Internet Use and Computer Attitudes. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

21. Gray, D. E., Ryan, M. & Coulon, A. (2003). The Training of Teachers and Trainers: Innovative Practices, Skills and Competencies in the use of eLearning. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

22. Hameed, T. (2007). ICT as an enabler of socio-economic development. Retrieved June, 24 2007, from http://www.itu.int/osg/spu/digitalbridges/materials/hameed-paper.pdf.

23. Hvorecký, J., Manažmentu, V. S. & Cesta, P. (2005). Can E-learning break the Digital Divide? European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

24. Jewels, T. & Ford, M. (2006). The Development of a Taxonomy of Desired Personal Qualities for IT Project Team Members and Its Use in an Educational Setting. Journal of Information Technology Education, 5. Retrieved July 14, 2007, from http://jite.org/documents/Vol5/.

25. Johnson, D. W., Bartholomew, K. W. & Miller, D. (2006). Improving Computer Literacy of Business Management Majors: A Case Study. Journal of Information Technology Education, 5. Retrieved July 14, 2007, from http://jite.org/documents/Vol5/.

26. Juniu, S. (2005). Digital Democracy in Higher Education Bridging the Digital Divide. Innovate Journal of Online Education, 2(1),


6

March

201

1 Users Training: The Predictor of Successful eLearning in HEIs

Early

View

http://ijedict.dec.uwi.edu/viewarticle.php?id=169&layout=html�http://ijedict.dec.uwi.edu/viewarticle.php?id=169&layout=html�http://www.eurodl.org/�http://www.fiankoma.org/pdf/ICTs%20in%20Teacher%20Education.pdf�http://www.fiankoma.org/pdf/ICTs%20in%20Teacher%20Education.pdf�http://www.eurodl.org/�http://jite.org/documents/Vol6/�http://www.eurodl.org/�http://www.ejel.org/volume-2/vol2-issue1/issue1-art23-drinkwater.pdf�http://www.ejel.org/volume-2/vol2-issue1/issue1-art23-drinkwater.pdf�http://www.eurodl.org/�http://jite.org/documents/Vol5/�http://jite.org/documents/Vol5/�http://www.ask4research.info/�http://innovateonline.info/�http://ijedict.dec.uwi.edu/�http://www.eurodl.org/�http://www.eurodl.org/�http://www.itu.int/osg/spu/digitalbridges/materials/hameed-paper.pdf�http://www.itu.int/osg/spu/digitalbridges/materials/hameed-paper.pdf�http://www.eurodl.org/�http://jite.org/documents/Vol5/�http://jite.org/documents/Vol5/�


October/November. Retrieved April 10, 2007, from http://Innovateonline.info.

27. Kopyc, S. (2007). Enhancing Teaching with Technology: Are We There Yet? Innovate Journal of Online Education. December 2006/January 2007, 3(2). Retrieved April 10, 2007, from http://Innovateonline.info.

28. Kundi, GM & Nawaz, A. (2010). From objectivism to social constructivism: The impacts of information and communication technologies (ICTs) on higher education. Journal of Science and Technology Education Research. 1(2):30-36. Available online http://www.academicjournals.org/JSTER

29. LaCour, S. (2005). The future of integration, personalization, and ePortfolio technologies. Innovate Journal of Online Education, 1(4), April/May. Retrieved April 10, 2007, from http://Innovateonline.info.

30. Laffey, J. M. & Musser, D. (2006). Shadow netWorkspace: An open source intranet for learning communities. Canadian Journal of Learning and Technology, 32(1), Winter. Retrieved May 14, 2007, from http://www.cjlt.ca/.

31. Loing, B. (2005). ICT and Higher Education. General delegate of ICDE at UNESCO. 9th UNESCO/NGO Collective Consultation on Higher Education (6-8 April 2005). Retrieved June 24, 2007, from http://ong-comite-liaison.unesco.org/ongpho/acti/3/11/rendu/20/pdfen.pdf.

32. Luthans, F. (1995). Organizational behavior. 7th ed. McGraw-Hill.

33. Lynch, J., Sheard, J., Carbone, A. & Collins, F. (2005). Individual and Organizational Factors Influencing Academics’ Decisions to Pursue the Scholarship of Teaching ICT. Journal of Information Technology Education, 4. Retrieved July 14, 2007, from http://jite.org/documents/Vol4/.

34. Macleod, H. (2005). What role can educational multimedia play in narrowing the digital divide? International Journal of Education and Development using ICT, 1(4). Retrieved May 11, 2007, from http://ijedict.dec.uwi.edu//.

35. Manochehr, N. (2007). The Influence of Learning Styles on Learners in E-Learning Environments: An Empirical Study. Computers in Higher Education and Economics Review, 18. Retrieved April 10, 2007, from http://www.economicsnetwork.ac.uk/cheer.htm.

36. Marcella, R. & Knox, K. (2004). Systems for the management of information in a university context: An investigation of user need. Information Research, 9(2), Paper 172. Retrieved April 10, 2007, from http://InformationR.net/ir/9-2/paper172.html.

37. McPherson, M. & Nunes, M. B. (2004). The Role of Tutors as an Integral Part of Online Learning Support. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

38. Mehra, P. & Mital, M. (2007). Integrating technology into the teaching-learning transaction: Pedagogical and technological perceptions of management faculty. International Journal of Education and Development using ICT, 3(1). Retrieved October 11, 2007, from http://ijedict.dec.uwi.edu//.

