Measuring Semantic Similarity and Relatedness in the Biomedical Domain : Methods and Applications

Ted Pedersen, Ph.D. Department of Computer ScienceUniversity of Minnesota, Duluth

tpederse@d.umn.eduhttp://www.d.umn.edu/~tpederse

Topics

Semantic similarity vs. semantic relatedness

How to measure similarity With ontologies and corpora

How to measure relatednessWith definitions and corpora

Applications? Word Sense Disambiguation

Sentiment Classification

What are we measuring?

Concept pairsAssign a numeric value that quantifies how similar or related two concepts are

Not wordsMust know concept underlying a word form

Cold may be temperature or illnessConcept Mapping

Word Sense Disambiguation

This tutorial assumes that's been resolved

Why?

Being able to organize concepts by their similarity or relatedness to each other is a fundamental operation in the human mind, and in many problems in Natural Language Processing and Artificial Intelligence

If we know a lot about X, and if we know Y is similar to X, then a lot of what we know about X may apply to YUse X to explain or categorize Y

GOOD NEWS!
Free Open Source Software!

WordNet::Similarity http://wn-similarity.sourcforge.net

General English

Widely used (+750 citations)

UMLS::Similarityhttp://umls-similarity.sourceforge.net

Unified Medical Language System

Spun off from WordNet::SimilarityBut has added a whole lot!

Similar or Related?

Similarity based on is-a relations

How much is X like Y?

Share ancestor in is-a hierarchy LCS : least common subsumer

Closer / deeper the ancestor the more similar

Tetanus and strep throat are similarboth are kinds-of bacterial infections

Least Common Subsumer (LCS)

Similar or Related?

Relatedness more generalHow much is X related to Y?

Many ways to be relatedis-a, part-of, treats, affects, symptom-of, ...

Tetanus and deep cuts are related but they really aren't similar(deep cuts can cause tetanus)

All similar concepts are related, but not all related concepts are similar

Measures of Similarity
(WordNet::Similarity & UMLS::Similarity )

Path Based

Rada et al., 1989 (path)

Caviedes & Cimino, 2004 (cdist)*cdist only in UMLS::Similarity

Path + Depth

Wu & Palmer, 1994 (wup)

Leacock & Chodorow, 1998 (lch)

Zhong et al., 2002 (zhong)*

Nguyen & Al-Mubaid, 2006 (nam)*zhong and nam only in UMLS::Similarity

Measures of Similarity
(WordNet::Similarity & UMLS::Similarity)

Path + Information ContentResnik, 1995 (res)

Jiang & Conrath, 1997 (jcn)

Lin, 1998 (lin)

Path Based Measures

Distance between concepts (nodes) in tree intuitively appealing

Spatial orientation, good for networks or maps but not is-a hierarchiesReasonable approximation sometimes

Assumes all paths have same weight

But, more specific (deeper) paths tend to travel less semantic distance

Shortest path a good start, but needs corrections

Shortest is-a Path

1path(a,b) = ------------------------------ shortest is-a path(a,b)

We count nodes...

Maximum = 1 self similarity

path(tetanus,tetanus) = 1

Minimum = 1 / (longest path in is-a tree)path(typhoid, oral thrush) = 1/7

path(moccasin athlete's foot, strep throat) = 1/7

etc...

path(strep throat, tetanus) = .25

path (bacterial infection, yeast infection) = .25

?

Are bacterial infection and yeast infection similar to the same degree as are tetanus and strep throat ?

The path measure says yes, they are.

Path + Depth

Path only doesn't account for specificity

Deeper concepts more specific

Paths between deeper concepts travel less semantic distance

Wu and Palmer, 1994

2 * depth (LCS (a,b))wup(a,b) = ---------------------------- depth (a) + depth (b)

depth(x) = shortest is-a path(root,x)

wup(strep throat, tetanus) = (2*2)/(4+3) = .57

wup (bacterial infections, yeast infections) = (2*1)/(2+3) = .4

?

Wu and Palmer say that strep throat and tetanus (.57) are more similar than are bacterial infections and yeast infections (.4)

Path says that strep throat and tetanus (.25) are equally similar as are bacterial infections and yeast infections (.25)

Information Content

ic(concept) = -log p(concept) [Resnik 1995]Need to count concepts

Term frequency +Inherited frequency

p(concept) = tf + if / N

Depth shows specificity but not frequency

Low frequency concepts often much more specific than high frequency onesRelated to Zipf's Law of Meaning? (more frequent word have more senses)

Information Content
term frequency (tf)

Information Content
inherited frequency (if)

Information Content (IC = -log (f/N)
final count (f = tf + if, N = 365,820)

Lin, 1998

2 * IC (LCS (a,b))lin(a,b) = -------------------------- IC (a) + IC (b)

Look familiar?

