WWW.DISSERTATION.XLIBX.INFO
FREE ELECTRONIC LIBRARY - Dissertations, online materials
 
<< HOME
CONTACTS



Pages:   || 2 | 3 | 4 | 5 |   ...   | 13 |

«Dissertation submitted to the faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the ...»

-- [ Page 1 ] --

Visualizing Users, User Communities, and Usage Trends in

Complex Information Systems Using Implicit Rating Data

Seonho Kim

Dissertation submitted to the faculty of the

Virginia Polytechnic Institute and State University

in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

in

Computer Science and Applications

Advisory Committee:

Edward A. Fox, Chair

Weiguo Fan

Christopher North

Deborah Tatar

Ricardo da Silva Torres

April 14, 2008 Blacksburg, Virginia Keywords: digital library, information system, visualization, 5S, social network, user modeling, personalization, implicit data, quantity-based evaluation Copyright © 2008 Seonho Kim All Rights Reserved Visualizing Users, User Communities, and Usage Trends in Complex Information Systems Using Implicit Rating Data Seonho Kim

ABSTRACT

Research on personalization, including recommender systems, focuses on applications such as in online shopping malls and simple information systems. These systems consider user profile and item information obtained from data explicitly entered by users. There it is possible to classify items involved and to personalize based on a direct mapping from user or user group to item or item group. However, in complex, dynamic, and professional information systems, such as digital libraries, additional capabilities are needed to achieve personalization to support their distinctive features: large numbers of digital objects, dynamic updates, sparse rating data, biased rating data on specific items, and challenges in getting explicit rating data from users. For this reason, more research on implicit rating data is recommended, because it is easy to obtain, suffers less from terminology issues, is more informative, and contains more user-centered information.

In previous reports on my doctoral work, I discussed collecting, storing, processing, and utilizing implicit rating data of digital libraries for analysis and decision support. This dissertation presents a visualization tool, VUDM (Visual User-model Data Mining tool), utilizing implicit rating data, to demonstrate the effectiveness of implicit rating data in characterizing users, user communities, and usage trends of digital libraries. The results of user studies, performed both with typical end-users and with library experts, to test the usefulness of VUDM, support that implicit rating data is useful and can be utilized for digital library analysis software, so that both end users and experts can benefit.

To my father Kiro Kim and mother Heeja Lee To my brothers

–  –  –

First and foremost, I thank my advisor Dr. Edward A. Fox, for his warm guidance and advice not just on academic knowledge but also about the right way of thinking and studying. I admire his passion for new ideas and enthusiasm for work. I admire his being gentle in appearance, but sturdy in spirit. I would have never survived my graduate program without his endless support. I also give heartfelt thanks to my other committee members, Dr. Fan, Dr. North, Dr. Tatar, and Dr. Torres for their excellent suggestions regarding research directions. They offered many helpful critiques and thoughtful comments on my work that lead to improving the quality of this dissertation.

Also, I thank members of the Digital Library Research Laboratory (DLRL) for their words of encouragement, and helpful suggestions. These people from the DLRL include, but are not limited to, Venkataraghavan, Yi, Yifei, Xiaomo, Xiaoyu, Pengbo, Ming, Bing, Dr. Weon, Dr. Chakraborty, Srinivas, Ryan, Yuxin, Wensi, Baoping, Lokeya, Hussein, Douglas, Johnny, Vikram, Rohit, and Unni. I specially thank Uma, Seungwon, Xiaoyan, and Rao for their valuable discussion, comments and being domain experts to verify my experiments. I don‟t want to forget those who have worked together with me. These people include, Nithiwat, Nicholas, Kapil, Subodh, Sandi, and Sheng. Their co-work formed the basis of my doctoral work.

In addition, I thank my good Korean friends in Torgersen Hall and the CS department, Jaehoon, Ji-Sun, Jason, Yongju, Yoon-Soo, Hannah, Si-Jung, Taehyuk, Joon Seok, Sunghee, Seung In, Myoungkyu, Jae-seung, and Haeyong, for helping my experiment.

