Mining of User Behavior Patterns in Computer Big Data Based on Apriori Algorithm

In the current era of information explosion, the mining of user behavior patterns in computer big data has immeasurable value for enterprises to gain insights into consumer needs, optimize services, and formulate market strategies. In this article, the Apriori algorithm is utilized to conduct in-depth analysis of user behavior data, aiming to identify and predict user behavior patterns, so that more precise decision support for enterprises can be provided. This study first involves data collection and preprocessing and then forms transaction sets suitable for Apriori algorithm analysis through feature extraction and transaction construction. In this article, the minimum support and confidence are determined, and the prediction ability of the rules is evaluated by the improvement degree. The experimental results show that the Apriori algorithm is superior to the Eclat algorithm and the Frequent Pattern Growth (FP-Growth) algorithm in terms of execution time and accuracy. Specifically, the average execution time of the Apriori algorithm is 296.84ms, and the accuracy is between 90.4% and 99.8%, demonstrating its high efficiency and accuracy in handling user behavior pattern mining of big data.