Abstract

Abstract We experiment with text classification of threads from Linux web user forums, in the context of improving information access to the problems and solutions described in the threads. We specifically focus on classifying threads according to: (1) them describing a specific problem vs. containing a more general discussion; (2) the completeness of the initial post in the thread; and (3) whether problem(s) in the initial post are resolved in the thread or not. We approach these tasks in both classification and regression frameworks using a range of machine learners and evaluation metrics.

Citations