Web usage mining is the process of extracting useful information from web server logs, user interactions, and other web data in order to understand and optimize how users interact with a website. This technique involves analyzing data generated by user activities to discover patterns, trends, and insights that can inform various business and technical decisions. In this research article, we use the Hadoop technique to analyze the behavior of user navigation patterns from log data. This method helps to understand how users interact with websites. For instance, we track the quantity of user logins, the quantity of item sets, the frequency of item sets’ searches, and determine the execution time of various web usage mining algorithms across varying support counts. This article proposes a joint SPM-a priori web mining technique that uses semantic information to improve the generation of frequent sequences and reduces execution time for different support counts. We com-pared existing a priori, web log search, and modified web log search with the proposed join SPM a priori algorithm in Sci-Hub, Pubg game log, Amazon e-commerce, and cryptocurrency datasets. We also found out how long each one took to run for different support counts. The results section says that the proposed SPM-a priori algorithm runs faster than all other algorithms. This means that the proposed system is better for analyzing execution time for user navigational pattern analysis.