39. Mujahid, Y. H. (2002). Digital opportunity initiative: for Pakistan. A paper evaluating Pakistani eReadiness Initiatives. National Post Graduate Institute of Telecommunications & Informatics, PTCL. IT & Telecom Division, Ministry of Science and Technology, Government of Pakistan. Retrieved May 11, 2007, from http://www.most.gov.pk/

40. Nawaz, A. & G M Kundi (2010) Demographic implications for the eLearning user perceptions in HEIs of NWFP, Pakistan. Electronic Journal of Information Systems for Developing Countries, 41(5), 1-17. (a)

41. Nawaz, A. & Kundi, GM. (2010) Predictor of e-learning development and use practices in higher education institutions (HEIs) of NWFP, Pakistan. Journal of Science and Technology Education Research. 1(3):44-54. Available online http://www.academicjournals.org/JSTER (b)

42. Nawaz, A. & Kundi, GM. (2010) Digital literacy: An analysis of the contemporary paradigms. Journal of Science and Technology Education Research. 1(2):19-29. Available online http://www.academicjournals.org/JSTER (c)

43. Nawaz, A., Kundi, G. M. & Shah, D. B. (2007). Metaphorical interpretations of information systems failure. Peshawar University Teachers’ Association Journal, 14, 15-26.

44. Oh, Eunjoo & French, Russell (2004). Pre-service Teachers’ Perceptions of an Introductory Instructional Technology Course. Electronic Journal for the Integration of Technology in Education, 3(1). Retrieved April 10, 2007, from http://ejite.isu.edu/Volume3No1/.

45. Oliver, R. (2002). The role of ICT in higher education for the 21st century: ICT as a change agent for education. Retrieved April 14, 2007 from http://elrond.scam.ecu.edu.au/oliver/2002/he21.pdf.

46. Qureshi, Q.A., Ahmad, S., Najibullah., Nawaz, A., & Shah, B. (2009) eLearning development in HEIs: Uncomfortable and comfortable zones for



7

March

201

1

Early

View

http://innovateonline.info/�http://innovateonline.info/�http://innovateonline.info/�http://www.cjlt.ca/�http://ong-comite-liaison.unesco.org/ongpho/acti/3/11/rendu/20/pdfen.pdf�http://ong-comite-liaison.unesco.org/ongpho/acti/3/11/rendu/20/pdfen.pdf�http://ong-comite-liaison.unesco.org/ongpho/acti/3/11/rendu/20/pdfen.pdf�http://jite.org/documents/Vol4/�http://ijedict.dec.uwi.edu/�http://www.economicsnetwork.ac.uk/cheer.htm�http://informationr.net/ir/9-2/paper172.html�http://www.eurodl.org/�http://ijedict.dec.uwi.edu/�http://www.most.gov.pk/�http://ejite.isu.edu/Volume3No1/�http://elrond.scam.ecu.edu.au/oliver/2002/he21.pdf�http://elrond.scam.ecu.edu.au/oliver/2002/he21.pdf�


developing countries. Gomal University Journal of Research, 25(2), 47-56. (GUJR)

47. Radosevich, D. & Kahn, P. (2006). Using Tablet Technology and Recording Software to Enhance Pedagogy. Innovate Journal of Online Education, 2(6). Aug/Sep. Retrieved April 10, 2007, from http://Innovateonline.info.

48. Reilly, C. (2005). Teaching by Example: A Case for Peer Workshops about Pedagogy and Technology. Innovate Journal of Online Education, 1(3), February/March. Retrieved April 10, 2007, from http://Innovateonline.info.

49. Sanyal, B. C. (2001). New functions of higher education and ICT to achieve education for all. Expert Roundtable on University and Technology-for-Literacy/Basic Education Partnership in Developing Countries. Paris, 10 to 12 Sep, 2001. International Institute for Educational Planning, UNESCO. Retrieved April 10, 2007, from http://www.literacyonline.org/products/ili/pdf/UTLPsanyal.pdf.

50. Sasseville, B. (2004). Integrating Information and Communication Technology in the Classroom: A Comparative Discourse Analysis. Canadian Journal of Learning and Technology, 30(2), Spring. Retrieved May 14, 2007, from http://www.cjlt.ca/.

51. Sirkemaa, S. (2001). Information technology in developing a meta-learning environment. European Journal of Open, Distance and E-Learning. Retrieved April 10, 2007, from http://www.eurodl.org/.

52. Stockley, D. (2004). Strategic Planning for Technological Innovation in Canadian Post Secondary Education. Canadian Journal of Learning and Technology, 30(2), Spring. Retrieved May 14, 2007, from http://www.cjlt.ca/.

53. Thurab-Nkhosi, D., Lee, M. & Giannini-Gachago, D. (2005). Preparing Academic Staff for e-Learning at the University of Botswana. Innovate Journal of Online Education, 2(1), Oct/Nov. Retrieved April 10, 2007, from http://Innovateonline.info.

54. Tinio, V. L. (2002). ICT in education. Presented by UNDP for the benefit of participants to the World Summit on the Information Society. UNDP’s regional project, the Asia-Pacific Development Information Program (APDIP), in association with the secretariat of the Association of Southeast Asian Nations (ASEAN). Retrieved July 14, 2007 from http://www.apdip.net/publications/iespprimers/eprimer-edu.pdf.

55. Higher education in a globalized Society. (2004). UNESCO Education Position Paper. Retrieved April 10, 2007, from

http://unesdoc.unesco.org/images/ 0013/001362/136247e.pdf.

56. Educational Decision Support Systems. (2006). UNESCO. Retrieved July 14, 2007, from http://portal.unesco.org/en/ev.php-URL_ID=5559&URL_DO=DO_TOPIC&URL_SECTION=201.html.

57. Policies and trends in higher education. (2001). The University of Queensland Australia (UQA). Retrieved April 10, 2007, from http://www.tedi.uq.edu.au/largeclasses/pdfs/LitReview_6_ policies&trend.pdf.

58. Valcke , M. (2004). ICT in higher education: An uncomfortable zone for institutes and their policies. In R. Atkinson, C. McBeath, D. Jonas-Dwyer & R. Phillips (Eds), Beyond the comfort zone: Proceedings of the 21st ASCILITE Conference (pp. 20-35). Perth, 5-8 December. Retrieved April 10, 2007, from http://www.ascilite.org.au/conferences/perth04/procs/valcke-keynote.html.

