Using Clustering and Modiﬁed Classiﬁcation algorithm without a learning corpus for automatic text summarization
In this paper we describe a modiﬁed classiﬁcation method destined for extractive summarization purpose. The classiﬁcation in this method doesn’t need a learning corpus; it uses the input text to do that. First, we cluster the document sentences to exploit the diversity of topics, then we use a learning algorithm (here we used Naive Bayes) on each cluster considering it as a class. After obtaining the classiﬁcation model, we calculate the score of a sentence in each class, using a scoring model derived from classiﬁcation algorithm. These scores are used, then, to reorder the sentences and extract the ﬁrst ones as the output summary. We conducted some experiments using a corpus of scientiﬁc papers, and comparing our system to another system which is UNIS system. Also, we experiment the impact of clustering threshold tuning, on the resulted summary, as well as the impact of adding more features to the classiﬁer. We found that this method is interesting, and gives good performance, and the addition of new features (which is simple using this method) can improve summary’s accuracy.
NLP, IR, Automatic text summarization, Clustering