Lin, 1998

2 * IC (LCS (a,b))lin(a,b) = -------------------------- IC (a) + IC (b)

Look familiar? 2* depth (LCS (a,b) )

wup(a,b) = ------------------------------ depth(a) + depth (b)

lin (strep throat, tetanus) =
2 * 2.26 / (5.21 + 4.11) = 0.485

lin (bacterial infection, yeast infection) =
2 * 0.71 / (2.26+2.81) = 0.280

?

Lin says that strep throat and tetanus (.49) are more similar than are bacterial infection and yeast infection (.28)

Wu and Palmer say that strep throat and tetanus (.57) are more similar than are bacterial infection and yeast infection (.4)

Path says that strep throat and tetanus (.25) are equally similar as are bacterial infection and yeast infection (.25)

How to decide??

Hierarchies best suited for nouns

If you have a hierarchy of concepts, shortest path can be distorted/misleading

If the hierarchy is carefully developed and well balanced, then wup can perform well

If the hierarchy is not balanced or unevenly developed, the information content measures can help correct that

What about concepts
not connected via is-a relations?

Connected via other relations?Part-of, treatment-of, causes, etc.

Not connected at all?In different sections (axes) of an ontology (infections and treatments)

In different ontologies entirely (SNOMEDCT and FMA)

Relatedness!Use definition information

No is-a relations so can't be similarity

Measures of relatedness

Path based

Hirst & St-Onge, 1998 (hso)

Definition based

Lesk, 1986

Adapted lesk (lesk)Banerjee & Pedersen, 2003

Definition + corpus

Gloss Vector (vector)Patwardhan & Pedersen, 2006

Path based relatedness

Ontologies include relations other than is-a

These can be used to find shortest paths between conceptsHowever, a path made up of different kinds of relations can lead to big semantic jumps

Aspirin treats headaches which are a symptom of the flu which can be prevented by a flu vaccine which is recommend for children . so aspirin and children are related ??

Measuring relatedness with definitions

Related concepts defined using many of the same terms

But, definitions are short, inconsistent

Concepts don't need to be connected via relations or paths to measure themLesk, 1986

Adapted Lesk, Banerjee & Pedersen, 2003

Two separate ontologies...

Could join them together ?

Each concept has definition

Find overlaps in definitions...

Overlaps

Oral Thrush and Alopeciaside effect of chemotherapyCan't see this in structure of is-a hierarchies

Oral thrush and folliculitis just as similar

Alopecia and Folliculitis hair disorder & hairReflects structure of is-a hierarchies

If you start with text like this maybe you can build is-a hierarchies automatically!Future work...

Lesk and Adapted Lesk

Lesk, 1986 : measure overlaps in definitions to assign senses to wordsThe more overlaps between two senses (concepts), the more related

Banerjee & Pedersen, 2003, Adapted LeskAugment definition of each concept with definitions of related conceptsBuild a super gloss

Increase chance of finding overlaps

lesk in WordNet::Similarity & UMLS::Similarity

The problem with definitions ...

Definitions contain variations of terminology that make it impossible to find exact overlaps

Alopecia : a result of cancer treatment

Thrush : a side effect of chemotherapyReal life example, I modified the alopecia definition to work better with Lesk!!!

NO MATCHES!!

How can we see that result and side effect are similar, as are cancer treatment and chemotherapy ?

Gloss Vector Measure
of Semantic Relatedness

Rely on co-occurrences of terms Terms that occur within some given number of terms of each other

Allows for a fuzzier notion of matching

Exploits second order co-occurrencesFriend of a friend relation

Suppose cancer_treatment and chemotherapy don't occur in text with each other. But, suppose that survival occurs with each.

cancer_treatment and chemotherapy are second order co-occurrences via survival

Gloss Vector Measure
of Semantic Relatedness

Replace words or terms in definitions with vector of co-occurrences observed in corpus

Defined concept now represented by an averaged vector of co-occurrences

Measure relatedness of concepts via cosine between their respective vectors

Patwardhan and Pedersen, 2006 (EACL)Inspired by Schutze, 1998 (CL)

vector in WordNet::Similarity & UMLS::Similarity

Experimental Results

Vector > Lesk > Info Content > Depth > PathClear trend across various studies

Dramatic differences when comparing to human reference standards (Vector > Lesk >> Info Content > Depth > Path)Banerjee and Pedersen, 2003 (IJCAI)

Pedersen, et al. 2007 (JBI)

Differences less extreme in extrinsic task-based evaluations Human raters mix up similarity & relatedness?

