Integrating Genetic Algorithms with Conditional Random Fields to Enhance Question Informer Prediction

Question informers play an important role in enhancing question classification for factual question answering. Previous works have used conditional random fields (CRFs) to identify question informer spans. However, in CRF-based models, the selection of a feature subset is a key issue in improving the accuracy of question informer prediction. In this paper, we propose a hybrid approach that integrates genetic algorithms (GAs) with CRF to optimize feature subset selection in CRF-based question informer prediction models. The experimental results show that the proposed hybrid GA-CRF model improves the accuracy of question informer prediction of traditional CRF models