Schema-Driven Relationship Extraction from Unstructured Text

Transcription

1 Wright State University CORE Scholar Kno.e.sis Publications The Ohio Center of Excellence in Knowledge- Enabled Computing (Kno.e.sis) 2007 Schema-Driven Relationship Extraction from Unstructured Text Cartic Ramakrishnan Wright State University - Main Campus Follow this and additional works at: Part of the Bioinformatics Commons, Communication Technology and New Media Commons, Databases and Information Systems Commons, OS and Networks Commons, and the Science and Technology Studies Commons Repository Citation Ramakrishnan, C. (2007). Schema-Driven Relationship Extraction from Unstructured Text.. This Presentation is brought to you for free and open access by the The Ohio Center of Excellence in Knowledge-Enabled Computing (Kno.e.sis) at CORE Scholar. It has been accepted for inclusion in Kno.e.sis Publications by an authorized administrator of CORE Scholar. For more information, please contact corescholar@

2 Schema-Driven Relationship Extraction from Unstructured Text Cartic Ramakrishnan Kno.e.sis Center, Wright State University, Dayton, OH

3 Outline Motivation Problem Description & Approach Results Future Work

4 Anecdotal Example mentioned_in UNDISCOVERED PUBLIC KNOWLEDGE Discovering connections hidden in text Harry Potter Nicolas Flammel mentioned_in Nicolas Poussin The Hunchback of Notre Dame member_of painted_by Holy Blood, Holy Grail written_by Victor Hugo member_of Priory of Sion cryptic_motto_of Et in Arcadia Ego The Da Vinci code mentioned_in member_of displayed_at mentioned_in Leonardo Da Vinci painted_by painted_by painted_by The Last Supper The Mona Lisa displayed_at The Louvre displayed_at The Vitruvian man Santa Maria delle Grazie

5 Motivation 1 Undiscovered Public knowledge in biology Migraine Stress? Calcium Channel Blockers Magnesium Swanson s Discoveries Spreading Cortical Depression PubMed Associations Discovered based on keyword searches These associations were discovered in 1986 followed by manually analysis of text to establish possible relevant relationships

6 Motivation 2 - Hypothesis Driven retrieval of Scientific Literature Migraine affects Stress inhibit isa Magnesium Patient Calcium Channel Blockers Keyword query: Migraine[MH] + Magnesium[MH] Complex Query PubMed Supporting Document sets retrieved

7 Motivation 3 -- Growth Rate of Public Knowledge Data captured per year = 1 exabyte (10 18 ) (Eric Neumann, Science, 2005) How much is that? Compare it to the estimate of the total words ever spoken by humans = 12 exabyte A small but significant portion is text data PubMed 16 Million abstracts MedlinePlus health information OMIM catalog of human genes and genetic disorders Undiscovered public knowledge may have also increased by a large amount

8 Our past work in Connection Discovery Semantic Associations over RDF graphs Discovery and Ranking Semantically Connected affects Migraine It is therefore critical to bridge the gap between unstructured and structured data Magnesium Assumption: by extracting Rich entities Semantic and relationships Stress Metadata inhibit containing between resulting entities isa related in semantic by a diverse set metadata of relationships Patient Calcium Channel Blockers

9 Outline Motivation Problem Description & Approach Results Future Work

10 Problem Extracting relationships between MeSH terms from PubMed Biologically active substance complicates causes affects UMLS Semantic Network Lipid causes affects Disease or Syndrome instance_of Fish Oils??????? instance_of Raynaud s Disease MeSH 9284 documents 5 documents 4733 documents PubMed

11 Background knowledge used UMLS A high level schema of the biomedical domain 136 classes and 49 relationships Synonyms of all relationship using variant lookup (tools from NLM) 49 relationship + their synonyms = ~350 mostly verbs MeSH T147 effect 22,000+ topics organized as a forest of 16 trees Used to query PubMed PubMed Over 16 million abstract Abstracts annotated with one or more MeSH terms T147 induce T147 etiology T147 cause T147 effecting T147 induced

12 Method Parse Sentences in PubMed SS-Tagger (University of Tokyo) SS-Parser (University of Tokyo) Entities (MeSH terms) in sentences occur in modified forms (TOP (S adenomatous (NP (NP (DT An) modifies (JJ excessive) hyperplasia (ADJP (JJ endogenous) (CC or) (JJ exogenous) An excessive ) (NN stimulation) endogenous ) (PP or exogenous (IN by) (NP stimulation (NN estrogen) modifies ) ) ) (VP (VBZ induces) estrogen (NP (NP (JJ adenomatous) (NN hyperplasia) ) (PP (IN of) (NP (DT the) Entities (NN endometrium) can also occur ) as ) ) composites ) ) ) of 2 or more other entities adenomatous hyperplasia and endometrium occur as adenomatous hyperplasia of the endometrium

