Study On Information Filtering Based On Artificial Immune System

The Information Retrieval becomes a necessary way for people normal living along with the wide use of network. At the same time, to filter nocuous or futile information emerged as side effects of the popularity of information processing over network. So the fast and veracious Retrieval for kinds of information has become an important issue. Artificial Immune System which is one of the important fields of intellective information processing means to filter non-self information is receiving increasing attention.Research works focus on the excellent properties of Artificial Immune System and its use in information filtering, which include: (1) Analysis of the properties of Artificial Immune System; (2) the information filtering methods based on Artificial Immune Algorithm; (3) An idiographic information processing method to Email and Web. The main research fruits are as follows:1. Summarize and analyze the status of the use of Artificial Immune System in information filtering at present and give some examples of the use of information filtering, then point out some problems of exiting information retrieval.2. Analyze the methods of information processing in existence, especially Natural Language Understanding for Chinese and automatic segmentation of Chinese words. Summarize the problem and elicitation of information processing based intelligent technology.3. The function of the immune system is to protect the living body against invaders through the use of defensive mechanisms. This thesis looks at application of the artificial immune system model to protect email users effectively from spam and presents an immune algorithm called EIS that is capable of continuously classifying email as spam and non-spam. Comparisons are drawn with other classifier methods and it is shown that the proposed system performs as well as the other classifier system, especially on recall, and has a great potential for information classifying.