Application Of Web Log Mining Technology In Fishery Information

With the rapid development of networking and information technology, the numberof networking information increases fiercely, which result in netizen soaking in theocean of information. Netizen get the information what they wanted so difficultly frommass information that they feel confused. In this background, the Web Mining, as asolution to above problem, is put forward. Web Mining is a new cross-discipline, basedon data mining, and involves many fields, such as Statistics, Machine Learning,Artificial Intelligence and so on, and extracts potential valuable user access pattern fromlots of isomeric and unstructured Web Information by integrating various knowledgeand technology. After analysis and study of these patterns, it brings these patterns inproduction application, so as to resolve some difficulty or improve statement. Inpractical applications, site administrators can adjust site structure according the user’sbehavior patterns, and information service platform based on user behavior patterns canprovide users with the information resources to meet user needs. The main work in thepaper includes:(1) Have summarized the background, significance and main content of researchsubject. With the process of development of fishery information, fishery science datagradually increases. In order to manage these data more scientific and rational, andbuild convenient and effective fishery information service platform, the Web log miningtechnology is brought into the field of fishery information, and fishery informationrecommendation system and intelligent RSS reader are designed and developed.(2) Have systematically stated a series of related conception, from Data Mining,Web Mining to Web Log Mining. A detailed analysis of Web log mining research statusat home and abroad is made, and the Web log mining technology application in the fieldof agriculture and fisheries is summarized, analyzed and forecasted. The whole processand related technologies of Web Log Mining is simply introduced then.(3) Algorithm improvement, including research of mining algorithm andapplication research of Web Log Ming in fishery information. According the procedureof Web Log Mining, the paper researched and analyzed User Identification Algorithmand proposed IASR algorithm at the step of preprocess data, which is proved excellentby contrastive experiment. At the step of patterns discovering, the classic algorithmApriori is developed. After that, the Apriori is more efficient when it deal with massdata, which is also proved by contrastive experiment.(4) Design and development of personalized service system. Based on fisheryscience data share platform, the frequent visiting pattern is mined from mass log data via developed web log mining algorithm, and use recommend technology of associationrules and user tracking technology to realize recommendation on the platform so as tohelp user get what they really need. In addition, Aim at the traditional RSS readersubscribing burdensome information, intelligent RSS reader is be designed anddeveloped that can automatically filter redundant information. The intelligent readeruses vector space model represent of text and the user’s characteristics interest, and usesChinese word segmentation extract the user’s interest features from the user’s history,according to the user feature vector reader automatically filters subscription informationthat is not related to user, reducing the user reader pressure, improving the userexperience. At last, fishery information system and intelligent RSS reader are integratedon fishery science data platform. It improves fishery science data platform quality ofservice and user experience, effectively promotes the process of fishery information.