Customer centric sales analysis and prediction

Abstract

For successful business management, sales predic- tion plays an inevitable part. Data mining techniques have been employed since a long time for sales analysis. In the past, the prediction has been done from various point of views always keeping mind the needs of the customer and the profitability of the business in consideration. Initially, sales prediction has been done using Market Basket analysis, wherein using the previous data the next item, which is most likely to be purchased is predicted. At later stages, the products on the shelf were only considered for predicting sales in a supermarket. Thereafter, for a range of supermarkets the location and time was considered to predict the sales of items. For predicting the sales a number of algorithms such as AIS, Apriori, FP Growth, FP Bonsai have been employed. In this paper, the price or the amount to be likely spent by the customer will be predicted using various other algorithms and a comparison between the different algorithms is outlined.