report on song recommender

27
Raga (Music Recommender using Data Mining Analysis) 1

Upload: kruthika-vishwanath

Post on 09-Aug-2015

27 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Report on Song Recommender

Raga

(Music Recommender using Data Mining Analysis)

1

Page 2: Report on Song Recommender

DEPARTMENT OF INFORMATION SCIENCE AND ENGINEERING M.S.RAMAIAH INSTITUTE OF TECHNOLOGY

(Autonomous Institute Affiliated to VTU)

VIDYA SOUDHA, M. S. R Nagar, M. S. R. I. T. Post, Bangalore – 560054

CERTIFICATE

This is to certify that the following students, who were working under my guidance,

have completed their work as per my satisfaction with the topic Raga­Music

Recommender using Data Mining Analysis.The project has been approved as it

satisfies the academic requirements in respect of project work described for the

course in INFORMATION STORAGE AND MANAGEMENT(IS6E22)

1MS11IS044 Kruthika Vishwanath

1MS11IS063 Navya H C

1MS11IS075 Pooja J

1MS11IS081 Priyanka U Pandit T Signature of the Guide Mr. George Philip. C

Dept. of ISE, MSRIT, Bangalore­54

2

Page 3: Report on Song Recommender

Contents

Introduction…………………………………………………………………………...04

Literature survey……………………………………………………………………..05

Naive­bayes Algorithm……………………………………………………………..05

Apriori Algorithm…………………………………………………………………..06

Codes generated for implementing Recommendation System……………...07

File No­1……………………………………………………………………………….07

File No­2……………………………………………………………………………….10

File No­3……………………………………………………………………………….12

File No­4……………………………………………………………………………….13

File No­5……………………………………………………………………………….15

File No­6……………………………………………………………………………….23

Importance……………………………………………………………………………..25

Bibliography…………………………………………………………………………...26

Acknowledgment……………………………………………………………………...27

3

Page 4: Report on Song Recommender

Introduction

Recommender systems or recommendation systems are a subclass of information filtering system that seek to predict the 'rating' or 'preference' or ‘liking’ that user would give to an item.Basically what we are trying to do is embracing elephant in a room i.e filtering big data and obtaining useful stuffs.

Recommender systems have become extremely common in recent years, and are applied in a variety of applications. The most popular ones are probably movies, music, news, books, research articles, search queries, social tags, and products in general. However, there are also recommender systems for experts, jokes, restaurants, financial services, life insurance, online dating and twitter followers.

Recommender systems typically produce a list of recommendations in one of two ways ­ through collaborative or content­based filtering. Collaborative filtering approaches build a model from a user's past behavior as well as similar decisions made by other users; then use that model to predict items that the user may have an interest in.Content­based filtering approaches utilize a series of discrete characteristics of an item in order to recommend additional items with similar properties.These approaches are often combined. Our Raga recommendation system combines both the approach(Hybrid approach), i.e we use person’s liking towards songs and build a model using collaborative filtering and then implement content based filtering by using songs characteristics like artist,genre and eras. The data obtained from the hybrid approach is very huge and that is what is called big data. This big data can’t be processed using on­hand database management tools or traditional data processing application. There are separate algorithms used to scan them and extract useful business related stuffs(data mining) from big data. We have used two most popular algorithms, Naive­bayes Classfiers algorithm and Apriori algorithm. Former used for content­based filtering and latter used for setting a threshold for deciding which of the results to be included and which to be excluded. We give the user a set of songs chosen randomly from database and then he/she chooses songs they like and this is given as input to Naive­bayes Classifier algorithm and then the output obtained i.e person’s nature is analyzed, which artist does he/she like, which genre is he/she more inclined to or which era song does he/she like more is determined. This output is given as input to Apriori algorithm which filters all those data results below the threshold and provides user with songs which he/she may like.

