Exploring supervised and unsupervised methods to detect topics in biomedical text

Abstract

Background: Topic detection is a task that automatically identifies topics (e.g., "biochemistry" and "protein structure") in scientific articles based on information content. Topic detection will benefit many other natural language processing tasks including information retrieval, text summarization and question answering; and is a necessary step towards the building of an information system that provides an efficient way for biologists to seek information from an ocean of literature.

Results: We have explored the methods of Topic Spotting, a task of text categorization that applies the supervised machine-learning technique naïve Bayes to assign automatically a document into one or more predefined topics; and Topic Clustering, which apply unsupervised hierarchical clustering algorithms to aggregate documents into clusters such that each cluster represents a topic. We have applied our methods to detect topics of more than fifteen thousand of articles that represent over sixteen thousand entries in the Online Mendelian Inheritance in Man (OMIM) database. We have explored bag of words as the features. Additionally, we have explored semantic features; namely, the Medical Subject Headings (MeSH) that are assigned to the MEDLINE records, and the Unified Medical Language System (UMLS) semantic types that correspond to the MeSH terms, in addition to bag of words, to facilitate the tasks of topic detection. Our results indicate that incorporating the MeSH terms and the UMLS semantic types as additional features enhances the performance of topic detection and the naïve Bayes has the highest accuracy, 66.4%, for predicting the topic of an OMIM article as one of the total twenty-five topics.

Conclusion: Our results indicate that the supervised topic spotting methods outperformed the unsupervised topic clustering; on the other hand, the unsupervised topic clustering methods have the advantages of being robust and applicable in real world settings.

Figure 1

The number of the OMIM entries as a function of the number of topics.

Figure 2

Topic clustering in OMIM. The cost of detection of Single-pass (A) and Group-wise-average (B) with different features; namely, semantic types only (S); combined semantic types and MeSH terms (S+M); semantic types with MeSH terms and title (S+M+T); all four features (S+M+T+A); the MeSH terms alone (M); and combined title and abstract (T+A). (C) Comparison of the cost of detection between single-pass and group-wise-average with MeSH terms alone as features and similarity threshold τ = 0.5.

Figure 3

Topic Clustering as a function of number of topics (similarity of threshold τ = 0.5.).

Figure 4

Topic Clustering of articles cited in biological review articles. (A) The cost of detection of Group C with different features; namely, semantic types only (S); combined semantic types and MeSH terms (S+M); semantic types with MeSH terms and title (S+M+T); all four features (S+M+T+A); the MeSH terms alone (M); and combined title and abstract (T+A). (B) The cost of detection of three groups (A, B and C) with the semantic types as the feature.