ABSTRACT
This research study contributes towards understanding the customer’s behaviour dynamics. In business analysis, it is very important not to ignore the fact that the interaction between human beings implicitly includes an emotional dimension. The research methodology includes the following: (1) customer purchase pattern prediction methods based on correlation; (2) augmentation of data set by using genetic algorithms; and (3) multiple regression models. The analysis indicates how the hobby of a customer is directly related to the purchase patterns and satisfaction level. We applied business intelligence (BI) techniques and concluded that, by using multiple regression method is possible to evaluate the level of customer satisfaction up to the upper limit of security of about 90%. BI tools could be used to employ significant achievements in specific fields based on open innovations. This paper aims at providing further practical guidance in this innovative research field by using a mix of interdisciplinary methods and techniques.
1. Introduction
The core idea of business intelligence (BI) is to recognise the behavior of the customer and to predict their purchase pattern for improvement of the business as well as for a better environmental sustainability. Efficient decision making based on BI is essential to ensure competitiveness for sustainable growth (Jin & Kim, 2018). However, in literature, there is no universally accepted definition of BI. BI plays an essential role in areas such as sales representatives’ performance, customer loyalty, and product performance (Athanasoulias & Chountalas, 2019). Remarkable BI applications have also been reported in a wide variety of occupational fields, from health care and airlines to major IT and telecommunication firms (Watson, 2009). Cebotarean (2011) suggested that BI is related to computer-based techniques used in spotting, digging-out, and analyzing business data, such as sales revenue by products and/or departments, or by associated costs and incomes. Larson and Chang (2016) argued that BI includes an information value chain for gathering raw data, turning these data into useful information, management decision making, driving business results, and raising corporate value.
5. Conclusions
Initially, the data records are clustered on the basis of weights using correlation technique, which is explained in the section 3.1. It is observed from the result and analysis that, with the huge data records, it is still possible to improve the accuracy of the result. As the data collection through communicating with the people is tedious and time consuming, it is decided to enhance the data using GA and is demonstrated in the section 3.2. The techniques of GA are applied with different bits of crossover operations and mutation. The newly generated data record is further mutated to modify its characteristics as per the GA procedure. GAs provide a well-established framework for implementing artificial intelligence tasks such as classification, learning, and optimisation. The results revealed that out of total records, 0.6 percentages are rejected due to improper data and also due to manual filtering.