4

Page 5: Report on Song Recommender

Literature survey We have gone through some articles concerning recommendation system and big data. This is compared to embracing an elephant in a room i.e we filter big data and provide useful business related things and also build a recommendation model.We can use loads of algorithms which analyses big data.Factors like Diversity,Recommender Persistence,User Demographics, Robustness,Serendipity ,Trust , Privacy Labelling are considered to choose algorithms. The popular algorithm used in recommendation system are Naive­bayes and Apriori. Lets discuss both one by one. Naive­bayes Algorithm A naive Bayes classifier is a simple probabilistic classifier based on applying Bayes' theorem with strong (naive) independence assumptions. A more descriptive term for the underlying probability model would be "independent feature mode.In simple terms, a naive Bayes classifier assumes that the value of a particular feature is unrelated to the presence or absence of any other feature, given the class variable. For example, a fruit may be considered to be an apple if it is red, round, and about 3" in diameter. A naive Bayes classifier considers each of these features to contribute independently to the probability that this fruit is an apple, regardless of the presence or absence of the other features. The Bayes Naive classifier selects the most likely classification P(ai|vj) given the attribute values a1, a2, . . . an. P(ai|vj) = (nc + mp)/(n + m) where, n=Number of cases where v=vj. nc=Number of cases where v=vj and a=ai. m=Equivalent sample size. p= A priori estimate for P(ai|vj). Lets consider a case where in we have provided user with 5 Shreya Goshal songs and he/she has selected only 2, based on their liking. Then, n = 5 (Number of cases where v=Shreya Goshal) nc = 2 (Number of cases where v=Shreya Goshal and a=checked) m = 3 (Sample size has three parameters ­ artist,genre and eras) p = 0.5 (Probability of choosing a particular singer or not)

5

Page 6: Report on Song Recommender

Apriori Algorithm Aprioris an algorithm for frequent itemset mining and association rule learning over transactional databases. It proceeds by identifying the frequent individual items in the database and extending them to larger and larger item sets as long as those itemsets appear sufficiently often in the database. The frequent item sets determined by Apriori can be used to determine association rules which highlight general trends in the database: this has applications in domains such as market basket analysis.Find frequent itemsets using an iterative level­wise approach based on candidate generation.Apriori uses a "bottom up" approach, where frequent subsets are extended one item at a time , and groups of candidates are tested against the data. The algorithm terminates when no further successful extensions are found Input: D, a database of transactions. min_sup, the minimum support count threshold. Output: L, frequent itemsets in D. This algorithm has two steps, 1. Find all the data result sets which has threshold support. 2. Use the frequent itemset to generate rules(If needed) Lets consider a case where in the output of naive­Bayes classifier algorithm are, P(Shreya Goshal) = 0.4 P(Sonu Nigam) = 0.3 P(Sunidhi) = 0.2 P(KK) = 0.0 P(MJ) = 0.1 If we chose min_sup as 0.2, then we can filter out KK and MJ. Lets consider a case where in there is same probability for many singers, then this also can be soughted using this algorithm. In short it helps in handling data result which lead to clashing assumptions.

6

Page 7: Report on Song Recommender

Codes generated for implementing Recommendation System To implement this music recommendation system,we have not used any tool , Instead have written own codes based on two algorithms as explained above using written in HTML, PHP,JAVASCRIPT. The software tool we have used is Phpmyadmin which uses PHP, intended to handle the administration of MySQL over the Web. We are running our code on localhost via port 80 and 3306. Xampp is used, which is an Apache distribution containing both Mysql and PHP. Apache is run via port 80 and Mysql via port 3306. We have the entire code distributed in several files which are dependent on each other. Given below are the codes, File No­1 <html> <head> <style type="text/css"> .slideThree width: 80px; height: 26px; background: #333; margin: 20px auto; ­webkit­border­radius: 50px; ­moz­border­radius: 50px; border­radius: 50px; position: relative; ­webkit­box­shadow: inset 0px 1px 1px rgba(0,0,0,0.5), 0px 1px 0px rgba(255,255,255,0.2); ­moz­box­shadow: inset 0px 1px 1px rgba(0,0,0,0.5), 0px 1px 0px rgba(255,255,255,0.2); box­shadow: inset 0px 1px 1px rgba(0,0,0,0.5), 0px 1px 0px rgba(255,255,255,0.2); .slideThree:after content: 'DISLIKE'; font: 12px/26px Arial, sans­serif; color: #000; position: absolute; right: 10px; z­index: 0; font­weight: bold;