Special thanks go to Kibum, Pilsung and Dong Kwan. They took care of me brotherly and encouraged me to overcome many difficulties.

Also, I must thank all my advisors that helped me until now, Dr. Jong-Hyeok Lee, and Dr. Geunbae Lee, at POSTECH, Dr. Dong-Yul Ra, Dr. Jong Hyun Kim, and Dr. JungBong Suk, at Yonsei University, Dr. Sun-Hwa Hahn, and Dr. Taehee Kim (now moved to Yongsan Univ.) at KISTI. I also thank Dr. Kandogan, at IBM Almaden Research Center, for providing me a great research opportunity and for his effort toward patent work.

Finally, I thank my parents and my brothers for their love and patience. Their continuous support and trust were chief factors in my success.

iv Table of Contents

Abstract

……………………………………………………………………………. ii Acknowledgements ………………………………………………………………….. iii Table of Contents

List of Figures

List of Tables





Chapter 1. Introduction

1.1 Overview and Motivation

1.2 Problem Statement

1.2.1 Needs for Personalized User Interfaces

1.2.2 Difficulties in Classifying Digital Library Materials

1.2.3 Limitations of Available Data

1.2.4 Lack of Analysis Tools for Digital Libraries

1.2.5 Needs for User-Centered Evaluation

1.3 Research Objectives

1.4 Research Question and Hypotheses

1.5 Contributions

1.6 Organization of this Dissertation

Chapter 2. Review of the Literature

2.1 Data Mining Applied to the Web

2.2 Digital Library Behavior and Usage Analysis

2.3 Recommender System for Digital Library

2.4 Visualizing Social Networks, Documents, and Topics

2.5 User Modeling

2.6 Implicit Rating Data

2.7 Evaluation of Information Visualization

Chapter 3. 5S Perspective of Digital Library Personalization

3.1 Society

3.2 Scenario

v 3.3 Structure

3.4 Space

3.5 Stream

Chapter 4. Collecting Data and Preprocessing

4.1 User Tracking System: Collecting Implicit Rating Data

4.1.1 About NDLTD

4.1.2 User Tracking System

4.2 User Survey: Collecting Explicit Rating Data

4.3 User Models: Storing Raw Data and Statistics

4.4 Using Legacy Data: VTLS NDLTD log data

Chapter 5. Visualization and Knowledge Finding

5.1 VUDM: A Visualization Tool

5.1.1 Need for a Visualization Tool for Digital Library Users

5.1.2 Data Description and Preprocessing

5.2 Visualization Strategies

5.3 Loose Grouping Algorithm

5.4 Knowledge Finding

5.4.1 User Characteristics and Relations

5.4.2 Virtual Interest Group and Relations

5.4.3 Usage Trend

5.4.4 Concept Drift

5.5 User Evaluation

Chapter 6. Formative Evaluation

6.1 Evaluation Design

6.2 Results and Conclusions

Chapter 7. Algorithm Evaluation and Enhancing Scalability

7.1 Community Finding Performance

7.2 Building a Test Collection

7.3 Document Classification and Noun Phrasing

7.4 Application: Measuring Accuracy of Community Finding Algorithms........... 52 7.5 Enhancing Scalability of FSWMC

vi 7.6 Results and Conclusions

Chapter 8. Summative Evaluation

8.1 Experiment Design

8.2 General Data Obtained

8.3 Rating Data Obtained

8.3.1 Accuracy

8.3.2 Efficiency

8.4 Conclusions

Chapter 9. Experts Review

9.1 Study Design

9.2 General Information Obtained

9.3 Ratings and Comments for Enhancement

9.4 Conclusions

Chapter 10. Conclusions and Future Work

10.1 Deliverables

10.2 Conclusions

10.3 Future Work

References

Appendix A-1: Call for Participation Advertisement

Appendix A-2: Questionnaire

Appendix A-3: Institutional Review Board Approval Letter

Appendix B-1: Call for Participation Advertisement

Appendix B-2: Questionnaire

Appendix B-3: Online Tutorial of VUDM for Library Experts Survey.................. 113 Appendix B-4: Institutional Review Board Approval Letter

vii List of Figures

Figure 1. A tag cloud with terms related to “5S Model”

