Topic segmentation of message hierarchies for indexing and navigation support

Abstract

Message hierarchies in web discussion boards grow with new postings. Threads of messages evolve as new postings focus within or diverge from the original themes of the threads. Thus, just by investigating the subject headings or contents of earlier postings in a message thread, one may not be able to guess the contents of the later postings. The resulting navigation problem is further compounded for blind users who need the help of a screen reader program that can provide only a linear representation of the content. We see that, in order to overcome the navigation obstacle for blind as well as sighted users, it is essential to develop techniques that help identify how the content of a discussion board grows through generalizations and specializations of topics. This knowledge can be used in segmenting the content in coherent units and guiding the users through segments relevant to their navigational goals. Our experimental results showed that the segmentation algorithm described in this paper provides up to 80 − 85% success rate in labeling messages. The algorithm is being deployed in a software system to reduce the navigational load of blind students in accessing web-based electronic course materials; however, we note that the techniques are equally applicable for developing web indexing and summarization tools for users with sight.

Online High-Quality Topic Detection for Bulletin Board Systems

“…According to the differences between BBS and news web pages, a few research results in BBS environment are published. J. Kim et al [12] proposed a method to split the topics in a single thread by organizing the posts in a thread as a chain and calculating the similarity between consecutive posts in the chain. However, posts from different threads cannot be organized as a chain, so this method is not appropriate for detecting topics from a set of threads.…”

Organization and Tagging of Blog and News Entries Based on Content Reuse

“…In our prior work, we presented content-driven techniques for measuring specialization/generalization degrees between entries in hierarchical structures, such as IS-A concept hierarchies [9] and message hierarchies, such as discussion board [8]. In this paper, we highlight that, a particular advantage of these content-driven techniques is that they do not assume any advance knowledge about the structure and, thus, can be used when only pairwise content (and reuse such as quotations) are available.…”

“…where for any x, | x| denotes the distance from the origin O, and b j C is the common parts of b i and b j (See [9] and [8] for more detail).…”

Topic Detection and Tracking for Threaded Discussion Communities

“…Kim et al [6] proposed a method to segment topics in a single discussion thread (hierarchy). Topic segmentation is also developed for online chat data, such as in [2] and [13].…”