Word Class-based Language Models for Sentence Retrieval in Question Answering Systems
Speaker: Saeedeh Momtazi
Abstract:
In this research we are discussing the effect of term clustering approach for improving the performance of sentence retrieval in question answering systems. As the search in question answering is conducted over smaller segments of data than in a document retrieval task, the problems of data sparsity and exact matching become more critical. In this task we use language model-based techniques to overcome such problems and improve the sentence retrieval performance. Our proposed methods include building class-based models by term clustering, employing higher order n-grams to the new class-based model, and interpolate the class-based and the word-based models. In our experimental results, the investigated method enhanced the mean average precision of sentence retrieval.
