Volume II, Issue X, October 2015 IJRSI ISSN 2321 - 2705 www.rsisinternational.org Page 89 Optimal Customer Tracker using Behaviour Analysis Rohan Kishan 1 , Prashant Shinde 2 , Sachin Mane 3 , Sandhya Pati 4 Department of Computer Engineering Fr. Conceicao Rodrigues Institute of Technology Abstract—Owing to increasing competition between organizations, securing the support of customers has become a must for every company to survive in the world of business. Customer satisfaction has become the prime objective of organizations. In order to ensure this, an expert system is required to identify beneficial customers. In this way the company can provide them lucrative offers in order to retain their fellowship with the company for a longer period of time and in due course also earn extra profits. Providing offers to some selected customers is a better alternative to declare reduced prices for everyone as the revenue of the company can be spent efficiently. This strategy will also encourage other customers to buy more products and increase the company’s profits. We propose a system that has been developed with the sole objective of identifying such beneficial customers based on the frequency of their visits and the profits that they help the company to attain. Keywords-Expert System, Beneficial Customers, lucrative offers. I. INTRODUCTION ustomer Relationship Management (CRM) and Data Mining are two distinct areas whose use by both private and public organisations has risen dramatically [i][vi]. The possible competitive advantages of using these technologies are well documented, with examples in most markets. Organisations such as the Bank of America, Tesco, Walmart, AT&T BT have benefited from adopting Data Mining to better inform and execute their CRM operations [iv]. The aim of our system is to analyse the current CRM and Data Mining operations of Organisations and formulate innovative domain specific strategies that harness the power of these technologies to create real business value. Many literature studies are available to preserve the customer relationship but small drawbacks occur in the existing methods. One method to maintain the customer relationship is frequency based method i.e., the company will give declination to the customer based on the historical data i.e. the number of times the customer visits the company. These methods are not effective because the revenue generated by these customers is less. So the company revenue is affected. We propose a framework for analysing customer value and segmenting customers based on their value. II. LITERATURE REVIEW Han & Kamber (2006) suggest that a main driver behind the evolution of data mining is “the wide availability of huge amounts of data and the imminent need for turning such data into useful information and knowledge” [iii]. The development of Data Mining techniques has therefore been motivated by the need to overcome this challenge, and is a direct consequence of the continual development of the information storage paradigm. Data Mining is an important component of the Customer Portfolio Analysis (CPA) and Customer Intimacy phases of his CRM model [ii]. CPA is essentially about identifying significant or profitable customers which the organisations should look to continue its relationship with [ii]. For this to occur, Data Mining techniques such as Classification are used to identify which customers these are with Buttle providing an industrial example from Natwest Corporate Banking Services[ii]. Natwest used classification on attributes such as Lifetime Value and Credit Rating to sort its client base into a number of segments ranging in significance and importance. From this Natwest developed a CRM strategy that varied its assignment of staff to each client, with the top most significant receiving an individual relationship manager, while the least significant had access to a business advisor. 2.1 Existing Systems Most data mining tools resulted in customer models and visualization results. It means the calculation results were analysed and represented as graphs [iv]. So the human experts have to take the actions according to the results of ROC [viii][vi] (receiver operating characteristic) curve. The actions taken by the human experts may vary. As a result it is difficult to find a set of analysis to process customer relations. The actions will be taken as per the experts own idea. When data mining techniques are applied to CRM it resulted in finding out customer models and behaviours as graphical representations. III. PROBLEM STATEMENT The current Globalization of commerce, also denoted as “Globalization 3.0” has created an increasingly competitive market, where competition is no longer set by geographic constraints [ix]. Within this global market place, and with a vast choice of suppliers for consumers, Organisations have had to tailor their value proposition in order to remain competitive. CRM has therefore become a central focus of any organisation, and has coincided with the commercialisation of Data Mining, which has provided meaningful intelligence on the customers an Organisation serves [ii]. Data Mining and CRM strategy are common place in most large organisations, and this project aims to evaluate their current adoption, while also C