7

Page 8: Report on Song Recommender

text­shadow: 1px 1px 0px rgba(255,255,255,.15); .slideThree:before content: 'LIKE'; font: 12px/26px Arial, sans­serif; color: #00bf00; position: absolute; left: 10px; z­index: 0; font­weight: bold; .slideThree label display: block; width: 34px; height: 20px; ­webkit­border­radius: 50px; ­moz­border­radius: 50px; border­radius: 50px; ­webkit­transition: all .4s ease; ­moz­transition: all .4s ease; ­o­transition: all .4s ease; ­ms­transition: all .4s ease; transition: all .4s ease; cursor: pointer; position: absolute; top: 3px; left: 3px; z­index: 1; ­webkit­box­shadow: 0px 2px 5px 0px rgba(0,0,0,0.3); ­moz­box­shadow: 0px 2px 5px 0px rgba(0,0,0,0.3); box­shadow: 0px 2px 5px 0px rgba(0,0,0,0.3); background: #fcfff4; background: ­webkit­linear­gradient(top, #fcfff4 0%, #dfe5d7 40%, #b3bead 100%); background: ­moz­linear­gradient(top, #fcfff4 0%, #dfe5d7 40%, #b3bead 100%); background: ­o­linear­gradient(top, #fcfff4 0%, #dfe5d7 40%, #b3bead 100%); background: ­ms­linear­gradient(top, #fcfff4 0%, #dfe5d7 40%, #b3bead 100%);background: linear­gradient(top, #fcfff4 0%, #dfe5d7 40%, #b3bead 100%); filter: progid:DXImageTransform.Microsoft.gradient( startColorstr='#fcfff4', endColorstr='#b3bead',GradientType=0 );

8

Page 9: Report on Song Recommender

.slideThree input[type=checkbox]:checked + label left: 43px; </style> </head> <body> <form name="songs" action="ism.php" method="POST" > <p><input type="checkbox" name="chckbox[]" value="tum hi ho" id="slideThree" />Tum hi ho</input></p> <p><input type="checkbox" name="chckbox[]" value="balma" id="slideThree"/>balma</input></p> <p><input type="checkbox" name="chckbox[]" value="bahara" id="slideThree"/>Bahara re</input></p> <p><input type="checkbox" name="chckbox[]" value="soniyo" id="slideThree"/>Soniyo</input></p> <p><input type="checkbox" name="chckbox[]" value="main agar kahoon" id="slideThree"/>Main agar kahoon</input></p> <p><input type="checkbox" name="chckbox[]" value="abhi mujme kahin" id="slideThree"/>Abhi mujme kahin</input></p> <p><input type="checkbox" name="chckbox[]" value="kamali" id="slideThree"/>Kamali</input></p> <p><input type="checkbox" name="chckbox[]" value="beintehan" id="slideThree"/>Aaja nachale</input></p> <p><input type="checkbox" name="chckbox[]" value="beedi" id="slideThree"/>Beedi</input></p> <p><input type="checkbox" name="chckbox[]" value="indino" id="slideThree"/>In dino</input></p> <p><input type="checkbox" name="chckbox[]" value="jaaneman" id="slideThree"/>Jaaneman</input></p> <p><input type="checkbox" name="chckbox[]" value="dilnashin" id="slideThree"/>Dil nashi</input></p> <p><input type="checkbox" name="chckbox[]" value="bad" id="slideThree"/>Bad</input></p> <p><input type="checkbox" name="chckbox[]" value="beat it" id="slideThree"/>Beat it</input></p> <p><input type="checkbox" name="chckbox[]" value="thriller" id="slideThree"/>Thriller</input></p> <p> <input type="submit" name="submit" value="submit"/></p> </form> </body> </html>

9

Page 10: Report on Song Recommender

Screenshot of output of File1

File No­2 <html> <head></head> <body> <?php if(isset($_POST['chckbox'])) $a=$_POST['chckbox']; $n=count($a); $c=mysql_connect("localhost","root",""); mysql_select_db("ism",$c); mysql_query("delete from naive"); foreach($a as $key=>$val)

$b=mysql_query("select * from songs where song='$val'");

while($row=mysql_fetch_array($b))

10

Page 11: Report on Song Recommender

$p=$row['song']; $q=$row['artist']; $r=$row['genre']; $s=$row['eras'];

mysql_query("insert into naive(song,artist,genre,eras) values('$p','$q','$r','$s')"); if($n<7)

echo "<a href='more.php'>select more songs</a>"; else echo "<a href='cal.php'>survey review</a>"; ?> </body> </html> Screenshot of output of File2

