CCR

Previous Issue

Asian Journal of Information and Communications 12(1)
Asian Journal of Information and Communications 12(1)

Call for Paper

Add to Bookmark

Editorial Office

Asian Journal of Information and Communications

A novel method of fuzzy topic modeling based on transformer processing Ching-Hsun Tseng et al.

  • Page:0
  • 2020-10-07
Topic modeling is admittedly a convenient way to monitor markets trend. Conventionally,
Latent Dirichlet Allocation (LDA) is considered a must-do model to gain this type of
information. By the given merit of deducing keyword with token’s conditional probability
in LDA, we can know the most possible or essential topic. However, the results are not
intuitive because the given topics cannot wholly fit human knowledge. LDA offers the first
possible relevant keywords, which also bring out another problem of whether the
connection is reliable based on the statistic possibility. It is also hard to decide the topic
number manually in advance. As the booming trend of using fuzzy membership to cluster
and using transformers to embed words, this work presents the fuzzy topic modeling based
on soft clustering and document embedding form state-of-the-art transformer-based model.
In our practical application in a press release monitoring, the fuzzy topic modeling gives a
more natural result than the traditional output from LDA.

Keywords: fuzzy, soft clustering, transformer
  PDF Download
  • Facebook
  • Share onTwitter