K NOWLEDGE - BASED M ETHOD FOR D ETERMINING THE M EANING OF A MBIGUOUS B IOMEDICAL T ERMS U SING I...

KNOWLEDGE-BASED METHOD FOR DETERMINING THE MEANING OF AMBIGUOUS BIOMEDICAL TERMS USING INFORMATION CONTENT MEASURES OF SIMILARITY

Bridget McInnes

Ted Pedersen

Ying Liu

Genevieve B. Melton

Serguei Pakhomov

OBJECTIVE OF THIS WORK

Develop and evaluate a method than can disambiguate terms in biomedical text by exploiting similarity information extrapolated from the Unified Medical

Language System

Evaluate the efficacy of Information Content-based similarity measures over path-based similarity measures

for Word Sense Disambiguation, WSD

WORD SENSE DISAMBIGUATION

Word sense disambiguation is the task of determining the appropriate sense of a term

given context in which it is used.

TERM: tolerance

DrugTolerance

ImmuneTolerance

Word sense disambiguation is the task of determining the appropriate sense of a term

given context in which it is used.

Busprione attenuates tolerance to morphine in mice with skin cancer

DrugTolerance

ImmuneTolerance

SENSE INVENTORY: UNIFIED MEDICAL LANGUAGE SYSTEM

Unified Medical Language Sources (UMLS) Semantic Network Metathesaurus

~1.7 million biomedical and clinical concepts; integrated semi-automatically

CUIs (Concept Unique Identifiers), linked: Hierarchical: PAR/CHD and RB/RN Non-hierarchical: SIB, RO

Sources viewed together or independently Medical Subject Heading (MSH)

SPECIALIST Lexicon Biomedical and clinical terms, including variants

DrugTolerance: C0013220

ImmuneTolerance:C0020963

Concept Unique Identifiers: CUIs

SENSERELATE ALGORITHM

Each possible sense of a target word is assigned a score [sum similarity between it and its surrounding terms]

Assign target word the sense with highest score

Proposed by Patwardhan and Pedersen 2003 using WordNet

UMLS::SenseRelate is a modification of this algorithm using information from the UMLS

NEXT UP: an example

SENSERELATE EXAMPLEBusprione attenuates tolerance to morphine

in mice with skin cancer

Busprione:

C0006462

Morphine:C0026549

Mice: C0026809

Skin cancer:

C0007114

SENSERELATE EXAMPLE

0.090.16

Busprione:

C0006462

Morphine:C0026549

Mice: C0026809

Skin cancer:

C0007114

SENSERELATE EXAMPLE

0.090.16

Busprione:

C0006462

Morphine:C0026549

Mice: C0026809

Skin cancer:

C0007114

Drug ToleranceScore = 0.09 + 0.09 + 0.16 + 0.11 = 0.45

SENSERELATE EXAMPLE

0.090.16

0.090.050.04

0.090.09

Busprione:

C0006462

Morphine:C0026549

Mice: C0026809

Skin cancer:

C0007114

SENSERELATE EXAMPLE

0.090.16

0.090.050.04

0.090.09

Busprione:

C0006462

Morphine:C0026549

Mice: C0026809

Skin cancer:

C0007114

Immune ToleranceScore = 0.09 + 0.09 + 0.05 + 0.05 = 0.27

SENSERELATE EXAMPLE

0.090.16

0.090.050.04

0.090.09

Busprione:

C0006462

Morphine:C0026549

Mice: C0026809

Skin cancer:

C0007114

Immune ToleranceScore = 0.09 + 0.09 + 0.05 + 0.05 = 0.27

SENSE RELATE ASSUMPTION

An ambiguous word is often used in the sense

that is most similar to the sense of the terms that surround it

SENSERELATE COMPONENTS

Identifying the concepts of surrounding terms

Calculating semantic similarity

IDENTIFYING THE CONCEPTS OF THE SURROUNDING TERMS

Use the SPECIALIST LEXICON to identify the terms and map the terms doing a string match to the MRCONSO table in

the UMLS

the UMLSBusprione attenuates tolerance to morphine

the UMLS

...skin cancerskin graftingskin disease

SPECIALISTLEXICON

the UMLS

...skin cancerskin graftingskin disease

...skin cancer C0007114skin grafting C0037297skin disease C0037274

SPECIALISTLEXICON

MRCONSO

SEMANTIC SIMILARITY MEASURES

Path-based measures Path Wu and Palmer Leacock and Chodorow Ngyuen and Al-Mubaid

Information content (IC)-based measures Resnik Lin Jiang and Conrath

PATH-BASED SIMILARITY MEASURES

Use only the path information obtained from a taxonomy

Path measure sim(c1,c2) = 1 / minpath(c2,c2)

where minpath is the shortest path between the two concepts

Path measure sim(c1,c2) = 1/minpath(c2,c2)

Wu and Palmer, 1994 sim(c1,c2) = (2*depth(LCS(c2,c2))) /

(depth(c1)+depth(c2)) where LCS is the least common subsumer of the

two concepts

Path measure sim(c1,c2) = 1/ minpath(c2,c2)

Wu and Palmer, 1994 sim(c1,c2) = (2*depth(LCS(c2,c2))) / (depth(c1)+depth(c2))

where LCS is the least common subsumer of the two concepts

Leacock and Chodorow, 1998 sim(c1,c2) = -log( minpath(c1,c2) / (2D) )

where D is the total depth of the taxonomy

PATH-BASED SIMILARITY MEASURES Use only the path information obtained from a taxonomy

Path measure sim(c1,c2) = 1/ minpath(c2,c2)

Leacock and Chodorow, 1998 sim(c1,c2) = -log( minpath(c1,c2) / (2D) )

where D is the total depth of the taxonomy

Wu and Palmer, 1994 sim(c1,c2) = (2*depth(LCS(c2,c2))) / (depth(c1)+depth(c2))

where LCS is the least common subsumer of the two concepts

Nyguen and Al-Mubaid, 2006 sim(c1,c2) = log ( (2 + minpath(c1,c2) - 1) *

(D - depth(LCS(c1,c2))) )

USE ONLY THE PATH INFORMATION OBTAINED FROM A TAXONOMY

Disease: C0012634

Drug Related

Disorder: C0277579

Neoplasm: C1302761

Neoplastic Disease: C1882062

Malignant Neoplasm: C0006826

Skin cancer:

C0007114

INFORMATION CONTENT-BASED MEASURES

Incorporate the probability of the concepts

IC = -log(P(concept))

P(concept)

Calculated by summing the probability of the concept and the probability of its descendants

Probabilities are obtained from an external corpus

IC = -log(P(concept)

Resnik, 1995 sim(c1,c2) = IC(LCS(c1,c2))

Jiang and Conrath, 1997 sim(c1,c2) = 1 / (IC(c1)+IC(c2) – 2*

IC(LCS(c1,c2))

Jiang and Conrath, 1997 sim(c1,c2) = 1 ÷ (IC(c1)+IC(c2) – 2* IC(LCS(c1,c2))

Lin, 1998 sim(c1,c2) = (2*IC(LCS(c2,c2))) / (IC(c1)+IC(c2))

IC-BASED SIMILARITY MEASURES

Disease: C001263

Related Disorder: C0277579Drug

Tolerance:

C0013220

Neoplasm:

C1302761Neoplasti

c Disease: C188206

2Malignan

t Neoplas

m: C000682

6Skin cancer: C000711

PATH INFORMATIONPROBABILITY OF

CONCEPTS

EXTERNAL CORPUS

EXPERIMENTAL FRAMEWORK

Use open-source UMLS::Similarity package to obtain the similarity between the terms and possible senses in the SenseRelate algorithm

Path information: parent/child relations in MSH source

Information content: calculated using the UMLSonMedline dataset created by NLM

Consists of concepts from 2009AB UMLS and the frequency they occurred in Medline using the Essie Search Engine (Ide et al 2007)