59. Vrana, I. (2007). Changes required by ICT era are painful sometimes. CAUSE98, an EDUCAUSE conference, online proceedings. Retrieved October 10, 2007, from http://www.educause.edu/copyright.html.

60. Wells, R. (2007). Challenges and opportunities in ICT educational development: A Ugandan case study. International Journal of Education and Development using ICT, 3(2). Retrieved July 21, 2007 from http://ijedict.dec.uwi.edu//.

61. Wikipedia (2009) http://www.Wikipedia.org/ accessed 10 February, 2009.

62. Willis, J. (2006). Creating a Working Model for Technology Integration Through a Lesson Planning WebQuest. Electronic Journal for the Integration of Technology in Education, 5, 25-33. Retrieved April 10, 2007, from http://ejite.isu.edu/Volume5No1/.

63. Young, L. D. (2003). Bridging Theory and Practice: Developing Guidelines to Facilitate the Design of Computer-based Learning Environments. Canadian Journal of Learning and Technology, 29(3), Fall/Autumn. Retrieved May 14, 2007, from http://www.cjlt.ca/.

64. Zhao, Y. & LeAnna Bryant, F. (2006). Can Teacher Technology Integration Training Alone Lead to High Levels of Technology Integration? A Qualitative Look at Teachers’ Technology Integration after State Mandated Technology Training. Electronic Journal for the Integration of Technology in Education, 5, 53-62. Retrieved April 10, 2007, from http://ejite.isu.edu/Volume5No1/.

Users Training: The Predictor of Successful eLearning in HEIs Global Journal of Computer Science and Technology Volume XI Issue IV Version I

8

March

201

1

Early

View

http://innovateonline.info/�http://innovateonline.info/�http://www.literacyonline.org/products/ili/pdf/UTLPsanyal.pdf�http://www.literacyonline.org/products/ili/pdf/UTLPsanyal.pdf�http://www.cjlt.ca/�http://www.eurodl.org/�http://www.cjlt.ca/�http://innovateonline.info/�http://www.apdip.net/publications/%20iespprimers/eprimer-edu.pdf�http://www.apdip.net/publications/%20iespprimers/eprimer-edu.pdf�http://unesdoc.unesco.org/images/%200013/001362/136247e.pdf�http://unesdoc.unesco.org/images/%200013/001362/136247e.pdf�http://portal.unesco.org/en/ev.php-URL_ID=5559&URL_DO%20=DO_TOPIC&URL_SECTION=201.html�http://portal.unesco.org/en/ev.php-URL_ID=5559&URL_DO%20=DO_TOPIC&URL_SECTION=201.html�http://portal.unesco.org/en/ev.php-URL_ID=5559&URL_DO%20=DO_TOPIC&URL_SECTION=201.html�http://www.tedi.uq.edu.au/largeclasses/pdfs/LitReview_6_%20policies&trend.pdf�http://www.tedi.uq.edu.au/largeclasses/pdfs/LitReview_6_%20policies&trend.pdf�http://www.ascilite.org.au/conferences/perth04/procs/valcke-keynote.html�http://www.ascilite.org.au/conferences/perth04/procs/valcke-keynote.html�http://www.educause.edu/copyright.html�http://ijedict.dec.uwi.edu/�http://www.wikipedia.org/�http://ejite.isu.edu/Volume5No1/�http://www.cjlt.ca/�http://ejite.isu.edu/Volume5No1/�


Performance Comparison of Radial Basis Function Networks and Probabilistic Neural Networks for Telugu Character Recognition

T. Sitamahalakshmi1, Dr.A.Vinay Babu2, M. Jagadeesh3, Dr.K.V.V.Chandra Mouli4

Abstract--The research on recognition of hand written scanned images of documents has witnessed several problems,

some of which include recognition of almost similar characters. Therefore it received attention from the fields of image processing and pattern recognition. The system of pattern recognition comprises a two step process. The first stage is the feature extraction and the second stage is the classification. In this paper, the authors propose two classification methods, both of which are based on artificial neural networks as a means to recognize hand written characters of Telugu, a language spoken by more than 100 million people of south India(Negi et al. ,2001). In this model, the authors used Radial Basis Function (RBF) networks and Probabilistic Neural Networks (PNN) for classification. These classifiers were further evaluated using performance metrics such as accuracy, sensitivity, specificity, Positive Predictive Value (PPV), Negative Predictive Value (NPV) and F measure. This paper is a comparison of results obtained with both the methods. The values of F measure are quite satisfactory and this is a good indication of the suitability of the methods for classification of characters. The values of F-Measure for both the methods approach the value of 1, which is a good indication and out of the two, RBF is a better method than PNN.

Keywords:

Classification, sensitivity, specification, F-measure, PPV, NPV.

I.

Introduction

haracter recognition is a form of pattern recognition (Khawaja etal.,2006). Any pattern recognition system consists of two major steps,

feature extraction and classification. The main focus in this paper is on classification. Classification is one of the important decision making factor for many real world problems. In this model authors used the classification techniques for identifying similar shaped Telugu characters.

About1- Department of CSE E-mail- [email protected] About2- Department of CSE, JNT University, Hyderabad, India. E-mail- [email protected] About3- Department of CSE E-mail- [email protected] About4- Department of IPE, GITAM University, Visakhapatnam, India E-mail- [email protected]

classifiers. RBF neural networks have fast training and learning rate because of their locally tuned neurons. They also exhibit a universal approximation property and good generalization ability. Probabilistic neural network integrates the characteristics of statistical pattern recognition and Back Propagation Neural Network (BPNN) and it has the ability to identity boundaries between the categories of patterns. In this research work the aforementioned two classifiers have been chosen for identification

of Telugu script and then compared their performance.

