Title :
An improved algorithm of mining preferred browsing paths
Author :
Hongbo Li ; Ning Wang ; Yu Wu
Author_Institution :
Inst. of Web Intell., Chongqing Univ. of Posts & Telecommun., Chongqing, China
Abstract :
Existing algorithms of mining preferred browsing paths just consider the influence of user visiting times, but ignore the accuracy influenced by other factors. In order to solve the problem, an improved algorithm which imports page similarity and support-preference concepts is proposed. Firstly a Web-log-based user access matrix is set up. Then by calculating the angel cosine similarity and support-preference, the 2-items preferred browsing sub-path set is obtained. Finally all the sub-paths are combined. Experiments show that the algorithm is more accurate and efficient.
Keywords :
Internet; data mining; human computer interaction; system monitoring; UPBPA; Web-log-based user access matrix; angle cosine similarity; page similarity; preferred browsing path mining; support-preference concepts; use preferred browsing paths algorithm; user visiting times; Accuracy; Algorithm design and analysis; Computers; Data mining; Educational institutions; Uniform resource locators; Vectors; Preferred Browsing Paths; Similarity Matrix; Support-preference;
Conference_Titel :
Natural Computation (ICNC), 2014 10th International Conference on
Conference_Location :
Xiamen
Print_ISBN :
978-1-4799-5150-5
DOI :
10.1109/ICNC.2014.6975941
