What question did this study set out to answer?

The aim is to develop a comprehensive index to determine the optimal number of topics in an LDA topic model.

October 3, 2021EntropyOpen Access

Selection of the Optimal Number of Topics for LDA Topic Model—Taking Patent Policy Analysis as an Example

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates an advanced method for topic selection in patent policies, suggesting improved classification accuracy.

The aim is to develop a comprehensive index to determine the optimal number of topics in an LDA topic model.
Constructed a judgment index based on perplexity, isolation, stability, and coincidence.
Compared the proposed method with existing topic selection methods using three general datasets.
Analyzed patent policies from various provinces in China to evaluate classification effectiveness.
The proposed method shows improved predictive ability in topic selection.
Exhibits high isolation between topics with no duplicates.
Classified patent policies effectively using the optimal number of topics.

Gan et al. (2021) studied this question.

View Full Paper

Ask AI

Bookmark