Conference Publication Details
Mandatory Fields
Dunne J.;Malone D.;Penrose A.
24th International Conference on Collaboration and Technology (CRIWG 2018) Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Bundles: A framework to optimise topic analysis in real-time chat discourse
0 ()
Optional Fields
Chat segmentation Regression Topic modelling
© Springer Nature Switzerland AG 2018. Collaborative chat tools and large text corpora are ubiquitous in today’s world of real-time communication. As micro teams and start-ups adopt such tools, there is a need to understand the meaning (even at a high level) of chat conversations within collaborative teams. In this study, we propose a technique to segment chat conversations to increase the number of words available (19% on average) for text mining purposes. Using an open source dataset, we answer the question of whether having more words available for text mining can produce more useful information to the end user. Our technique can help micro-teams and start-ups with limited resources to efficiently model their conversations to afford a higher degree of readability and comprehension.
Grant Details