Freetext Medical Document Retrieval via Phrasebased Vector Space Model

1
Free-text Medical Document Retrieval via
Phrase-based Vector Space Model

• Wenlei Mao, MS and Wesley W. Chu, PhD
• wenlei_at_cs.ucla.edu and wwc_at_cs.ucla.edu
• Computer Science Department
• University of California, Los Angeles

2
Outline

• Vector space model (VSM) in document retrieval
• Stem-based VSM
• Concept-based VSM
• Conceptual similarity
• Phrase-based VSM
• Retrieval effectiveness comparison
• Conclusion

3
Document Retrieval

• Find free-text documents to answer queries like,
• Hyperthermia, leukocytosis, increased
  intracranial pressure, and central
  herniation.Cerebral edema secondary to
  infection, diagnosis and treatment.

4
Vector Space Model (VSM)
5
Stem-based VSM

• Morphological variants bear similar content
• E.g., edema and edemas
• Use stemmer to extract stems
• Lovins stemmer and Porter stemmer

• Baseline of comparison

6
Shortcomings of Stem-based VSM

• Inability to capture multi-word concepts
• Increased intracranial pressure
• Inability to utilize the relations between
  concepts
• Synonyms hyperthermia and fever
• IS-A relation hyperthermia and body
  temperature elevation

7
Concept-based VSM

• Uses concepts in knowledge base (KB) as terms
• KB Metathesaurus in UMLS

• Captures multi-word concepts
• Captures synonyms

8
Shortcomings of Concept-based VSM

• Concepts may be related
• E.g. hyperthermia and body temperature
  elevation are not identical but related concepts
• Need to quantify conceptual relations
• Knowledge bases are often incomplete, which
  reduces the retrieval effectiveness

9
Conceptual Similarity Evaluation
10
Deriving Conceptual Similarity From Hypernym
Hierarchy
11
Shortcomings of Concept-based VSM

• Concepts may be related
• The conceptual similarity measure, s(ci,cj),
  quantifies relations between concepts.
• Knowledge bases are often incomplete, which
  reduces the retrieval effectiveness.

12
Incompleteness of the Knowledge Bases

• In general, concept-based VSM cannot outperform
  stem-based VSM

13
Phrase-based Indexing Examples
14
Evaluate Phrase-based Document Similarity
15
To Compare Retrieval Effectiveness

• The test set OHSUMED
• 106 queries, 14K documents
• Expert relevance judgment R or N
• Retrieval effectiveness
• Recall the percentage of relevant documents
  retrieved so far
• Precision the percentage of retrieved documents
  that are relevant

16
Retrieval Effectiveness Comparison (Corpus
OHSUMED, KB UMLS)
16100 queries vs. 5 50 queries
17
Stem and Concept Similarity Contribution Weights
18
Sensitivity of Retrieval Effectiveness to fs and
fc
19
Computation Complexity Using Phrase-based VSM

• Data reorganization
• Build separate indexes on stems and concepts
• Keep a list of related concepts cjs and
  conceptual similarity s(ci,cj) with ci.
• Time complexities of document similarity
  calculation, same order of magnitude
• Stem-based VSM
• Phrase-based VSM

20
Conclusion

• A new document indexing paradigm based on phrases
  is proposed
• Use phrases (concept and its word stems) as terms
• Document similarity is derived from both the stem
  and the concept contributions
• Conceptual similarity quantifies the concept
  relations and improves retrieval effectiveness
• Stems remedy the incomplete coverage of the
  knowledge base (missing concepts and missing
  links between related concepts)
• Experimental results reveal a significant
  retrieval effectiveness improvement of the
  phrase-based VSM over the stem-based VSM

21
Acknowledgement
This research is supported in part by NIC/NIH
Grant4442511-33780
22
Model Comparison