II.

Literature Review

(Nawaz et al., 2004) developed a system for

recognition of Arabic characters with RBF network and Hu invariant moments are used as predictor variables. (Ashok and Rajan, 2010) designed a system for writer identification with hand writing using Radial Basis

function. The efforts published by (Vijay and Ramakrishnan, 2004) described a system for the recognition of Kannada text where they used the wavelet features as attributes and RBFN as a classifier. (Birijesh ,2010) designed the system for the hand written Hindi characters and in this work the performance of Multi Layer Perceptron (MLP) and RBF networks were compared and it was shown that RBF is superior to MLP. (Kunte and Samuel, 2007) developed a neural network classifier with Hu invariant moments, Zernike moments as predictor variables and RBF network as a classifier. (Vatkin and Selinger,2001) used RBF neural network for the classification of hand written Arabic numerals using Legendre moments as predictor variables.(Romera et al.,1997) described an advanced system of classification using probabilistic neural networks and they used the classifier for optical Chinese character recognition. (Khatatneh et al., 2006) proposed a new technique in developing a recognition system for handling Arabic hand written characters with probalistic neural networks which yields a significant improvement. The work published by (Koche, 2010) compared the classification results of template matching, probabilistic

C Global Journal of Computer Science and Technology Volume XI Issue IV Version I

9

March

201

1

Early

View

In the present work, authors used radial basis function network and probabilistic neural network as

neural network, and feed forward back propagation neural network where the performance of PNN was

Performance Comparison of Radial Basis Function Networks and Probabilistic


superior. (Jeatrakul and Wong, 2009) compared the performance of classifiers developed using RBF and PNN and according to them the performance of RBF was found to be superior.

From the literature survey it has been observed

that the recognition systems were developed for Arabic and Kannada and Chinese script using RBF and probabilistic work. Not much work had been reported for Telugu script using RBF and PNN. This literature review reveals a dearth in information regarding recognition of Telugu hand-written characters. It inspired us to develop a classifier for Telugu script using RBF and PNN and compare the performance of the networks.

III.

Problem Statement

Application of neural networks for optical character recognition is the problem domain. The goal of this paper is to construct classifiers with radial basis function networks, probabilistic neural networks and to compare the performance.

IV.

Proposed System

The model proposed in this paper builds a pattern recognition system. Any pattern recognition comprises of

two steps, feature extraction and classification. As the main aim of the paper is for classification a brief review of feature extraction is given.

1)

Feature Extraction

As predictor variables used in the classification play a major role in increasing the accuracy of the classifier, the feature extraction is an important step. The system proposed by us is for the classification of Telugu hand written letters. The Telugu characters are neither available commercially nor available on the net. So the authors collected images from 60 people covering

different educational back grounds and different age groups. Sample set of characters collected from one person and the corresponding Telugu alphabet and the class label are shown in figure 1.

Figure1: Sample set of Characters

As handwriting varies from person to person and from time to time with the same person, the following preprocessing steps are required before extracting the features.

2)

Normalization

All the scanned images are brought to a common size by identifying the tight fit rectangular boundary around the image and they are scaled to 32x32 image.

3)

Binarization and Thinning

The aim of this process is to separate the character from the back ground in the grey image color to

black and white and then the image is thinned down

to skeleton of unitary thickness.

After preprocessing a set of 41 features are extracted from the skeletal images covering the local, global and statistical features. A brief description of the features is given in Table 1.

Table 1: Description of Features V1

The number of pixels in skeletal image that are in excited state

V2

The number of pixels in skeletal image that have one exited neighbor V3

The number of pixels in skeletal image that have two excited neighbors

V4

The number of pixels in the skeletal image that have three excited neighbors V5

The number of pixels in the skeletal image that have two exited neighbors which are 180 degrees apart

V6 ,V7 ,V8

The densities of pixels in the exited state when the image is divided into three regions horizontally V9,v10.v11

The densities of the pixels in the exited state when the image is divided into three regions vertically

V12

Total number of crossings i.e., changes from 1’s to 0’s and from 0’s to 1’s as the image scanned horizontally V13

Total change in the horizontal crossings

V14

Total number of crossings i.e., changes from 0’s to 1’s and from 1’s to 0’s as the image is scanned in the vertical direction V15

Total change in the vertical crossings

V16

The number of connected components in the image V17

Euler number the binary matrix i,e., the skeletal image

V18

entropy: is a statistical measure of the randomness that can be used to characterize the texture of the input image Entropy=-sum (p*log2

(p));

V19

Energy: is the sum of squared elements in the grey level co occurrence matrix. Energy=∑ p(I) 2

for all i and j

V20

Contrast: returns a measure of the intensity contrast between a pixel and its neighbor over the whole image Contrast = ∑ | i-j| 2

p(i,j) for all I and j

V21

Correlation: is measure of how correlated a pixel to its neighbor over the whole image.

Correlation=∑((i-µi)(j-µj)p(I,j))/σi

σj

Neural Networks for Telugu Character Recogn RecognitionGlobal Journal of Computer Science and Technology Volume XI Issue IV Version I

10

March

201

1

Early

View


V22 Cluster tendency: Measure of the grouping of the pixels that have similar gray level values. Cluster tendency=∑ ∑ (I+j -2µ)k p(I,j)

V23 Standard deviation of the binary matrix V24 Maximum value of the gray level co occurrence matrix V25,V26 Co ordinates of the centroid of the binary skeletal image V27 ,V28 Number of crossings at the centroid in horizontal and vertical directions V29 Eccentricity: scalar that specifies the eccentricity of an ellipse that has same second moments as the region of the image V30 Orientation: scalar (in degrees) between the x axis and the major axis of the ellipse ,that has the same second moments as the image V31 Scalar that specifies the number of pixels in the convex area of the image V32 Diameter: scalar that specifies the diameter of the circle as the region of the image V33 Solidity: scalar specifying the proportion of pixels that are in the region of the image. V34 Extent: scalar that specifies the ratio of pixels to the total in the bounding box V 35 to v41

Hu invariant moments: seven moment based features which are invariant to size and orientation of the character

As the data obtained for different features are with different scales, standardization of the data is required before proceeding with any classification task. The standardization is performed with

X

1

SX

-X=

X

Where X is the median and Sx is the standard

deviation.