So far we've shown that ...

we can quantify the similarity and relatedness between concepts using a variety of sources of informationPaths

Depths

Information content

Definitions

Co-occurrence / corpus data

There is open source software to help you!

Sounds great! What now?

SenseRelate Hypothesis : Most words in text will have multiple possible senses and will often be used with the sense most related to those of surrounding wordsHe either has a cold or the fluCold not likely to mean air temperature

The underlying sentiment of a text can be discovered by determining which emotion is most related to the words in that text I cried a lot after my mother died. Happy?

SenseRelate!

In coherent text words will be used in similar or related senses, and these will also be related to the overall topic or mood of a text

First applied to WSD in 2002Banerjee and Pedersen, 2002 (WordNet)

Patwardhan et al., 2003 (WordNet)

Pedersen and Kolhatkar 2009 (WordNet)

McInnes et al., 2011 (UMLS)

Recently applied to emotion classificationPedersen, 2012 (i2b2 suicide notes challenge)

GOOD NEWS!
Free Open Source Software!

WordNet::SenseRelateAllWords, TargetWord, WordToSet

http://senserelate.sourceforge.net

UMLS::SenseRelateAllWords

http://search.cpan.org/dist/UMLS-SenseRelate/

SenseRelate for WSD

Assign each word the sense which is most similar or related to one or more of its neighborsPairwise

2 or more neighbors

Pairwise algorithm results in a trellis much like in HMMsMore neighbors adds lots of information and a lot of computational complexity

SenseRelate - pairwise

SenseRelate 2 neighbors

General Observations on WSD Results

Nouns more accurate; verbs, adjectives, and adverbs less so

Increasing the window size nearly always improves performance

Jiang-Conrath measure often a high performer for nouns (e.g., Patwardhan et al. 2003)

Info content measures perform well with clinical text (McInnes et al. 2011)

Vector and lesk have coverage advantagehandle mixed pairs while others don't

Recent Specific Experiment

Compare efficacy of different measures when performing WSD using UMLS::SenseRelate

Evaluate on MSH-WSD data (from NLM)

Information Content based on concept counts from Medline (UMLSonMedline, from NLM)

More details available McInnes, et al. 2011 (AMIA)

McInnes & Pedersen, in review

MSH-WSD data set

Contains 203 ambiguous terms and acronymsInstances are from Medline

CUIs from 2009 AB version of UMLS

Each word has avg. 187 instances, 2.08 possible senses, and 54.5% majority sense

Leverages fact that MedLine is manually indexed with Medical Subject Headings (associated with CUIs)

http://wsd.nlm.nih.gov/collaboration.shtml

Results

SenseRelate for
Sentiment Classification

Find emotion most related to contextSimilarity less effective since many words can be related an emotion, but fewer are similar Related to happy? : love, food, success, ...

Similar to happy? : joyful, ecstatic, pleased,

Pairwise comparisons between emotion and senses of words in context

Same form as Naive Bayesian model or Latent Variable modelWordNet::SenseRelate::WordToSet

SenseRelate - WordToSet

Experimental Results

Sentiment classification results in 2011 i2b2 suicide notes challenge were disappointing (Pedersen, 2012)Suicide notes not very emotional!

In many cases reflect a decision made and focus on settling affairs

Future Work

Find new domains and types of problemsEHR, clinical records,

Integrate Unsupervised Clustering with WordNet::Similarity and UMLS::Similarityhttp://senseclusters.sourceforge.net

Exploit graphical nature of of SenseRelatee.g., Minimal Spanning Trees / Viterbi Algorithm to solve larger problem spaces?

Attract and support users for all of these tools!

UMLS::Similarity Collaborators

Serguei Pakhomov : Assoc. Professor, UMTC

Bridget McInnes :PhD UMTC, 2009

Post-doc UMTC, 2009 - 2011

Now at Securboration, NC

Ying Liu :PhD UAB, 2007

Post-doc UMTC 2009 2011

Until recently at City of Hope, LA

Acknowledgments

This work on semantic similarity and relatedness has been supported by a National Science Foundation CAREER award (2001 2007, #0092784, PI Pedersen) and by the National Library of Medicine, National Institutes of Health (2008 2012, 1R01LM009623-01A2, PI Pakhomov)

The contents of this talk are solely my responsibility and do not necessarily represent the ocial views of the National Science Foundation or the National Institutes of Health.

