1 / 28

Classification of Semantic Relations in Noun Compounds using MeSH

Classification of Semantic Relations in Noun Compounds using MeSH. Marti Hearst, Barbara Rosario SIMS, UC Berkeley. LINDI Project Synopsis. Goal: Extract semantics from text Method: statistical corpus analysis Focus: BioMedical text Interesting inferences (Swanson) Rich lexical resources

Download Presentation

Classification of Semantic Relations in Noun Compounds using MeSH

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Classification of Semantic Relations in Noun Compounds using MeSH Marti Hearst, Barbara Rosario SIMS, UC Berkeley

  2. LINDI Project Synopsis • Goal: Extract semantics from text • Method: statistical corpus analysis • Focus: BioMedical text • Interesting inferences (Swanson) • Rich lexical resources • Difficult NLP problems • Noun Compounds

  3. Noun Compounds (NCs) • Any sequence of nouns that itself functions as a noun • asthma hospitalizations • asthma hospitalization rates • bone marrow aspiration needle • health care personnel hand wash • Technical text is rich with NCs Open-labeled long-term study of the subcutaneous sumatriptan efficacy and tolerability in acute migraine treatment.

  4. NCs: 3 computational tasks(Lauer & Dras ’94) • Identification • Syntactic analysis (attachments) • Baseline headache frequency • Tension headache patient • Semantic analysis • Headache treatment treatment for headache • Corticosteroid treatment treatment that uses corticosteroid [ ] [ ] [ ] [ ]

  5. NC Semantic Relations • Linguistic theories regarding the nature of the relations between constituents in NCs all conflict. • J. Levi ‘78 • P. Downing ’77 • B. Warren ‘78

  6. NC Semantic relations • 38 Relations found by iterative refinement based on 2245 NCs • Goals: • More specific than case roles • General enough to aid coverage • Allow for domain-specific relations

  7. Semantic relations • Examples • Frequency/time of • influenza season, headache interval • Measure of • relief rate, asthma mortality, hospital survival • Instrument • aciclovir therapy, laser irradiation, aerosol treatment • “Purpose” • headache drugs, voice therapy, influenza treatment • Defect • hormone deficiency, csf fistulas, gene mutation • Inhibitor • Adrenoreceptor blockers, influenza prevention

  8. Multi-class Assignment • Some NCs can be describe by more than one semantic relationships • eyelid abnormalities : location and defect • food allergy: cause and activator • cell growth: change and activity • tumor regression: change and ending/reduction

  9. Extraction of NCs • Titles and abstracts from Medline (medical bibliographic database) • Part of Speech Tagger • Extraction of sequences of units tagged as nouns • Collection of 2245 NCs with 2 nouns

  10. Models • Lexical (words) • headache pain • Class based model using MeSH descriptors for levels of descriptions • MeSH 2: C.23 G.11 • MeSH 3: C23.888 G11.561 • MeSH 4: C23.888.592 G11.561.796 • MeSH 5: C23.888.592 G11.561.796 • MeSH 6: C23.888.592.612 G11.561.796 .444

  11. MeSH Tree Structures 1. Anatomy [A] 2. Organisms [B] 3. Diseases [C] 4. Chemicals and Drugs [D] 5. Analytical, Diagnostic and Therapeutic Techniques and Equipment [E] 6. Psychiatry and Psychology [F] 7. Biological Sciences [G] 8. Physical Sciences [H] 9. Anthropology, Education, Sociology and Social Phenomena [I] 10. Technology and Food and Beverages [J] 11. Humanities [K] 12. Information Science [L] 13. Persons [M] 14. Health Care [N] 15. Geographic Locations [Z]

  12. 1. Anatomy [A] Body Regions [A01] + Musculoskeletal System [A02] Digestive System [A03] + Respiratory System [A04] + Urogenital System [A05] + Endocrine System [A06] + Cardiovascular System [A07] + Nervous System [A08] + Sense Organs [A09] + Tissues [A10] + Cells [A11] + Fluids and Secretions [A12] + Animal Structures [A13] + Stomatognathic System [A14] (…..) Body Regions [A01] Abdomen [A01.047] Groin [A01.047.365] Inguinal Canal [A01.047.412] Peritoneum [A01.047.596] + Umbilicus [A01.047.849] Axilla [A01.133] Back [A01.176] + Breast [A01.236] + Buttocks [A01.258] Extremities [A01.378] + Head [A01.456] + Neck [A01.598] (….) MeSH Tree Structures