To ensure accurate classification a large number of features are extracted in our models which are to be characteristic of each individual character. Different researchers used different number of variables to suit their purposes like (Huette et al., 1998) who used about 124 and (Patra et al., 2002) who used only 17 and the authors used 41 variables. As the number of features increases, the complexity of the pattern recognition system increases, so we reduced the dimensions by using factor analysis. Predictor variables are reduced to 18 variables from a total of 41 variables.

4)

Classification

Classification is a data mining technique used to

predict group membership for data instances. The objective of the data classification is to analyze the input data and to develop an accurate description or model for each class using the features present in the data. The model is used to predict the class label of unknown records and such modeling is referred as predictive modeling. The methodology used in the paper uses predictive modeling and developed using neural networks. As the goal of this work is to compare the performance of a classification model and is based on the counts of test samples correctly and incorrectly predicted by the model.

5)

Performance Metrics

Several criteria

may be used to evaluate the

performance of a classification algorithm in supervised learning. A confusion matrix is a useful tool for analyzing how well a classifier can identify test samples of different classes (Han and Kamber, 2009), which tabulates the records correctly and incorrectly predicted by the model. Each entry Cij in the confusion matrix denotes the number of records from class i predicted to be of class j.

Although confusion matrix provides the information needed to determine how well a classification model performs, summarizing this information with a single number would make it convenient to compare the performance of different models. This can be done by using the performance metrics such as sensitivity or recall, specificity, precision or positive predictive value, negative predictive value, F-measure and accuracy.

6) Sensitivity:

It measures the actual members of the class which are correctly identified as such. It is also referred as True Positive Rate (TPR) or recall. It is defined as the fraction of positive examples predicted correctly by the classification model

Sensitivity (recall) = )( FNTP

TP+

7) Specificity

It is also referred to as true negative rate .It is defined as the fraction of negative examples which are predicted correctly by the model

Specificity = )( FPTN

TN+

8) Precision (Positive predictive value)

It is also called as positive predictive value and determines the fraction of records that actually turns out to be positive in the group which has been declared as positive class by the classifier

Precision = )( FPTP

TP+

9) Negative predictive value (NPV)

It is proportion of the samples which do not belong to the class under consideration and which are correctly identified as non members of the class

NPV= )( FNTN

TN+

10) F-measure

It can be used as a single measure of performance of the test. The F measure is the harmonic mean of precision and recall

Performance Comparison of Radial Basis Function Networks and Probabilistic Neural Networks for Telugu Character Recogn Recognition


11

March

201

1

Early

View


F Measure = )(

**2recallprecision

recallprecision+

11) Accuracy

Accuracy is used as a statistical measure of how well a binary classification test identifies or excludes a condition. It is a measure of proportion of true results

Accuracy = )(

)(FNTNFPTP

TNTP+++

+

Where TP=true positives, TN =True negatives, FP=false positives, FN=false negatives In our model we used the following two networks for classification

1. Radial Basis Function Networks 2. Probabilistic Neural Networks

12) Radial Basis Function Approach

Radial basis function network which is a feed forward network consists of three layers input layer, hidden layer and the output layer. The architecture of RBF is shown in Figure 2. The RBF is different from the ordinary feed forward networks in calculating the activations of hidden neurons. The activations at the hidden neurons are computed by using the exponential of distance measures.

Each node in the input layer corresponds to a component of the input vector x. The second layer, the only hidden layer in the neural network applies non linear transformation from input space into hidden space by employing non-linear activation function such as Gaussian kernel. A linear node at the output layer corresponds to the classes of the problem. A simple way to choose the number of radial basis functions is to create a hidden neuron centered on each training pattern. However, this method is computationally very costly and takes up huge amount of memory. In our model, the training patterns are clustered into a reasonable number of groups using K-means clustering algorithm.

INPUT LAYER HIDDEN LAYER OUTPUT LAYER

Figure 2: Radial Basis Function Network

Then a neuron is assigned to each cluster centre. The output of each hidden neuron is calculated by using the Gaussian radial basis function

= 22

2||iµ - x||-exp||)iµ - x ||( σG

Where, x is the training sample, µi is the centre

of the hidden ith neuron and σ is the width of the neuron. The width of the basic functions are set to a value which is a multiple of the average distance between the centers. This value governs the amount of smoothing.

The activation at the output neurons is defined by the summation

( ) ∑ +=i

GwxY b ||)µ - x ||(* i

TT 1−

In our model,

we fixed the number of centers as

100 and width as 2.4 which is a multiple of the average width 0.6 of the hidden neuron. The percentages of characters correctly classified for different number of centers and for different widths (σ

values) are

shown in

Table 2 and table 3 respectively.

Table 2: Percentage of Characters Correctly Classified for Different Numbers of Centers

Table 3: Percentage of Characters Correctly Classified

for Different Values of σ σ % Characters Correctly Classified .6 72.5

1.2 78.2 1.8 77.8 2.4 78.8 3.0 78.2

With the above results, the authors fixed the

parameters, the number of hidden neurons as 100 and width of the basis function as 2.4. With these parameters the confusion matrix obtained as shown in Figure 3.

Number of Centers

% Characters Correctly Identified

90 75.8 100 77.7 110 75.8



12

March

201

1

Where, w is the weight vector. The weights arecomputed by W = (G G) G d Where d is the target class matrix.

