International Journal of Engineering Research ISSN:2319-6890)(online),2347-5013(print) Volume No.5, Issue No.4, pp : 264-268 1 April 2016 IJER@2016 doi : 10.17950/ijer/v5s4/409 Page 264 A New Model for Profitable Pattern Mining Jagrat Gupta 1 , Akhilesh Tiwari 2 1,2 Department of Computer Science and Engineering, Madhav Institute of Technology & Science, Gwalior, India 1 jagrat.webtech@gmail.com, 2 atiwari.mits@gmail.com Abstract: Profitable pattern mining is a captivating research area that accentuates to adjudicate the business objectives.One of the most prominent and unaddressed business objectives regarding this discussion is Profit.The research problem as well as the prominent objective behind this research paper is to extract profitable rules accurately, efficiently as well as in optimized manner. To accomplish the above objective, incorporation of rough set theory followed by conventional association rule mining algorithm and genetic based optimization is used in optimized and efficient manner. The proposed model overcomes the major findings to make model more beneficial for any business organization in the current scenario. Keywords: Association rules, Data Mining, Genetic Algorithm, Market Basket Analysis, Profitable Pattern mining (PPM), Rough Set Theory (RST). I. Introduction Now-a-days, the explosive growth of amount of data gathered by systems has needed to analyze as well as discover interesting and non obvious information from such huge amount of data. This explosive gathering of data is possible only due to the technological advancements and available storage facilities. So there is urgent challenge as well as requirement for the development of tools and techniques for analyzing immense data. Data mining emerged as the new research area to meet this challenge. Data mining, also called Knowledge mining, Knowledge extraction, Data archaeology, Data Dredging, is the process for the extraction of valuable information from the huge amount of data. It is one of the most important analysis step of the “Knowledge Discovery in Databases” or KDD Process [i]. An official definition of KDD given by Usama Fayyad in 1996 is: “KDD or Data Mining is non-trivial process of identifying valid, novel, potentially useful and ultimately understandable patterns in data”. It was also suggested that data mining should be used for the discovery as well as analysis stage of KDD process. Another aspect of Data Mining is that different applications are incorporated by researchers during their research work but the first application in the context of Data Mining is Market Basket Analysis (MBA). Market Basket Analysis or super market analysis is a modeling technique which is widely used to identify the purchasing relationship between items.So in precisely MBA is a tool used in business intelligent decisions [ii]. For Example, Market basket analysis gives essential sales information about group of goods basis Customers who buys Bread often also buy several products related to Bread like Milk, Butter or Jam.Such related groups of goods also must be located side-by-side in order to remind customers of related items and to lead them through the center in a logical manner. Typically the relationship will be represented in the form of rules referred as Association rules. Extraction of Association rules is one of the most prominent Data Mining tasks which were given by R. Agrawal [iii]. Such rules describe the co-occurrence relationship among the set of items in a dataset [iv]. For Example: {Bread} {Milk or Butter or Jam} The probability that a customer will buy Bread or Milk or Jam is referred as theSupport for the rule. The Conditional probability that a customer will purchase Milk or Butter or Jam is referred as the Confidence. These rules as well as measures lead to analyze various types of scenarios related to market or organization. Literature indicates that several measures are addressed by researchers during their research work but still there are some measures that are required to get more attention by the researchers [v], [vi]. One of such measure is Profit that leads to the evolution of Profitable Pattern Mining. Next section describes the interpretation as well as the objective of Profitable Pattern Mining. II. Profitable Pattern Mining Now-a-days, prominent issueis whether a customer purchases an item recommended by organization. Regarding this, different factors like items stocked, competitors` offers, prices, promotions, recommendations by individuals, psychological issues, individual interest etc are in consideration. As far as implementation of all such factors are concerned, it is difficult to implement all these factors in a single model but still some factors can be taken into considerations and build up a model for the enhancement of Profit. All the above Considerations lead to the evolution of Profit Mining. So the major concern in Profit mining problem is to determine an item that is of the interest to customer at affordable price and also profitable for organization. The prominent objective of Profit Pattern Mining (PPM) is to develop a model which generates profitable rules as well as recommender rules that recommend target items for future customer [vii]. It is a new technique as well as extension of association Rule Mining which aims to extract those patterns which contributes maximum profit for organization. Following figure clearly explains level wise hierarchy of relationship from KDD Process to profit mining.