11

Page 12: Report on Song Recommender

File No­3 <html> <head> </head> <body> <form name="songs" action="insert2.php" method="POST" > <p><input type="checkbox" name="chckbox[]" value="raabta" />raabta</input></p> <p><input type="checkbox" name="chckbox[]" value="chikni chameli"/>chikni chameli</input></p> <p><input type="checkbox" name="chckbox[]" value="deewana"/>deewana</input></p> <p><input type="checkbox" name="chckbox[]" value="kal ho na ho"/>kal ho na ho</input></p> <p><input type="checkbox" name="chckbox[]" value="dhoom"/>dhoom</input></p> <p><input type="checkbox" name=”chckbox[]" value="sur"/>sur</input></p> <p><input type="checkbox" name="chckbox[]" value="sajde"/>sajde</input></p> <p><input type="checkbox" name="chckbox[]" value="guzzarish"/>guzzarish</input></p> <p><input type="checkbox" name="chckbox[]" value="earth song"/>earth song</input></p> <p><input type="checkbox" name="chckbox[]" value="bille jean"/>bille jean</input></p> <p> <input type="submit" name="submit" value="submit"/></p> </form> </body> </html>

12

Page 13: Report on Song Recommender

Screenshot of output of File3

File No­4 <html> <head></head> <body> <?php if(isset($_POST['chckbox'])) $a=$_POST['chckbox'] $n=count($a); $c=mysql_connect("localhost","root",""); mysql_select_db("ism",$c);foreach($a as $key=>$val) $b=mysql_query("select * from songs where song='$val'"); while($row=mysql_fetch_array($b)) $p=$row['song']; $q=$row['artist']; $r=$row['genre']; $s=$row['eras'];

13

Page 14: Report on Song Recommender

mysql_query("insert into naive(song,artist,genre,eras) values('$p','$q','$r','$s')"); echo "<a href='cal.php'>survey review</a>"; ?> </body> </html> Screenshot of output of File4

14

Page 15: Report on Song Recommender

File No­5 <html> <head> <style type="text/css"> h3 color:white; </style> </head> <body style="background­color:black; background­repeat:repeat­x;background­position:center; "> <h3><i>After reviewing your likes, we suggest you the following songs which you may like..</i></h3> <?php $c=mysql_connect("localhost","root",""); mysql_select_db("ism",$c); $n=$n1=$n2=$n3=$n4=$n5=0; $q=mysql_query("SELECT COUNT(*) FROM naive"); $n=mysql_result($q,0); mysql_query("delete from display"); $q1=mysql_query("SELECT * FROM naive" ); $a=array("shreya goshal"=>"0","sonu nigam"=>"0","sunidhi chauhan"=>"0","KK"=>"0","MJ"=>"0"); while($row=mysql_fetch_array($q1)) foreach ($a as $singer=>$p1 ) $q8=mysql_query("SELECT COUNT(*) FROM naive WHERE artist='$singer' "); $rr=mysql_result($q8,0); $l=$rr/$n; $a[$singer]=$l; $yyy=arsort($a); $dups1=array(); $dups2=array(); $dups3=array(); foreach($a as $k1=>$v1) $a[$k1]=100*$v1; $s1=$a['shreya goshal']; $s2=$a['sonu nigam']; $s3=$a['sunidhi chauhan']; $s4=$a['KK']; $s5=$a['MJ'];

15

Page 16: Report on Song Recommender

echo "<script type='text/javascript' src='https://www.google.com/jsapi'></script> <script type='text/javascript'> google.load('visualization', '1', packages:['corechart']); google.setOnLoadCallback(drawChart); function drawChart() var data = google.visualization.arrayToDataTable([ ['artist', 'survey'], ['shreya goshal',$s1 ], ['sonu nigam',$s2], ['sunidhi chauhan',$s3], ['KK', $s4], ['MJ', $s5] ]); var options = title: 'Survey of Artist', backgroundColor : 'black', legend: textStyle: color: 'white', is3D: true, colors:['Red', 'Green','Yellow','Grey','Blue'] ; var chart = new google.visualization.PieChart(document.getElementById('piechart_3d')); chart.draw(data, options); </script>"; foreach($a as $key=>$val) if(isset($new_arr1[$val])) $new_arr1[$val]=$key; else if(isset($dups1[$val])) $dups1[$val][]=$key; else $dups1[$val]=array($key);

