Graph Structures and Algorithms for Query-Log Analysis

• Published in: Programs, Proofs, Processes pp. 126 - 131
• Main Author: Donato, Debora
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg
• Series: Lecture Notes in Computer Science
• Subjects: graph mining; link mining; query recommendation; query-log mining; web mining; world wide web
• ISBN: 9783642139611; 3642139612
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-642-13962-8_14

Query logs are repositories that record all the interactions of users with a search engine. This incredibly rich user behavior data can be modeled using appropriate graph structures. In the recent years there has been an increasing amount of literature on studying properties, models, and algorithms for query-log graphs. Understanding the structure of such graphs, modeling user querying patterns, and designing algorithms for leveraging the latent knowledge (also known as the wisdom of the crowds) in those graphs introduces new challenges in the field of graph mining. The main goal of this paper is to present the reader with an example of these graph-structures, i.e., the Query-flow graph. This representation has been shown extremely effective for modeling user querying patterns and has been extensively used for developing real time applications. Moreover we present graph-based algorithmic solutions applied in the context of problems appearing in web applications as query recommendation and user-session segmentation.