Early

View


Figure 3: Confusion Matrix

13) Probabilistic Neural Network Approach

The PNN as described in Figure 4 consists of input layer, two hidden layers, (one of the example/pattern and class/summation) and an output layer. The process based classification that differentiates PNN and RBF is that PNN works on the estimation of probability density function

Figure 4: Probabilistic Neural Network

The input layer does not perform any computation and simply distribute the input to the neurons in the pattern layer which has one node for each training example. On receiving the pattern x from the input layer, the neuron xij of the pattern layer computes its output as

( )( )

( ) ( )

−−−= 22/ 2

exp2

1σσπ

φ ijT

ijddij

xxxxx

Where, d denotes the dimension of the pattern vector x, σ is the smoothening parameter and xij is the neuron vector. The summation layer neurons compute the maximum likelihood of pattern x being classified into Ci by summarizing and averaging the output of all the neurons that belong to the same class,

( )( )

−

−

−∑=

= 22exp1

2/2

1 Ni

1 σσπ

ijxxT

ijxx

iNddxiP j

Where, Ni denotes the total number of samples in a class Ci. If the apriori probabilities for each class are the same, the decision layer classifies pattern x in accordance with the ayes decision rule based on the output of all summation layer neurons.

mixPxC i ,.....2,1 )}(max{ arg)( ==Λ

Where )(xCΛ denotes the estimated class of pattern x and m is the total number of classes in the training samples. In our model we fixed the value of σ as 1.4 and the values of σ and percentage of characters classified for each σ are shown in Table 4 and the best value was found to be at 1.4.

Table 4: Percentage of Characters Correctly Classified for Different Values of σ

σ % Characters Correctly Classified .9 70.7 1 71.3

1.1 71.7 1.2 72.0 1.3 72.3 1.4 72.5 1.5 72.2 1.6 71.7

With σ=1.4 the confusion matrix is shown in Figure 4.

Figure 4: Confusion Matrix

XI. Results and Discussions

In this paper the authors compared the classification models developed using radial basis function network and probabilistic neural network. The summary of the confusion matrix for both the methods is shown in table 5 and table 6 respectively.

Table 5: Summary of Performance Metrics for RBF Network



13

March

201

1

Early

View


85

90

95

100

1 2 3 4 5 6 7 8 9 10

CLASS

ACCURACY MEASURE

ACCURACY

FOR RBF

ACCURACY

FOR PNN

00.20.40.60.8

1

1 2 3 4 5 6 7 8 9 10

CLASS

F MEASURE

FOR RBF

Table 6: Summary of Performance Metrics for PNN Network.

1.

The Performance metric accuracy which is a function of specificity and sensitivity is a measure for comparing two classifiers. The accuracy of RBF network for all the classes except classes with labels 8 and 10 is above 95% where as with PNN the accuracy for four classes with labels 1, 3, 4, 5

are above 95% and for the remaining is less than 95%. The comparison of accuracy measure is shown in figure 5.

2.

Building a model that maximizes both precision and recall is a key challenge in classification algorithm (Tan et al., 2007). Precision and recall can be summarized into another metric known as F measure as explained in performance metrics. The F measure for both the classes is shown in the form of a graph in figure 6. With the first method the value of F is less than 0.7 for classes with the labels 8, 10 and with PNN the value is less than 0.7 for classes with labels 2, 6, 8 and 10.

Figure 5: Accuracy Measure

V. Conclusion In this paper, the authors presented two

classification models, one is radial basis function networks and the other is probabilistic neural networks and both being implemented using MATLAB. The work was carried out with 600 images collected from 60 people and the result is tested with 10 fold cross validation. With RBF network 474 characters are classified correctly while with PNN 435 characters are classified correctly. The following observations are made from the results.

1. Only for class with label 3 the values of accuracy and F measure are found to be good with PNN and for all the remaining classes RBF is showing good results.

2. Percentage of characters classified correctly with RBF network is 78.8% and with PNN the percentage of characters classified correctly is 72.5.

3. Except for class with label 10 the value of F measure is nearer to one, the reason being the character considered for class with label 10 has similar structure with classes with labels 2, 6 and 7. The accuracy of all the classes is above 90%

with both the methods. But the overall accuracy of the RBF network is found to be better from the results. In future we are planning to extend the work to other characters of Telugu script and to use the techniques which are less dependent on the sample size considered for training of the classifier which may reduce the time required for classification

References Références Referencias

1.

Negi.A, Krishna.B, Bhagavati.C(2001), “An OCR system for Telugu” In Proceedings of Int. conf. on Document Analysis and Research ,1110-1114.

2.

Khawaja . A,Tinghi .S, Menon .N.M, Rajpur,(2006) .A “ Recognition of printed



14

March

201

1

Early

View


Chinese characters by using neural network “ , Proceedings of the Int. Multitopic conf. INMIC 2006,169-172.

3. Nawaz .S.N, Sarfraz.M ,Zidouri.A and G.Al-Khatib,(2004)” An Approach to offline Arabic character recognition using neural networks” Proceedings of the 11th IEEE Int. Conf. on Electronics, Circuits and Systems.

4. Ashok.J, Rajan.E.G, (2010)”Writer Identification and Recognition Using Radial Basis Function”, Int. Jour. of Computer Science and Information Technologies, 1(2), 51-57.

5. Vijay.K.B, Ramakrishnan A.G,(2004) “Radial Basis Function and Subspace Approach For Printed Kannada Text Recognition” Proceedings of the IEEE Int. Conf. on Acoustics, Speech and Signal Processing.

6. Birijesh.K.V. (2010)“Handwritten Hindi Character Recognition Using Multilayer Perceptron and: Radial Basis Function Neural Networks”, Int. Jour. of Computer Science & Communication,1(2),141-144.

7. Kunte.R.S and Samuel.R.D.S,(2007) “A simple and efficient optical character recognition system for basic symbols in printed Kannada text”, SADHANA ,32(5),521–533.