13 Method Identify entities and Relationships in Parse Tree DT the NP NP JJ excessive ADJP NN stimulation JJ endogenous CC or JJ exogenous IN by PP TOP S NN estrogen VBZ induces VP JJ adenomatous NP NP NN hyperplasia Modifiers Modified entities Composite Entities IN of PP DT the NP NN endometrium

14 Entities The simple, the modified and the composite To capture the various types of entities we define Simple entities as MeSH terms Modifiers as siblings of entities that are Determiners Y induces no X Noun Phrases An excessive endogenous or exogenous stimulation Adjective phrases adenomatous Prepositional phrases M is induced by the X in the Z Modified Entities as any entity that has a sibling which is a modifier Composite Entity as any entity that has another entity as a sibling

15 Resulting RDF adenomatous hyperplasia hasmodifier An excessive endogenous or exogenous stimulation hasmodifier modified_entity2 haspart haspart modified_entity1 induces composite_entity1 haspart estrogen haspart Modifiers Modified entities Composite Entities endometrium

16 Outline Motivation Approach Results Future Work

17 Results Dataset 1 Swanson s discoveries Associations between Migraine and Magnesium [Hearst99] stress is associated with migraines stress can lead to loss of magnesium calcium channel blockers prevent some migraines magnesium is a natural calcium channel blocker spreading cortical depression (SCD) is implicated in some migraines high levels of magnesium inhibit SCD migraine patients have high platelet aggregability magnesium can suppress platelet aggregability

18 Results Creation of Dataset 1 Keywords pairs e.g. stress + migraine etc. against PubMed return PubMed abstracts that are annotated (by NLM) with both terms 8 pairs of terms in this scenario result in 8 subsets of PubMed Semantic Metadata Represented in RDF With complex entities and relationships connecting them Pointers to original document and sentence Size ~2MB RDF for Migraine Magnesium subset of PubMed

19 Evaluating the Result of Extraction Ideal method to evaluate the Extraction method Domain experts read a set of abstract given a set of relationship names and entities to look for In addition to this give them the extracted triples and entities For every abstract the expert judges counts the correct, incorrect and missed triples Measure precision and recall

20 Evaluating the Result of Extraction In the absence of a domain expert we focus of getting a feel for the utility of the extracted data We know the association manually discovered between Migraine and Magnesium We locate paths of various lengths between them and manually inspect these paths If the paths are indicative of the manually discovered associations the extracted data is useful

21 Paths between Migraine and Magnesium Paths are considered interesting if they have one or more named relationship Other than haspart or hasmodifiers in them

22 An example of such a path stimulated caused_by migraine (D008881) haspart platelet (D001792) haspart collagen (D003094) haspart magnesium (D008274) stimulated me_2286 _13%_and_17%_adp_and_collagen_induced_platelet_aggregation me_3142 by_a_primary_abnormality_of_platelet_behavior

23 Results Dataset 2 Neoplasm (C04) For subtree of MeSH rooted at Neoplasms all topics under this subtree are used as query terms against PubMed The resulting dataset contains ~500,000 PubMed abstracts The extraction process run on this data returns ~150MB Processing the tagged and parsed sentences for Dataset 2 (Neoplasm) to generate RDF took approx. 5 minutes Stats 211 different named relationships found 500,000 instance-property-instance statements 260,000 instance-property-literal statements Currently setting up to extract RDF from all of PubMed

24 Outline Motivation Problem Description & Approach Results Future Work

25 Future Extensions to the Extraction process Short-term goals (1 month) MeSH qualifiers (blood pressure, contraindications) Curate and release Migraine-Magnesium RDF Long-Term goals More complex structures Conjunctions X causes Y to inhibit Z Rule-action language to test new extraction rules Finding new terms to enrich existing vocabularies Perhaps ontology enrichment

26 The projected future of research in Biology From Hypothesis driven wet lab experiments To Data-driven reduction/pruning of hypothesis space leading to new insight and possibly discovery What challenges does this transition bring?

27 Use of Generated Semantic Metadata Semantic Browsing of PubMed based on named relationships between MeSH terms Path/hypothesis based document retrieval Knowledge discovery from literature Coprus-based complex relationship discovery and ranking Corpus-based relevant connection subgraph discovery

28 Support such retrieval and discovery operations across multiple data sources Extract Semantic Metadata about entities in all of these databases that might occur in PubMed text Resulting metadata will contain relationships between genes (OMIM), diseases (MeSH), nucleotide anomalies (SNP) hypothesis validation and knowledge discovery in biology.

29 THANK YOU!