Summarization of changes in dynamic text collections using Latent Dirichlet Allocation model
作者：
Highlights：
•
• The summarization of changes addresses a new challenge – the automatic summarization of changes in dynamic text collections.
• Four different approaches are proposed for the summarization of changes.
• A system based on Latent Dirichlet Allocation model is used to find the hidden topic structures of changes.
• The approach based on LDA model outperforms all the other approaches.
• The differences in ROUGE scores for LDA based approach is statistically significant at 99% over baseline.
摘要
•The summarization of changes addresses a new challenge – the automatic summarization of changes in dynamic text collections.•Four different approaches are proposed for the summarization of changes.•A system based on Latent Dirichlet Allocation model is used to find the hidden topic structures of changes.•The approach based on LDA model outperforms all the other approaches.•The differences in ROUGE scores for LDA based approach is statistically significant at 99% over baseline.
论文关键词：Changes summarization,Temporal term weighting,Sentence ranking,Latent Dirichlet Allocation,Wikipedia
论文评审过程：Received 22 September 2014, Revised 26 May 2015, Accepted 1 June 2015, Available online 30 June 2015, Version of Record 27 October 2015.
论文官网地址：https://doi.org/10.1016/j.ipm.2015.06.002