Medline: database of citations of biomedical/clinical articles

EVALUATION DATA: MSH WSD

MSH-WSD dataset (Jimeno-Yepes, et al 2011) 203 target words (ambiguous word) from Medline

106 terms e.g. tolerance 88 acronyms e.g. CA (calcium, california) 9 mixtures e.g. bat (brown adipose tissue)

Each target word contains ~187 instances (Medline abstracts) abstract = ~ 500 words

Each target word in the instances assigned a concept from MSH by exploiting the manually assigned MSH concepts assigned to the abstract

Average of 2.08 possible senses per target word Majority sense over all the target words is 54.5%

RESULTS

0.7200000000000010.690000000

0000010.700000000

000001

0.720000000000001

0.730000000000001

0.740000000000001

baseline path

lch wup

res jcn

accuracy

Path-based IC-based

COMPARISON ACROSS SUBSETS OF MSH-WSD

Terms Acronyms Mixture Overall0

0.55 0.54 0.53 0.55

0.670000000000002

0.8 0.730000000000001

0.7400000000000020.71000000000

0.870000000000002 0.88

0.670000000000002

0.850000000000001

BaselineSenseRe-lateMRD2-MRD

accuracy

0.55 0.54 0.53 0.55

0.670000000000002

0.8 0.730000000000001

0.7400000000000020.71000000000

0.870000000000002 0.88

0.670000000000002

0.850000000000001

accuracy

0.55 0.54 0.53 0.55

0.670000000000002

0.8 0.730000000000001

0.7400000000000020.71000000000

0.870000000000002 0.88

0.670000000000002

0.850000000000001

accuracy

0.55 0.54 0.53 0.55

0.670000000000002

0.8 0.730000000000001

0.7400000000000020.71000000000

0.870000000000002 0.88

0.670000000000002

0.850000000000001

accuracy

0.55 0.54 0.53 0.55

0.670000000000002

0.8 0.730000000000001

0.7400000000000020.71000000000

0.870000000000002 0.88

0.670000000000002

0.850000000000001

accuracy

WINDOW SIZES

Use the terms surrounding the target word within a specified window: 1, 2, 5, 10, 25, 50, 60, 70

Busprione attenuates tolerance to morphine in mice with skin_cancer

WINDOW SIZE = 2

COMPARISON OF WINDOW SIZES FOR LIN

0 1 2 5 10 25 50 60 700

0.50.53

0.650000000000002

0.690000000000001

0.710000000000001

0.740000000000002

accuracy

window size

SURROUNDING TERMS

Not all terms have a concept in the UMLS

therefore

Not all surrounding terms in the window mapped to CUIs

WINDOW SIZES VERSUS MAPPED TERMS

0 1 2 5 10 25 50 60 700

0 0.27 0.79

number

mappings

window size

FUTURE WORK: MAPPING TERMS

Currently looking at mapping the terms to CUIs using information from the concept mapping system MetaMap

Obtain the terms from MetaMap and do a dictionary look up in MRCONSO Hypothesis – the terms obtained by MetaMap are more

accurate than using the SPECIALIST Lexicon

Obtain the CUIs from MetaMap Hypothesis – the CUIs obtained by MetaMap will be

more accurate than the dictionary look-up

OBJECTIVE #1

Develop and evaluate a method than can disambiguate terms in biomedical text by

exploiting similarity information extrapolated from the UMLS

UMLS::SenseRelate statistically significantly higher disambiguation accuracy than the baseline

On par with previous unsupervised methods for terms

OBJECTIVE #2

Evaluate the efficacy of IC-based similarity measures over path-based measures on a

secondary task

There is no statistically significant difference between the accuracies obtained by the IC-based measures

There is a statistically significant difference between the IC-based measures and the path-based measures

TAKE HOME MESSAGE:

An ambiguous word is often used in the sense

that is most similar to the sense of the concepts

of the terms that surround it

RESOURCES

Software: UMLS::SenseRelate

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

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

Data MSH-WSD

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

RESOURCES

Data MSH-WSD

THANK YOU

RESOURCES

Data MSH-WSD

QUESTIONS?

