Heuristics Miner for E-Commerce Visitor Access Pattern Representation

• Published in: Communications in science and technology Vol. 2; no. 1; pp. 1 - 5
• Main Authors: Wardhani, Kartina Diah Kesuma, Yunanto, Wawan
• Format: Journal Article
• Language: English
• Published: Komunitas Ilmuwan dan Profesional Muslim Indonesia 01.06.2017
• Subjects: Access Pattern; Heuristic Miner; Visitor Behavior
• ISSN: 2502-9258; 2502-9266
• DOI: 10.21924/cst.2.1.2017.21

E-commerce click stream data can form a certain pattern that describe visitor behavior while surfing the e-commerce website. This pattern can be used to initiate a design to determine alternative access sequence on the website. This research use heuristic miner algorithm to determine the pattern. σ-Algorithm and Genetic Mining are methods used for pattern recognition with frequent sequence item set approach. Heuristic Miner is an evolved form of those methods. σ-Algorithm assume that an activity in a website, that has been recorded in the data log, is a complete sequence from start to finish, without any tolerance to incomplete data or data with noise. On the other hand, Genetic Mining is a method that tolerate incomplete data or data with noise, so it can generate a more detailed e-commerce visitor access pattern. In this study, the same sequence of events obtained from six-generated patterns. The resulting pattern of visitor access is that visitors are often access the home page and then the product category page or the home page and then the full text search page.