Certain investigations on customer behaviour analysis for effective management using enhanced model in big data analytics

Abstract

The online retailing become an integral part of retailing today. The growth on internet brings sophisticated shopping on fly starting from electronics till groceries and veggies. The covidand#8223;19 pandemic situation puts even cars and on online stores. Sensing customers is important to serve them better, which has direct impact on revenue. The data produced by online stores are huge and informative, many of the data collected are either not used or seldom visited. The customer behaviour analysis is the recommendation framework to analyse these customer data. The behaviour analytics analyses the insight and revels about which customers bought what and when. It also recommends the channel which is competitively advantageous based on the data-based marketing decisions. Such customer behaviour analysis also recommends the system about web design to enhance the customer experience, predictions about their likelihood to increase the customer satisfaction and sales, inventory maintenance. There many systems established for instigating the customer behaviour analysis but still it has many rooms to explore and greater potential for enhancing the same. The Bigdata Analytics is a rising technology which is promising one for handling the e-commerce data for customer behaviour analysis. The bigdata has the advantage of accommodating the heterogenous data in the native format and processing in relatively low cost. As the growth of social media and personal digital assistant yield data in different structure or even unstructured. Although the data is huge, big data tools can perform parallel processing which reduces significant processing time. The machine learning algorithms are capable of providing projections of metrics such as loyalty of the customer, affinity, transaction value, and probability of purchase. This helps the retailer in adjecting their stock and campaigns, and changing the business strategies. This thesis work proposes two models namely Mouse Movement Pattern based Analysis of Customer Behaviour (CBA-MMP) and Enhanced Model for Customer Behaviour and Purchase Analysis newline