X-tracking the usage interest on web sites [Elektronische Ressource] / von Long Wang

Long Wang

Sujets

• Informatik

Informations

• Publié par: universitat_potsdam
• Publié le: 01 janvier 2009
• Nombre de lectures: 77
• Langue: Deutsch
• Poids de l'ouvrage: 3 Mo

Extrait

Hasso–Plattner–Institut fur¨ Softwaresystemtechnik
an der Universitat¨ Potsdam
X-tracking the Usage Interest on Web Sites
Dissertation
zur Erlangung des akademischen Grades
”Doctor rerum naturalium”
(Dr. rer. nat.)
am Fachgebiet Internet Technologien und Systeme
eingereicht an der
Mathematisch–Naturwissenschaftlichen Fakultaet
der Universitaet Potsdam
von
Long Wang
Potsdam, December 8, 2009
i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Published online at the 
Institutional Repository of the University of Potsdam: 
URL http://opus.kobv.de/ubp/volltexte/2011/5107/ 
URN urn:nbn:de:kobv:517‐opus‐51077 
http://nbn‐resolving.org/urn:nbn:de:kobv:517‐opus‐51077 Contents
Acknowledgements vii
Abstract viii
1 Introduction 1
1.1 Related Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Thesis Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
I Discovering the Changes of Usage Interest on a Portal Site 7
2 Web-Cares: A Platform to Track the Web Usage Interest 9
2.1 Input of Web-Cares . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Work Flow of Web-Cares . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Summary for This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3 Data Preparation in Web Usage Mining 15
3.1 Collecting Server Logs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 Robots Removal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.3 Sessions Reconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.4 Rebuilding Individual Accessing Behaviors . . . . . . . . . . . . . . . . . . . . . . . 16
3.4.1 Problem statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.4.2 Algorithms on rebuilding individual access behaviors . . . . . . . . . . . . 19
3.4.2.1 Rebuilding simple behaviors . . . . . . . . . . . . . . . . . . . . . . 19
3.4.2.2 tree structure behaviors . . . . . . . . . . . . . . . . . . 20
3.4.2.3 Rebuilding acyclic routing . . . . . . . . . . . . . . . . . 20
3.4.2.4 cyclic routing behaviors . . . . . . . . . . . . . . . . . . 22
3.5 URL Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.6 Summary for This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
iiiContents
4 Modeling and Discovering Usage Patterns 25
4.1 Starting from Four Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.2 Mining Usage Patterns on a Web Site . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2.1 Mining frequently co-accessed pages . . . . . . . . . . . . . . . . . . . . . . 28
4.2.2 fr accessed page sequences . . . . . . . . . . . . . . . . . . . 29
4.2.3 Mining frequently tree structures . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.3 Page Clustering based on Pair Browsing . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.3.1 Heuristic importance within a page pair . . . . . . . . . . . . . . . . . . . . . 34
4.3.2 Clustering method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.3.3 Site modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.4 X-tracking the Changes of Usage Patterns . . . . . . . . . . . . . . . . . . . . . . . . 37
4.4.1 Measuring internal change . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.4.2 external . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.4.3 Measuring local popularity change . . . . . . . . . . . . . . . . . . . . . . . . 41
4.4.4 global change . . . . . . . . . . . . . . . . . . . . . . . 42
4.4.5 Algorithms for X-tracking the changes . . . . . . . . . . . . . . . . . . . . . . 42
4.5 Pattern Interpretation and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.5.1 Objective: Technique standard . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.5.2 Subjective: Expert standard . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.5.3 Task oriented standard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.6 Summary for This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5 Results Discussion on Web-Cares 47
5.1 General Usage Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.2 Rebuilding Individual Behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.3 Discovering Usage Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.3.1 Frequent navigation patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.3.2 Content clusters based on pair browsing . . . . . . . . . . . . . . . . . . . . 53
5.3.3 Case studies on frequent patterns . . . . . . . . . . . . . . . . . . . . . . . . 53
5.4 X-tracking the Changes of Usage Patterns . . . . . . . . . . . . . . . . . . . . . . . . 55
5.4.1 Site semantics extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.4.2 Frequent patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.4.3 Case studies on change detections . . . . . . . . . . . . . . . . . . . . . . . . 60
5.4.4 Time analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.5 Summary for This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
II Mining the Learning Interest in a Web-Streaming E-learning Site 63
6 TASK-Moniminer: An Engine to Query the Learning Interest on tele-TASK 65
6.1 tele-TASK: A Web-Based e-Learning Environment . . . . . . . . . . . . . . . . . . . 65
6.2 Implementation of TASK-Monimier . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
ivContents
7 Issues on Mining Learning Interest in the tele-Teaching Environment 71
7.1 Relate Works in Web Video-based e-Learning Environment . . . . . . . . . . . . . . 71
7.2 Related Works on Evaluating Participation in e-Learning . . . . . . . . . . . . . . . 72
7.3 Difficulties and Arguments on Evaluation . . . . . . . . . . . . . . . . . . . . . . . . 73
7.4 Our Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
8 Modeling and Discovering Learning Interest in Different Questions 75
8.1 Data Preparation in e-Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.1.1 Cleaning and integrating learning usage data . . . . . . . . . . . . . . . . . . 75
8.1.2 Modeling student learning profiles . . . . . . . . . . . . . . . . . . . . . . . . 76
8.1.2.1 Computing the number of operations . . . . . . . . . . . . . . . . . 77
8.1.2.2 the duration . . . . . . . . . . . . . . . . . . . . . . . . 77
8.2 Answering the Six Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
8.2.1 Is there any difference between viewing the live broadcasting lectures and
browsing lectures after they are recorded and edited? . . . . . . . . . . . . . 79
8.2.2 Is there any preference on the different lectures in a course and preference
on the different pieces of one lecture? . . . . . . . . . . . . . . . . . . . . . . 79
8.2.3 Is there any favor among real, mp4 and flash formats? . . . . . . . . . . . . 79
8.2.4 For one lecture, is the real video viewed together with its mp4 and flash clips? 80
8.2.5 Do the students view other lectures when they access one lecture? . . . . . 80
8.2.6 For the same named course supplied for different years, is there any change
on the students’ interest? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
8.2.6.1 Statements on types of changes . . . . . . . . . . . . . . . . . . . . 81
8.2.6.2 Measuring changes of usages . . . . . . . . . . . . . . . . . . . . . 83
8.2.6.3 Similarity comparison between two learning objects . . . . . . . . 83
8.2.6.4 Measuring changes of learning interest . . . . . . . . . . . . . . . . 84
8.3 Summary for This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
9 Results Discussion on TASK-Moniminer 87
9.1 General Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
9.2 Results on Answering 5 Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
9.3 on Detecting Difference of Learning Interest on Similar Courses . . . . . . 91
9.4 Summary for This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
III Recommending High Reputation Articles in a Social Site 95
10 Re-Blog: A Mechanism to Recommend High Reputation Articles on IT-Gipfelblog 97
10.1 IT-Gipfelblog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
10.1.1 Content of IT-Gipfelblog . . . . . . . . . . . .