Weighted Clickstream Mining Using Pre-order Linked Web-Access Pattern Tree

Data mining is the knowledge-discovery process by analyzing the massive volumes of data from many perspectives, summarizing it into useful information. The recent growth of Internet usage has opened new doors to commercial advertisements. Such advertisements can be placed more effectively, analyzing customers’ web-access behavior. Click-stream pattern mining, a cardinal branch of data mining, focuses on the users’ web-access behavior and discovers such patterns. Many methods have been introduced for click-stream mining. Yet, most of these methods produce an enormous number of pointless patterns which need more time to execute. This paper proposes an efficient way to mine interesting patterns using weight constraint through an efficient anti-monotonic pruning measure. Execution assessment on real datasets shows that this method functions admirably faster than other existing methodologies.