Conclusion

Measures of semantic similarity and relatedness are supported by a rich body of theory, and open source softwarehttp://wn-similarity.sourceforge.net

http://umls-similarity.sourceforge.nethttp://atlas.ahc.umn.edu

These measures can be used as building blocks for many NLP and AI applicationsWord sense disambiguation

Sentiment classification

References

S. Banerjee and T. Pedersen. An adapted Lesk algorithm for word sense disambiguation using WordNet. In Proceedings of the Third International Conference on Intelligent Text Processing and Computational Linguistics, pages 136145, Mexico City, February 2002.

S. Banerjee and T. Pedersen. Extended gloss overlaps as a measure of semantic relatedness. In Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, pages 805-810, Acapulco, August 2003.

J. Caviedes and J. Cimino. Towards the development of a conceptual distance metric for the UMLS. Journal of Biomedical Informatics, 37(2):77-85, April 2004.

J. Jiang and D. Conrath. Semantic similarity based on corpus statistics and lexical taxonomy. In Proceedings on International Conference on Research in Computational Linguistics, pages 19-33, Taiwan, 1997.

References

C. Leacock and M. Chodorow. Combining local context and WordNet similarity for word sense identification. In C. Fellbaum, editor, WordNet: An electronic lexical database, pages 265-283. MIT Press, 1998.

M.E. Lesk. Automatic sense disambiguation using machine readable dictionaries: how to tell a pine code from an ice cream cone. In Proceedings of the 5th annual international conference on Systems documentation, pages 24-26. ACM Press, 1986.

D. Lin. An information-theoretic definition of similarity. In Proceedings of the International Conference on Machine Learning, Madison, August 1998.

B. McInnes, T. Pedersen, Y. Liu, G. Melton and S. Pakhomov. Knowledge-based Method for Determining the Meaning of Ambiguous Biomedical Terms Using Information Content Measures of Similarity. Appears in the Proceedings of the Annual Symposium of the American Medical Informatics Association, pages 895-904, Washington, DC, October 2011.

References

H.A. Nguyen and H. Al-Mubaid. New ontology-based semantic similarity measure for the biomedical domain. In Proceedings of the IEEE International Conference on Granular Computing, pages 623-628, Atlanta, GA, May 2006.

S. Patwardhan, S. Banerjee, and T. Pedersen. Using measures of semantic relatedness for word sense disambiguation. In roceedings of the Fourth International Conference on Intelligent Text Processing and Computational Linguistics, pages 241257, Mexico City, February 2003.

S. Patwardhan and T. Pedersen. Using WordNet-based Context Vectors to Estimate the Semantic Relatedness of Concepts. In Proceedings of the EACL 2006 Workshop on Making Sense of Sense: Bringing Computational Linguistics and Psycholinguistics Together, pages 1-8, Trento, Italy, April 2006.

T. Pedersen. Rule-based and lightly supervised methods to predict emotions in suicide notes. Biomedical Informatics Insights, 2012:5 (Suppl. 1):185-193, January 2012.

References

T. Pedersen and V. Kolhatkar. WordNet :: SenseRelate :: AllWords - a broad coverage word sense tagger that maximizes semantic relatedness. In Proceedings of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies 2009 Conference, pages 17-20, Boulder, CO, June 2009.

T. Pedersen, S. Pakhomov, S. Patwardhan, and C. Chute. Measures of semantic similarity and relatedness in the biomedical domain. Journal of Biomedical Informatics, 40(3) : 288-299, June 2007.

R. Rada, H. Mili, E. Bicknell, and M. Blettner. Development and application of a metric on semantic nets. IEEE Transactions on Systems, Man and Cybernetics, 19(1):17-30, 1989.

References

P. Resnik. Using information content to evaluate semantic similarity in a taxonomy. In Proceedings of the 14th International Joint Conference on Artificial Intelligence, pages 448-453, Montreal, August 1995.

H. Schtze. Automatic word sense discrimination. Computational Linguistics, 24(1):97-123, 1998.

J. Zhong, H. Zhu, J. Li, and Y. Yu. Conceptual graph matching for semantic search. Proceedings of the 10th International Conference on Conceptual Structures, pages 92-106, 2002.

Click to edit the title text format

WindowsizePath based Information ContentRelatedness

pathwupjcnlinleskvector

2.63.63.65.65.67.68

5.66.67.68.69.68.68

10.68.69.70.71.68.67

25.70.70.73.74.68.65