In The increasing craze of internet has geared a number of modern firms for using web technology in their day to day lives. A remarkable ability to analyze web log data is provided to them by the web mining technology, that are putatively full of information, but frequently lacking with meaningful information. This signifies the need to the development of an inference mechanism which is advanced and that draws out richer implication from mining's output. This paper presents a web mining algorithm that aims at amending the interpretations of the draft's output of association rule mining. This algorithm is being tremendously used in web mining. The results obtained prove robustness of the algorithm proposed in this paper.