Computer-Assisted Creation of Boolean Search Rules for Text Classification in the Legal Domain
Hannes Westermann, Jaromir Savelka, Vern R. Walker, Kevin D. Ashley et Karim Benyekhlef, Computer-Assisted Creation of Boolean Search Rules for Text Classification in the Legal Domain, 2019, Frontiers in Artificial Intelligence and Applications, Volume 322, 123-132.
In this paper, we present a method of building strong, explainable classiﬁers in the form of Boolean search rules. We developed an interactive environment called CASE (Computer Assisted Semantic Exploration) which exploits word co-occurrence to guide human annotators in selection of relevant search terms. The system seamlessly facilitates iterative evaluation and improvement of the classiﬁcation rules. The process enables the human annotators to leverage the beneﬁts of statistical information while incorporating their expert intuition into the creation of such rules. We evaluate classiﬁers created with our CASE system on 4 datasets, and compare the results to machine learning methods, including SKOPE rules, Random forest, Support Vector Machine, and fastText classiﬁers. The results drive the discussion on trade-offs between superior compactness, simplicity, and intuitiveness of the Boolean search rules versus the better performance of state-of-the-art machine learning models for text classiﬁcation.
This content has been updated on 01/08/2020 at 14 h 00 min.