8. Vatkin.M, Selinger.M(2001) ”The system of Handwritten Characters Recognition on the Basis of Legendre Moments and Neural Network” ,The Intr. Wor. on Discrete- Event System Design, DESDes’01, June 27-29.

9. Romero.R, Touretzky.D, and Thibadeau.R,(1997) “Optical Chinese Character Recognition Using Probabilistic Neural Networks,” Pattern Recognition, 30,1279-1292.

10. Khatatneh.K, Ibrahiem M.M El Emary and Basem Al- Rifai,(2006)” Probabilistic Artificial Neural Network For Recognizing the Arabic Hand Written Characters”, Jour. of Computer Science 2 (12): 879-884.

11. Koche.K,(2010)” Comparison of Neural Network and Template Matching Technique for Identification of Characters in License Plate”, Proceedings of the Int. Conf. on Information Science and Applications ICISA 2010,6 February 2010, Chennai, India, 2010.

12. Jeatrakul.P and Wong.K.W,(2009)” Comparing the Performance of Different Neural Networks for Binary Classification Problems”, Proceedings of the Eighth Int. Sym. on Natural Language Processing.

13. Heutte.l,Paquet .T,Moreau .J.V,Lecourtier.Y and Oliver.C ,(1998)”A structural /stastical feature based vector for handwritten character recognition”, Pattern Recognition Letters ,629-641.

14. Patra.P.K, Nayak.M, Nayak.S.K and Gobbak.N.K ,(2002) ”Probabilistic Neural Network for Pattern Classification”, Proceedings of Int. Joint Conf. on Neural Networks, 1200-1205.

15. Han. J and Kamber. M, Data Mining concepts and Techniques, Elsevier publishers,(2nd ed.) 2009.

16. Tan .P.N, Steinback .M and Kumar .V, Introduction to Data Mining, Pearson Education, 2007.



15

March

201

1

Early

View




16

March

201

1

This page is intentionally left blank

Early

View


Adaptive Routing Based on Delay Trusted Routing in Adhoc Network

T.K.Shaik Shavali1, Dr T. Bhaskara Reddy2

Abstract-Existing network hardware is constantly being improved and new communication technology continues to be developed. Together with the trend that computing hardware becomes smaller and portable, this network technology progress has led to dynamic networks. Next generation wireless networks are characterized as heterogeneous networks, particularly in terms of its underlying technology. One of the challenges of these heterogeneous networks is to manage handoff. Mobile IP is chosen for managing the handoff to accommodate the all-IP vision of the future interconnected networks. However, the handoff management of the mobile IP is mainly for data services where delay is not of a major concern. Therefore, it would be considerable challenge to achieve low latency handoff for real-time services. In this paper, we propose a multicasting scheme for delay-sensitive applications. Keywords- MobileIP,Handoff, Heterogeneous Network, Mobile stations, Quality of Service.

I. INTRODUCTION n the near future, a large number of Mobile Stations (MSs) will be equipped with multiple radio interfaces for wireless access to the Internet. A multi-mode MS

with multiple air interfaces (cellular interface, Bluetooth, IEEE 802.11 and IEEE 802.16 etc) and different data rates will be able to access cellular Base Stations (BSs), WLAN or WMAN Access Points (APs). In this scenario, the integration of multi-hop ad hoc communications with infrastructure based (or single-hop) wireless networks, such as wireless WANs (e.g., 2.5G, 3G, and 4G), wireless LAN (e.g., IEEE 802.11 a/b/e/g and HiperLAn/2) and wireless MANs (e.g., IEEE 802.16), is fundamental to improving the coverage and performance of the integrated network [3]. In addition, multi-hop communications can be used to increase the utilization and capacity of a BS by decreasing the co-channel interference via lowering the transmission power either of the BS or of the MSs [5] [9].Also, the integration can be useful in achieving load-balancing by forwarding part of the traffic from an overloaded cell to a free neighboring cell [6] [7]. From the protocol stack perspective, the network layer is the lowest possible layer where the convergence of heterogeneous wireless systems can be About1- Professor, Department of Computer Science, Lords institute of Engineering & Tech, Hyderabad-08, A.P., INDIA E-mail:- [email protected] About2- Department of Computer Science & Technology, S.K. University, Anantapur-03, A.P.,INDIA E-mail:[email protected]

developed. Furthermore, the desire to extend the great success of the Internet Protocol (IP) from the wired world to wireless leads to an all-IP vision [3]. So far, the IP is the best integration technology for heterogeneous networks and there is currently no foreseeable alternative to the IP [4]. To allow for seamless handoff to take place in IP-based heterogeneous networks, the IP must support users' mobility. In an effort to do that, the Internet Engineering Task Force (IETF) has developed the mobile IP standard to support mobility in IP-based networks [5]. In recent years, there has been a considerable amount of works that address the mobile IP-based handoff problem in heterogeneous networks [2],[3], [6]-[9]. Since data packets could be lost during the latency period, mobile IP-based handoff may not meet the quality-of-service (QoS) requirements for real time voice applications. Even though, mobile IP describes a scheme to recover the lost packets from the old foreign agent to the new one, this process takes some time as the signal experiences a random delay when it travels through the network. This makes the latency even longer. For non-real time services, this additional delay will not create a major problem. However, for real time services, this will dramatically degrade the QoS requirements. This problem can be solved if multicasting is employed. In this case, data packets are sent to the neighboring foreign agents as soon as the Received Signal Strength (RSS) of the mobile host goes below a certain threshold level. When this occurs, the data packets are stored in the buffer at the new foreign agent, and in the process, the latency can be reduced.

In this paper, we consider a multicasting scheme to solve the handoff latency problem in heterogeneous networks. The proposed handoff technique offers two main advantages:

a. It reduces the handoff latency in hybrid networks,

b. Recovers lost packets during the handoff process, which increases the system throughput.

