Comparative Study of Consumer Purchasing and Decision Pattern Analysis using Pincer Search Based Data Mining Method

Big data analysis is now widely used as a result of the Internet's quick development. Researchers have focused a lot of attention on data mining because it is essential for obtaining potentially valuable information from big data. Python is a popular programming language used in data mining. Python is regarded as an indispensable tool for data mining because of its robust scientific calculation capabilities and rich database. There have been many consumer purchasing behaviour studies that have been presented and applied to actual issues. In-depth consumer behaviour analysis will likely benefit from the use of data mining techniques. Both benefits and drawbacks of the data mining approach exist, though. To mine databases effectively, it is crucial to choose the right techniques. This paper aims to apply several methods, such as pincer search-based association rule generation, to enhance conventional data mining analysis. The a priori algorithm uses a bottom-up, breadth-first search approach. The computation begins with the lowest set of frequent item sets and advances until it reaches the greatest frequent item set. The greatest size of the frequent item collection is equivalent to the number of database passes. The algorithm must go through numerous iterations as a result of any one of the frequent item sets getting longer, which lowers performance. To overcome this challenge, this paper aims to propose pincer searching-based association rule generation based on big data mining and classification algorithms for effective accuracy. The wholesale data market is used for simulation purposes.