مقالات
Automatic Hypertext Construction in Persian Texts Using Self- Organizing Map Neural Network
نویسندگان
Mahdieh HajiMohammadHosseini, Behrouz Minaei-Bidgoli
چکیده
With the availability of electronic texts, users are encouraged to study them. Therefore users may encounter during their study with different information needs and want more information or related information about a particular word or phrase within that document. If so, it is necessary to search the entire corpus of texts and then they are faced with problems related to the search. Using hypertext is a fast method for retrieving information. Manually converting large amounts of documents into hypertext is time consuming and sometimes impossible. The purpose of this paper is to implement an automated way to convert texts into hypertext. This is the first activity and implementation in Persian documents. In this approach, two types of links are made using Selforganizing Map neural network, two labeling processes and analyzing them. In this study, in addition to single words links, two-word phrases links are produced too. Some of links sources in generated links were in title of the destination document that shows high correlation between source and destination; but about other sources, we specified most related paragraph in the destination document. The average precision rate of the two types of links for single words and phrases was calculated 0.71
کلیدواژهها
Automatic hypertext construction, Information Retrieval, Neural networks, Text mining