ISSN:2229-6093 G. N. Shinde,Inamdar S.A, Int. J. Comp. Tech. Appl., Vol 2 (2), 280-283
Web Data Mining Using An Intelligent Information System Design G. N. Shinde Indira Gandhi College, CIDCO, Nanded-431603, Maharashtra, INDIA
[email protected]
Abstract— To use the large amounts of information efficiently on the Web to make the information processing intelligent, personalized and automatic is the most important applications of the current data mining technology. Model Driven Architecture (MDA) which is used for code generation has many benefits over traditional software development methods. An intelligent
mining system of information is built with combining the data mining. In this paper, concept of Web data mining is introduced where the role of MDA is defined. MDA using J2EE (Java to Enterprise Edition) to describe behavior of agents are used in this proposed architecture. JADE (Java Agent Development Environment) Framework provides a standard for developing MAS (multi-agent systems). Also Agent’s Modeling Language (AUML) is being defined to effective implementation of defining agent roles. Keywords-Multi-agent systems, Modeling Language (AUML)
J2EE,
JADE,
Agent’s
INTRODUCTION The WWW serves as a huge, wide, distributed, global information service center for news, dvertisements, consumer information, financial management, education, government, ecommerce, and m y other services[14']. With the rapid increasing of information in the WWW, the Web Mining has gradually become more and more important in Data Mining. Web Mining can be classified into three domains: Web Structure Mining, Web Content Mining and Web Usage Mining. There are generally three tasks in Web Usage Mining: preprocessing, knowledge discovery and pattern analysis. I.
Inamdar S.A. School of Computational Science Swami Ramanand Teerth Marathwada University, Nanded, Maharashtra, , INDIA
[email protected]
WEB MINING TECHNOLOGY Data mining is extracting potential, unknown, useful information, patterns and trends from abundant, incomplete, noise and random data. The web mining technology is an important branch of data mining. The Web includes huge t amount of data, using the data mining technology on the Web, namely the Web mining rechnology, becomes the most important Iesearch along with the rapid development of tnternet [2]. The Web grows and evolves faster shan we would like and expect, imposing scalability and relevance problems to Web earch engines. II.
A.
Web-based data mining concept
Web usage mining is to mine Web log to discover user accessing patterns of Web pages. Through analyzing and exploring regularities in Web log records engines. The Web can be seen as a structure containing information about hyperlinks, Web usage information and Web contents in itself. Web site usage data, which s contain records of how user has visited a Web ite [9, 11]. Agent technology and multi-agent system Agents are used to perform some action s activity on behalf of a user of a computer or ystem. Agent refers to the entities which run in dynamic environment and have higher selfgovernment capacity. Agent software is a type i computer program which simulates human of ntelligence behaviour and provides the B.
280
ISSN:2229-6093 G. N. Shinde,Inamdar S.A, Int. J. Comp. Tech. Appl., Vol 2 (2), 280-283
corresponding services. Agent has characteristics of autonomy, reaction, initiative and sociality. Multi-agent system is composed by a number of Agents and has certain organization structure, as an effective method of solving complex systems. Web search engines are one of the most popular services to help users find useful information on the Web. C.
profiles to organize and interpret the discovered information. •
Information filtering/categorisation agents : These agents use a number of techniques and characteristics of hypertext documents to automatically retrieve, filter and categorize web documents according to some predefined criteria or user interaction.
•
Personalized web agents : This type of web agent learns user preferences and then automatically discovers web documents and resources based on the created user profile.
Web data mining categories
Web mining research can be classified into three categories: Web content mining (WCM), Web structure mining (WSM), and Web usagemining (WUM). Web content mining refers to the discovery of useful information from web contents, including text, image, audio, video, etc. Web content mining describes the discovery of useful information from the content of web pages or Web documents[4].
III.
WEB-BASED DATA MINING PROCESS
Web mining is to explore interesting information and potential patterns in the contents of Web page, by using techniques of data mining, which can help people extract knowledge, improve Web sites design.
Figure 2. Web mining functions A.
Figure 1. Taxonomy of Web mining •
Intelligent search agents : This class of agent deployed on the web automatically searches for information and it deems relevant to a particular search query using domain characteristics and user
Data acquisition
The data acquisition is composed by three relatively independent processes which are data search, data selectionand data collection. B. Data preprocessing Data preprocessing is the transform process of the result of resource discovery.
281
ISSN:2229-6093 G. N. Shinde,Inamdar S.A, Int. J. Comp. Tech. Appl., Vol 2 (2), 280-283
C. Data mining Data mining refer to find the hidden, potentially useful information from a large amount of data. The web mining technology is an important branch of data mining. D.
Analysis and evaluation
Analysis and evaluation module is to analyze the credibility and effectiveness of the knowledge mode the data mining obtained. E.