K NOWLEDGE - BASED M ETHOD FOR D ETERMINING THE M EANING OF A MBIGUOUS B IOMEDICAL T ERMS U SING I...

Documents

Transcript of K NOWLEDGE - BASED M ETHOD FOR D ETERMINING THE M EANING OF A MBIGUOUS B IOMEDICAL T ERMS U SING I...

Determining Attendance and Alternatives to Seat-Time · 2020. 11. 12. · 2 etermining Attendance and Alternatives to Seat-Time States can ask districts to develop policies to determine

Proving Triangles are Similar 8.5 Proving Triangles are Similar 8.5 Chapter 8 Similarity Section 8.5 Proving Triangles are Similar U SING S IMILARITY.

D ON ' T M ATCH T WICE : R EDUNDANCY - FREE S IMILARITY C OMPUTATION WITH M AP R EDUCE Lars Kolb, Andreas Thor, Erhard Rahm Database Group Leipzig University.

W い い ぃ W d - Taketoyo...m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m

ETERMINING WITHDRAWAL RATES USING HISTORICAL DATAkyestates.com/wp-content/uploads/2015/02/Bengen1.pdf · 2017-04-04 · Number of Years Portfolio Assets Will Last (Withdrawals Vary

m m m m m m m m m m m m m m m m m m m m m ... - CFA …

A S CENARIO A GGREGATION –B ASED A PPROACH FOR D ETERMINING A R OBUST A IRLINE F LEET C OMPOSITION FOR D YNAMIC C APACITY A LLOCATION Ovidiu Listes, Rommert.

m m m m m m m m m m m m m m m m m m m m m m m mPart III · m m m m m m m m m m m m m m m m m m m m m m m mPart III Statement of Program Service Accomplishments Check if Schedule O

Lessons in Fijian - LIVE LINGUA Corps Lessons... · of Spanl-h or Latin AlthoUgh'fOur of the Fijian ioweIs sound. A. But there. the. ame vowel imilarity ends. to the Engli-h vowels

UNITED STATES DISTRICT COURT NORTHERN …€¦ · Due Process Clause of the Fifth ... a ft er d etermining tha t Bl ak ely did no t a pp ly to t he U. S. Se nte nc ing ... serving

Configura JumpStarthink MOVE vs. Sell! elationships matter imilarity works (in Biz) valuate others

F IRST V ERSUS N ON -F IRST G ENERATION S TUDENTS : D ETERMINING V ARIABLES OF A CADEMIC S UCCESS Michelle Coffman Heather Osterman Hanover College.

G-36-1020 Real Estate Agent Checklist for etermining if a ... · Web viewFlat Plate A metal collector plate on the roof The plate is usually copper or aluminum and is coated with

NPR FOUNDATION FORM 990-T EXEMPT … deduction (Generally $1,000, but see line 33 instructions for exceptions) m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m

C ORRECTING FOR I NDIRECT R ANGE R ESTRICTION IN M ETA -A NALYSIS : D ETERMINING THE U T D ISTRIBUTION Huy Le University of Central Florida In-Sue Oh University.

BONUS! 7 INSIDER TIPS ON NEW HOME CONSTRUCTION …7 INSIDER TIPS ON NEW HOME CONSTRUCTION COSTS D etermining new home construction costs can be a tricky business. With so many vari-

U SING S IMILARITY T HEOREMS THEOREM S THEOREM 8.2 Side-Side-Side (SSS) Similarity Theorem If the corresponding sides of two triangles are proportional,

Free-Gait of Quadruped Robots Using Neural Networks for Etermining the Order of Swing Leg

ETERMINING WITHDRAWAL RATES USING HISTORICAL DATA · recession was the most devastating be- cause it occurred during a period of high inflation. Not only did investors suffer large

D ETERMINING E LIGIBILITY FOR M C K INNEY -V ENTO R IGHTS AND S ERVICES National Center for Homeless Education (800) 308-2145 homeless@serve.org.

W いいぃ W d - Taketoyo...m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m