16

Page 17: Report on Song Recommender

krsort($dups1); $q2=mysql_query("SELECT * FROM naive" ); $b=array("romantic"=>"0","jazz"=>"0","rock"=>"0","saga"=>"0"); while($row=mysql_fetch_array($q2)) foreach ($b as $genree=>$p2 ) $q9=mysql_query("SELECT COUNT(*) FROM naive WHERE genre='$genree' "); $rr1=mysql_result($q9,0); $l1=$rr1/$n; $b[$genree]=$l1; $xxx=arsort($b); foreach($b as $k1=>$v1) $b[$k1]=100*$v1; $g1=$b['romantic']; $g2=$b['jazz']; $g3=$b['rock']; $g4=$b['saga']; echo "<script type='text/javascript' src='https://www.google.com/jsapi'></script> <script type='text/javascript'> google.load('visualization', '1', packages:['corechart']); google.setOnLoadCallback(drawChart); function drawChart() var data = google.visualization.arrayToDataTable([ ['genre', 'survey'], ['romantic', $g1 ], ['jazz', $g2], ['rock', $g3], ['saga', $g4] ]); var options = title: 'Survey of Genre', backgroundColor : 'black', legend: textStyle: color: 'white', is3D: true, colors:['Red', 'Green','Yellow','Grey','Blue'] ; var chart = new google.visualization.PieChart(document.getElementById('piechart1_3d')); chart.draw(data, options); </script>";

17

Page 18: Report on Song Recommender

foreach($b as $key=>$val) if(isset($new_arr2[$val])) $new_arr2[$val]=$key; else if(isset($dups2[$val])) $dups2[$val][]=$key; else $dups2[$val]=array($key); krsort($dups2); $q3=mysql_query("SELECT * FROM naive" ); $c=array("old"=>"0","new"=>"0"); while($row=mysql_fetch_array($q3)) foreach ($c as $erass=>$p3 ) $q10=mysql_query("SELECT COUNT(*) FROM naive WHERE eras='$erass' "); $rr3=mysql_result($q10,0); $l3=$rr3/$n; $c[$erass]=$l3; $zzz=arsort($c); foreach($c as $k1=>$v1) $c[$k1]=100*$v1; $e1=$c['old']; $e2=$c['new']; echo "<script type='text/javascript' src='https://www.google.com/jsapi'></script> <script type='text/javascript'> google.load('visualization', '1', packages:['corechart']); google.setOnLoadCallback(drawChart); function drawChart() var data = google.visualization.arrayToDataTable([ ['eras', 'survey'], ['old', $e1 ],

18

Page 19: Report on Song Recommender

['new', $e2] ]); var options = title: 'Survey of Eras', backgroundColor : 'black', legend: textStyle: color: 'white', is3D: true, colors:['Red', 'Green','Yellow','Grey','Blue'] ; var chart = new google.visualization.PieChart(document.getElementById('piechart2_3d')); chart.draw(data, options); </script>"; foreach($c as $key=>$val) if(isset($new_arr3[$val])) $new_arr3[$val]=$key; else if(isset($dups3[$val])) $dups3[$val][]=$key; else $dups3[$val]=array($key); krsort($dups3); foreach($dups1 as $key=>$val) if($key>=28) foreach($val as $key1) $w=mysql_query("select * from songs WHERE artist='$key1' and song not in(select song from naive) limit 10"); while($row=mysql_fetch_array($w)) $m1=$row['song'];

19

Page 20: Report on Song Recommender

mysql_query("insert into display(song) values('$m1')"); foreach($dups1 as $a=>$b) foreach($dups2 as $c=>$d) foreach($dups3 as $e=>$f) if($a>28 && $c>35 && $e>60) foreach($b as $p) foreach($d as $q) foreach($f as $r) $poo=mysql_query("select * from songs where artist='$p' and genre='$q' and eras='$r' and song not in(select song from naive) limit 4"); while($row=mysql_fetch_array($poo)) $m4=$row['song']; mysql_query("insert into display(song) values('$m4')"); ?> <form id="feedback" action="fb.php" method="POST" > <?php $dis=mysql_query("SELECT DISTINCT song FROM display"); $rows = mysql_num_rows($dis); while($row=mysql_fetch_array($dis)) echo "<p></p>";echo "<p style='color:white;'>".$row['song']."<input type='checkbox' name='a[]'/></p>"; echo "<p></p>"; echo "<input type='submit' name='submit' value='submit'/>";

