Related Experiment Video
Updated: Jun 24, 2026
05:56
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
Combining hidden Markov models and latent semantic analysis for topic segmentation and labeling: method and clinical
Filip Ginter1, Hanna Suominen, Sampo Pyysalo
1Department of Information Technology, University of Turku, Turku, Finland. filip.ginter@utu.fi
International Journal of Medical Informatics
|March 31, 2009
Summary
This study introduces an unsupervised topic segmentation method using hidden Markov models and latent semantic analysis. It effectively identifies short text segments without needing annotated data, outperforming baseline methods.
Area of Science:
- Information Science
- Natural Language Processing
- Computational Linguistics
Background:
- Topic segmentation and labeling are crucial for fine-grained information retrieval.
- Existing methods often require extensive annotated data and struggle with short text segments.
Purpose of the Study:
- To develop an unsupervised method for topic segmentation and labeling.
- To address limitations of existing methods regarding data annotation and segment length.
Main Methods:
- A novel unsupervised approach combining Hidden Markov Models (HMM) and Latent Semantic Analysis (LSA).
- This method allows for free definition of topics without manual annotation.
- Capable of identifying topics within short text segments.
Main Results:
- The method was evaluated on intensive care nursing narratives.
- It significantly outperformed a keyword-based heuristic baseline.
- Achieved performance comparable to supervised methods trained on substantial annotated data.
Conclusions:
- The proposed unsupervised method offers an effective solution for topic segmentation, particularly for short texts.
- It reduces the need for manual annotation, making it more adaptable and efficient.
- Demonstrates potential for improved information retrieval in specialized domains.