A LDA Based Model for Topic Evolution: Evidence from Information Science Journals

DOI

10.2991/msota-16.2016.12How to use a DOI?

Keywords

evolution of topic; topic intensity; topic content; Latent Dirichlet Allocation

Abstract

Mining evolution of topic from papers plays an important role in learning about the trend and monitoring the hot topic of research. The paper proposes a model based on Latent Dirichlet Allocation (LDA) for the purpose of mining evolution of topic. Firstly, we deal with the collection of all the papers using LDA to find out topics and their key words, and get probability distribution of document - topic on different time windows so that we can figure out the trend of topic intensity. Secondly, we apply LDA in papers on every single time window to get probability distribution of topic - word, through which we can compute similarity of topics from different time windows, and the words probability of similar topics can help us figure out trend of topic content.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Maoran Zhu
AU  - Xiaopeng Zhang
AU  - Hongwei Wang
PY  - 2016/12
DA  - 2016/12
TI  - A LDA Based Model for Topic Evolution: Evidence from Information Science Journals
BT  - Proceedings of 2016 International Conference on Modeling, Simulation and Optimization Technologies and Applications (MSOTA2016)
PB  - Atlantis Press
SP  - 49
EP  - 54
SN  - 2352-538X
UR  - https://doi.org/10.2991/msota-16.2016.12
DO  - 10.2991/msota-16.2016.12
ID  - Zhu2016/12
ER  -