A sliding window method for finding top-k path traversal patterns over streaming Web click-sequences

  • Authors:
  • Hua-Fu Li

  • Affiliations:
  • Department of Computer Science, Kainan University, Taoyuan, Taiwan

  • Venue:
  • Expert Systems with Applications: An International Journal
  • Year:
  • 2009

Quantified Score

Hi-index 12.05

Visualization

Abstract

Online mining of path traversal patterns from Web click-streams is one of the most important problems of Web usage mining. In this paper, we propose a sliding window-based Web data mining algorithm, called Top-SW (Top-kpath traversal patterns of Stream sliding Window), to discover the set of top-k path traversal patterns from streaming maximal forward references, where k is the desired number of path traversal patterns to be mined. A new summary data structure, called Top-list (a list of Top-kpath traversal patterns) is developed to maintain the essential information about the top-k path traversal patterns from the current maximal forward references stream. Experimental studies show that the proposed Top-SW algorithm is an efficient, single-pass algorithm for mining the set of top-k path traversal patterns from a continuous stream of maximal forward references.