20

Page 21: Report on Song Recommender

session_start(); $_SESSION["total"] = $rows ; ?> <i><p style='color:white;font­size:xx­large;text­align: center'>Survey of artist </p></i>; <div id="piechart_3d" style="width: 900px; height: 500px;"></div> <i><p style='color:white;font­size:xx­large;text­align: center'>Survey of genre</p></i>; <div id="piechart1_3d" style="width: 900px; height: 500px;"></div> <i><p style='color:white;font­size:xx­large;text­align: center'>Survey of eras</p></i>; <div id="piechart2_3d" style="width: 900px; height: 500px;"></div> </body> </html> Screenshot of output of File5

21

Page 22: Report on Song Recommender

22

Page 23: Report on Song Recommender

File No­6 <html> <head> </head> <body style="background­color:black; background­repeat:repeat­x;background­position:center; "> <i><p style='color:white;font­size:xx­large;text­align: center'>Feedback Result</p></i> <?php session_start(); if(isset($_POST['a'])) $a1=$_POST['a']; $n=count($a1); //echo $n; $t=$_SESSION["total"];

23

Page 24: Report on Song Recommender

echo "<p></p>"; //echo $t; echo "<p></p>"; $FDB=($n/$t)*100; echo "<p style='color:white;font­size:large'>The feedback system is $FDB% efficient</p>"; echo "<script type='text/javascript' src='https://www.google.com/jsapi'></script> <script type='text/javascript'> google.load('visualization', '1', packages:['corechart']); google.setOnLoadCallback(drawChart); function drawChart() var data = google.visualization.arrayToDataTable([ ['feedback', 'survey'], ['liked', $n ], ['not liked', $t­$n] ]); var options = title: 'Survey of Artist', backgroundColor : 'black', legend: textStyle: color: 'white', is3D: true, ; var chart = new google.visualization.PieChart(document.getElementById('piechart3_3d')); chart.draw(data, options); </script>"; ?> <div id="piechart3_3d" style="width: 900px; height: 500px;"></div> <a href="main.php" style="color:white;">Take another survey if you wish</a> </body> </html>

24

Page 25: Report on Song Recommender

Screenshot of output of File6

Importance Of Recommendation System

Predict the 'rating' or 'preference' that user would give to an item.

It can also help in predicting movies, music, news, books, research articles, search queries, social tags, and products in general.

It can also help in recommending experts, jokes, restaurants, financial services, life

insurance, persons (online dating), and twitter followers. Recommender system research is being conducted with a strong emphasis on practice and commercial applications, since, aside from its theoretical contribution, is generally aimed at practically improving commercial RSs. Thus, RS research involves practical aspects that apply to the implementation of these systems. These aspects are relevant to different stages in the life cycle of a RS, namely, the design of the system, its implementation and its maintenance and enhancement during system operation. The aspects that apply to the design stage include factors that might affect the

25

Page 26: Report on Song Recommender

choice of the algorithm. The first factor to consider, the application’s domain, has a major effect on the algorithmic approach that should be taken. [72] provide a taxonomy of RSs and classify existing RS applications to specific application domains. Based on these specific application domains, we define more general classes of domains for the most common recommender systems applications:

Entertainment ­ recommendations for movies, music, and IPTV.

Content ­ personalized newspapers, recommendation for documents, recommendations

of Web pages, e­learning applications, and e­mail filters.

E­commerce ­ recommendations for consumers of products to buy such as books, cameras, PCs etc.

Services ­ recommendations of travel services, recommendation of experts for consultation, recommendation of houses to rent, or matchmaking services. Bibliography Referred the following sites to collect information for doing this project :

http://www.wikipedia.org/

http://stackoverflow.com/

http://www.kdnuggets.com/

http://www.abbottanalytics.com/data­mining­resources­websites.php

https://www.youtube.com/watch?v=8yvBqhm92xA

https://www.youtube.com/watch?v=QMPBawsYR­I

26

Page 27: Report on Song Recommender

Acknowledgment We would like to express our special thanks of gratitude to our professor, Mr. George Philip. C who gave us this golden opportunity to do this wonderful report on the topic on Data Mining, which also helped us in doing a lot of Research and we came to know about so many new things we are really thankful to him. Secondly we would also like to the thank Mr.Vijaykant N who helped us a lot in finishing this project within the limited time.

1MS11IS044 Kruthika V +91­9731119104 [email protected]

1MS11IS063 Navya H C +91­9916986549 [email protected]

1MS11IS075 Pooja J +91­8971613674 [email protected]

1MS11IS081 Priyanka U Pandit T +91­8762315991 [email protected]

27