American Journal of Data Mining and Knowledge Discovery 2019; 4(1): 19-23 http://www.sciencepublishinggroup.com/j/ajdmkd doi: 10.11648/j.ajdmkd.20190401.14 ISSN: 2578-7810 (Print); ISSN: 2578-7837 (Online) Extermination of Obsolete Relationship Through KTMIN-JAK-MAXAM Algorithm in Confusion Mining Kittappa Thiagarajan 1, * , Jeyavel Kavitha 2 , Karunakaran Sarukesi 3 , Avudaiappan Maheshwari 4 1 Academic Research and Development, Jeppiaar Engineering College, Chennai, India 2 Department of Mathematics, R&D Centre, Bharathiar University, Coimbatore, India 3 KCG College Engineering and Technology, Chennai, India 4 Department of Computer Science and Engineering, Jeppiaar Engineering College, Chennai, India Email address: * Corresponding author To cite this article: Kittappa Thiagarajan, Jeyavel Kavitha, Karunakaran Sarukesi, Avudaiappan Maheshwari. Extermination of Obsolete Relationship Through KTMIN-JAK-MAXAM Algorithm in Confusion Mining. American Journal of Data Mining and Knowledge Discovery. Vol. 4, No. 1, 2019, pp. 19-23. doi: 10.11648/j.ajdmkd.20190401.14 Received: February 25, 2019; Accepted: April 10, 2019; Published: May 11, 2019 Abstract: At the present time web contains many indistinguishable documents. Much effort made towards in investigates mechanism with identical detection algorithms, still the retrieved web documents with outmodedlink. In this proposed system, we are successfully identifying and minimize the redundant information and like link in web documents. We introduce the correct graph theory based KTMIN-JAK-MAXAM algorithm filters out the redundant link. From the proposed system, we have relevant information with more accuracy. Using this KTMIN-JAK-MAXAM algorithm accessing of web pages with reduced time and space complication. Keywords: Data Mining, Degree, Maximum Degree, Minimum Degree, Link, Path 1. Introduction The data and information available on web is exponentially improving, duplication of web content also increase simultaneously [1, 2]. Retrieving relevant information from web without redundancy is more challenge task nowadays where in web mining communities [3, 4]. Web content mining is the way toward extracting the applicable information, data and learning from World Wide Web. Utilizing customary data recovery [5] and information mining systems it get to the known and obscure data from the Web content. Web mining is categorized into three group Web Content mining [6], Web structure mining, Web usage mining. Traditional web mining algorithms handle with structured document [7-15] than the advanced methodology of mining algorithm can dealthe entire heterogeneous document comprises of images [9], graphs, videos [16], etc. 2. Architecture of Proposed System A query is searched in a web search tool to recover some significant and required data for the client, either the search query is known or unintelligible to the client, it generally to reply with relevant data rather than redundant, however we can’t guarantee that the reply for the query about the significance and redundancy. Once the input query is requested, the search engine generate the document with multiple web pages along with the links, the user will be unaware about the content of the web pages, the extracted web documents contain multiple web pages either be redundant or not. The Document retrieved must follow some constraints which have less time & space requirements, based upon the criteria the extracted web document must be preprocessed, for preprocessing & information selection, need to apply some techniques such as stop word removal, Stemming of word, phrasing, normalization of tokens. Once the document is preprocessed, Normalization of tokens is generated to further process the web content document.