【作者单位】1 Center for Data Engineering, International Institute of Information Technology, Gachibowli, Hyderabad 500 032, India.2 Center for Data Engineering, International Institute of Information Technology, Gachibowli, Hyderabad 500 032, India.3 Center for Data Engineering, International Institute of Information Technology, Gachibowli, Hyderabad 500 032, India
【出 处】International Journal of Computational Science and Engineering
【摘 要】 In the literature, research efforts are going on to extract interesting information from text documents to improve the performance of information–based services. Interesting information is extracted after identifying features from each document. In t... 更多 >> In the literature, research efforts are going on to extract interesting information from text documents to improve the performance of information–based services. Interesting information is extracted after identifying features from each document. In this paper, we have proposed the notion of 'special feature' which is a new kind of knowledge that can be used to improve the performance of information–based services. A feature is a special feature if only very few documents in the dataset possess it. Given a text document dataset, we have proposed a methodology to extract special features. By using the notion of special features, we have also proposed frameworks to improve the performance of product selection in the e–commerce environment and the process of resume selection. The experiment results on real datasets show that it is possible to improve the efficiency of the applications with the proposed approach. << 收起