II. MOBILE IP AND HANDOFFS First, second- and third-generation mobile

systems depended on the employment of the radio spectrum that was either unlicensed (available for public use) or licensed for use by a very small number of service providers and network operators in each region.

I


17

March

201

1

Early

View

Adaptive Routing Based on Delay Trusted Routing in Adhoc Network


Differences in bandwidth and coverage areas have led to the necessity of developing multi-network interface devices (terminals) that are capable of using the variety of different network services provided.

1) Mobile IP

Mobile IP is an Internet protocol, defined by the Internet Engineering Task Force (IETF) that allows users keep the same IP address, and stay connected to the Internet while roaming between networks. The key feature of Mobile IP design is that all required functionalities for processing and managing mobility information are embedded in well-defined entities, the Home Agent (HA), Foreign Agent (FA), and Mobile Nodes (MNs) [1, 2]. When a MN moves from its Home Network (HN) to a Foreign Network (FN), the correct delivery of packets to its current point of attachment depends on the MN's IP address, which changes at every new point of attachment. Therefore, in order to guarantee packets delivery to the MN, Mobile IP allows the MN to use two IP addresses: The Home address, which is static and assigned to the MN at the home network; and the Care-of-Address (CoA), which represents the current location of the MN [2]. One of the main problems that face the implementation of the original Mobile IP is the Triangle Routing Problem. When a CN sends traffic to the MN, the traffic gets first to the HA, which encapsulates this traffic and tunnels it to the FA. The FA de-tunnels the traffic and delivers it to the MN. The route taken by this traffic is triangular in nature, and the most extreme case of routing can be observed when the CN and the MN are in the same subnet [4, 5].

In mobile IP, two network entities are defined to support users mobility namely; the home agent and the foreign agent. These two agents periodically send advertisement messages to their corresponding networks (i.e., home and foreign networks) to acknowledge the mobile of its present location. Based on these advertisement messages, and the present location of the mobile host, the mobile host decides whether it belongs to its home network or to a new foreign network. If the mobile host discovers that it has migrated to a new foreign network, it sends a registration request to the corresponding new foreign agent to obtain a care-of-address. Also the foreign agent registers the new address (i.e., new location) with the mobile host home agent. After this process, any data packets that are received at the mobile's home network will be encapsulated with a new IP address and tunneled to the new foreign agent to which the mobile host resides. The foreign agent (at the other end of the tunnel) takes care of the de-encapsulation of the arriving data packets, and then forwards them to the mobile host using the new IP address. In the same way, if the mobile host transmits data packets to its correspondent host, it uses the foreign agent for the tunneling process

to forward these data packets to the home agent for subsequent transmission to the correspondent host.

2) Classification of Handoffs

In principle, each mobile terminal (node) is, at all times, within range of at least one network access point, also known as a base station. The area serviced by each base station is identified as its cell. The dimensions and profile of every cell depend on the network type, size of the base stations, and transmission and reception power of each base station. Usually, cells of the same network type are adjacent to each other and overlap in such a way that, for the majority of time, any mobile device is within the coverage area of more than one base station. Cells of heterogeneous networks, on the other hand, are overlaid within each other. Therefore, the key issue for a mobile host is to reach a decision from time to time as to which base station of which network will handle the signal transmissions to and from a specific host and handoff the signal transmission if necessary. We classify handoffs based on several factors as shown in Fig. 1. No longer is the network type the only handoff classification factor. Many more factors constitute categorization of handoffs including the administrative domains involved, number of connections and frequencies engaged. The following are categorization factors along with the handoff classifications that are based on them.

Figure1- Hierarichal Classification of Handoff

Handoffs can be classified as either horizontal

or vertical. This depends on whether a handoff takes place between a single type of network interface or a variety of different network interfaces. Horizontal Handoff: The handoff process of a mobile terminal between access points supporting the same network technology. For example, the changeover of signal transmission (as the mobile terminal moves around) from an IEEE 802.11b base station to a geographically neighboring IEEE 802.11b base station is considered as a horizontal handoff process. Vertical Handoff: The handoff process of a mobile terminal among access points supporting different network technologies. For example, the changeover of


18

March

201

1

Early

View


signal transmission from an IEEE 802.11b base station to an overlaid cellular network is considered a vertical handoff process.

III. SYSTEM ARCHITECTURE The proposed interconnection architecture

using mobile IP is shown in Fig. 2. The following are the network parameters and assumptions used in our handoff technique:

1. The home agent (HA), the foreign agents (FAs) and the correspondent host (CH) are interconnected through Internet

2. FAs are connected to the Internet through a wireless or a wired medium with large bandwidth.

3. The CH can be a fixed or mobile host. The time taken to switch from the home agent of

the mobile user to the new foreign agent is known as the mobile IP handoff latency. In addition to this handoff latency if the mobile host enters into a new foreign agent (from another foreign agent) during the tunneling process between the home agent and the old foreign agent, and before registering with the new foreign agent, data packets destined to the mobile host will be lost. These packets will then be retransmitted leading to an increase in the overall system delay.

Figure2- Proposed IP-based handoff architecture.

In delay-sensitive applications, handoff latency

can cause serious degradation in the quality of the underlying application. As a result of the frequent handoffs, this handoff latency becomes a major problem if the coverage area of the sub-networks gets smaller. Recent works on the existing problems of the handoff latency of mobile IP based networks and possible solutions can be found in [12]-[14].

IV. PROPOSED IMPROVEMENT IN LATENCY

1) Improvement in Registration Time

The improvement in Registration Time is achieved by starting to forward data packets after a small fixed delay (termed as the 'Fixed Registration

Delay') following the Registration

early - global journals · 2012. 9. 11. · iii. users of e-learning all users of ict-based tools...

Documents