HUIM-SMP: High utility itemset mining based stock market analogy

Stock market prediction is a method of predicting the futuristic value of a company stock. The predictive analysis can be performed using fundamental, technical, or data mining techniques. One of the most prominent method of data mining is high utility itemset mining which yields the profit of itemsets in a transaction database equal or more than a user-defined minimum utility threshold. HUI-miner is a one phase HUI mining algorithm which uses a novel list structure called utility-list. Utility-list stores the information to apply the pruning method and utility check. In this paper, we mainly focus on prediction through HUI mining algorithms. We introduce a new method called HUI Miner for Stock Market Prediction (HUIMSMP) to predict stockset from a stock dataset that yields the maximum profit. HUIMSMP uses utility-list structure to store the utility and heuristic information of a stockset to prune weak candidates from search space. The proposed application revolves around the issues of a customer regarding a safe investment in the stock market with high profits. Extensive experiments are conducted to evaluate the performance of HUIMSMP. The results show that the proposed algorithm reduces the number of candidates and runs effectively when a database contains number of long transactions.