Knowledge formulation
Knowledge expression module refers to the knowledge modes mined from the Web data by using data mining tools. IV. Intelligence mining system design The model is designed to cover the limitations of traditional search engines and its purpose is to design the information service intelligence system. The basic idea is: the first step, from the Web pages of particular Website analyzing and extracting some necessary parameters which are title or distinction sign of text beginning and ending,link address[11].
Figure 3. Information intelligence mining system structure
A. Approach To capture the behavior of agents the concept of agent roles is used . Roles of agents are defined by their behavior. Implementastion of proposed design Model Driven Architecture (MDA) which is used for code generation has many benefits over traditional software development methods. MDA using J2EE (Java to Enterprise Edition) to describe behavior of agents are used in this proposed architecture. JADE (Java Agent Development Environment) Framework provides a standard for developing MAS (multiagent systems) [16].Also for Agent’s Modeling B.
282
ISSN:2229-6093 G. N. Shinde,Inamdar S.A, Int. J. Comp. Tech. Appl., Vol 2 (2), 280-283
Language (AUML) is being defined to effective implementation of defining agent roles. V. CONCLUSIONS MDA is a promising approach for software development. MDA using J2EE (Java to Enterprise Edition) is used to describe behavior of agents. JADE (Java Agent Development Environment) Framework provides a standard for developing MAS (multi-agent systems). Web Usage Mining, as well as Web Mining, is a new research field, which has a long way to go. For the Web-based data warehouse and data mining technology, the development of the Internet provides a broad application scope. With the rapid development of Internet, communications technology, the research of Web based data mining will be further in-depth and Web site design and so on. Also for Agent’s Modeling Language (AUML) is being defined to effective implementation of defining agent roles . REFERENCES [1] Li Zhan, Liu Zhijing, , ‘ Web Mining Based On Multi-Agents ’, COMPUTER SOCIETY,IEEE(2003) [2] Margaret H. Dunham and Sridhar, Data Mining, Introduction and Advanced Topics, (Prenticce Hall Publication), ISBN 81-7758785-4, chap nos.1,7, pp.3,4,195-218. [3] YAN LI , XIN-ZHONG CHEN , BING-RU YANG, ‘RESEARCH ON WEB MININGBASED INTELLIGENT SEARCH ENGINE’, proceedings of first international conference on machine learning and cybernetics,Biejing, IEEE(2002). [4] WangBin, LiuZhijing, , ‘ Web Mining Research’ , International Conference On Computational Intelligence and Multimedia Applications, IEEE (2003). [5] Hiroyuki Kawano, ‘Web Archiving Strategies by using Web Mining Techniques’, IEEE (2003).
[6] Sung Ho Ha, Sung Min Bae, Sang Chan Park, ‘WEB MINING FOR DISTANCE EDUCATION’, ICMIT, IEEE (2000). [7] Sanjay Kumar Madria, white paper, ‘ Web Mining : A Bird’s Eye View ’(2008). [8] Arun K Pujari , Data Mining Techniques ,Universities Press (India) Limited, ISBN 817371-380-4 [9] Wang Jicheng, Huang Yuan, Wu Gangshan and Zhang Fuyan, ‘Web Mining: Knowledge Discovery on the Web’ , IEEE (1999). [10] Jakub Snopek, Ivan Jelínek, ‘Web Access Predictive Models’, International Conference on Computer Systems and Technologies – CompSysTech (2005). [11] Xinlin Zhang, Xiangdong Yin, ‘ Design of an Information Intelligent System based on Web Data Mining’ ,International Conference on Computer Science and Information Technology 2008. [12] Feng Zhang, HuiI-You Chang, ‘RESEARCH AND DEVELOPMENT IN WEB USAGE MINING SYSTEM--KEY ISSUES AND PROPOSED SOLUTIONS: A SURVEY’ International Conference on Machine Learning and Cybernetics, Beijing, 4-5 November 2002 [13] Wu Gangshan, Huang Yuan, Shian-Shyong Tseng, Zhang Fuyan, ‘A knowledge sharing and collaboration system model based on Internet’, 1999, IEEE. [14] Lizhen Liu, Junjie Chen, Hantao Song, ‘The Research of Web Mining’, Proceedings of the 4th World Congress on Intelligent Control and Automation June 10-14, 2002, Shanghai, P.R.China ,IEEE. [15] Shakirah Mohd Taib, Soon-Ja Yeom, Byeong-Ho Kang, ‘Elimination of Redundant Information for Web Data Mining’, Proceedings of the International Conference on Information Technology: Coding and Computing (ITCC’05) IEEE. [16] James Huamonte, Kevin Smith, ‘The Use of Roles to Model Agent Behaviors for Model Driven Architecture’, 2005 IEEE.
283