  13. Mapping Nouns to MeSH Concepts • headache recurrence • C23.888.592.612.441 C23.550.291.937 • headache pain • C23.888.592.612.441 G11.561.796.444 • breast cancer cells • A01.236 C04 A11

  14. Levels of Description headache pain (C23.888.592.612.441 G11.561.796.444) • Only Tree: C G • C(Diseases) • G (Biological Sciences) • Level 1 : C 23 G 11 • C 23 (Diseases: Pathological Conditions) • G 11 (Biological Sciences: Musculoskeletal, Neural, and Ocular Physiology) • Level 2 : C 23 888 G 11 561 • C 23.888 (Diseases:Pathological Conditions: Signs and symptoms) • G 11.561 (Biological Sciences: Musculoskeletal, Neural, and Ocular Physiology:Nervous System Physiology) • Level 3 : C 23 888 592 G 11 561 796 • C 23.888.592 (Diseases :Pathological Conditions: Signs and symptoms: Neurologic Manifestations) • G 11.561.796 (Biological Sciences: Musculoskeletal, Neural, and Ocular Physiology:Nervous System Physiology:Sensation)

  15. Classification Task & Method • Multi-class (18) classification problem • Multi layer Neural Networks to classify across all relations simultaneously. • Evaluation: distinguish between • Seen: NCs where 1 or 2 words appeared in the training set • Unseen: NCs in which neither word appeared in the training set

  16. Lexical MeSH Accuracy for 18-way Classification Correct answer in first three (76%-78%) Correctanswer in first two (71%-73%) Correct answer ranked first (61%-62%) Baseline (guessing most frequent class)

  17. Accuracies for 18-way classification: generalization on unseen NCs MeSH MeSH on unseen Lexical Lexical on unseen

  18. Accuracies by Unseen Noun

  19. Accuracy for each relation

  20. Accuracy for sample relations Produces (genetic) Ex. Test Set: thymidine allele tumor dna csf mrna acetylase gene virion rna (…)

  21. Accuracy for sample relations Frequency/time of Test Set: disease recurrence headache recurrence enterovirus season influenza season mosquito season pollen season disease stage transcription stage drive time injection time ischemia time travel time

  22. Accuracy for sample relations Purpose Test Set: varicella vaccine tb vaccines poliovirus vaccine influenza vaccination influenza immunization abscess drainage acne therapy asthma therapy asthma treatment carcinogen treatment disease treatment hiv treatment

  23. Related work • Finin (1980) • Detailed AI analysis, hand-coded • Rindflesch et al. (2000) • Hand-coded rule base to extract certain types of assertions

  24. Related work • Vanderwende (1994) • automatically extracts semantic information from an on-line dictionary • manipulates a set of handwritten rules • 13 classes • 52% accuracy • Lapata (2000) • classifies nominalizations into subject/object binary distinction • 80% accuracy • Lauer (1995): • probabilistic model • 8 classes • 47% accuracy

  25. Related work • Prepositional Phrase Attachment • The problem • Eat spaghetti with a fork • Eat spaghetti with sauce • V N1 P N2 • Attachment/association, not semantics • Approaches • Word occurrences (Hindle & Rooth ’93) • Using a lexical hierarchy • Conceptual association (Resnik ’93, Resnik & Hearst ’93) • Transformation-based (Brill & Resnik ’94) • MDL to find optimal tree cut (Li & Abe ’98) • Lindi: use ML techniques to determine appropriate level of lexical hierarchy, classify into semantic relations

  26. Conclusions • A simple method for assigning semantic relations to noun compounds • Does not require complex hand-coded rules • Does make use of existing lexical resources • High accuracy levels for an 18-way class assignment • Small training set gets ~60% accuracy on mixed seen and unseen words • Tiny training set (73 NCs) gets ~40% accuracy on entirely unseen words • Off-the-shelf, unoptimized ML algorithms

  27. Future work • Analysis of cases where it doesn’t work • NC with > 2 terms • How to generalize patterns found for noun compounds to other syntactic structures? • How can we best formally represent semantics? • How can we deal with non medical words? • Should we use other ontologies (e.g.,WordNet)?

  28. Using Relations • Eventual plan: combine relations with constituents’ ontology memberships • Examples • Instrument_2 (biopsy,needle) -> Instrument_2(Diagnostic, Tool) • Procedure(brain,biopsy) -> Procedure(Anatomical-Element, Diagnostic) • Procedure(tumor, marker) -> Procedure(Disease-element, Indicator)

More Related