Figure 2. A Themescape with documents related to the Virginia Tech April 16 incident

Figure 3. A GalaxyView with documents related to the Virginia Tech April 16 incident

Figure 4. 5S perspective of digital library personalization

Figure 5. Structure of user model data

Figure 6. A JavaScript based user interface

Figure 7. System diagram of user tracking system

Figure 8. XML schema of user model data

Figure 9. Iteration of data mining process using visualization tools

Figure 10. The main window of VUDM shows an overview of virtual user communities

Figure 11. Windows for detailed information about users and user communities

Figure 12. Zoomed user space

Figure 13. Filtering user space

Figure 14. Detailed information on demand

Figure 15. Illustration of the kNN (top) and FSWMC (bottom) algorithms

Figure 16. Fixed-size window multi-classification algorithm

Figure 17. User and user community characteristics and relations

Figure 18. Visualizing usage trends of digital library

Figure 19. Detecting drift of concepts makes it possible to provide timely recommendation.

..............42 Figure 20. Workflow to build a collection of research interests and learning topics

Figure 21. Hierarchical structure of the collection

Figure 22. An example of ETD metadata

Figure 23. Visualization of user communities in the NDLTD collection as of June 2006.

User communities were found by kNN (top), FSWMC (middle), and Tanimoto (bottom) algorithm. 53 Figure 24. EstimateCategories Function

Figure 25. Counting shared interests using the output of EstimateCategories Function

Figure 26. Change of the number of total users and active users

Figure 27. Change of the number of active users in terms of DOA, zoomed-in from Figure 24.

.........58 Figure 28. Completion time of FSWMC runs in O(n2)

Figure 29. Completion time of FSWMC runs in O(n√n)

Figure 30. Completion time of FSWMC runs in O(n)

Figure 31. Comparison of completion times of the FSWMC and its two variations

Figure 32. Completion times of the kNN, Tanimoto, FSWMC and its two variations

Figure 33. Accuracies of five community finding algorithms

viii Figure 34. Accuracy and size of community

Figure 35. Academic level of participants

Figure 36. Age distribution of participants

Figure 37. Preferred statistical analysis software package

Figure 38. Average ratio of correct, given-up, and wrong answers of VUDM (left bar), and other software packages (right bar)

Figure 39. Average ratio of correct, given-up, and wrong answers of VUDM (left bar), and Excel (right bar)

Figure 40. Average accuracy and confidence intervals of overall user tasks

Figure 41. Averages accuracy and confidence interval of each user task

Figure 42. Averages of overall completion time and confidence intervals

Figure 43. Average completion time and confidence interval of each user task

Figure 44. Average scores and confidence intervals of two different academic levels in terms of software used (left), and in terms of academic levels of participants (right)

Figure 45. Average completion times and confidence intervals of two different software within academic levels (left), and academic levels within two different software (right)

Figure 46. IRB approval letter for the summative evaluation

Figure 47. Java version checking

Figure 48. VUDM demo is installed under C:\VUDM

Figure 49. Invoke VUDM

Figure 50. Open new data in current window

Figure 51. Open new data in a new window

Figure 52. Select a data folder

Figure 53. Data loading

Figure 54. User data of February 2006 is successfully loaded and visualized

Figure 55. VUDM overview

Figure 56. VUDM visualization strategy

Figure 57. Zoomed image of a community

Figure 58. Control the strictness of community finding by adjusting the θ value

Figure 59. User community of January 2006 (θ = 0.

2)

Figure 60. User community of January 2006 (θ = 0.

18)

Figure 61. User community of January 2006 (θ = 0.

15)

Figure 62. User community of February 2006 (θ = 0.

15)

Figure 63. Details on demand

Figure 64. IRB approval letter for the experts review

ix List of Tables

Table 1. Data set: User model data used for VUDM formative evaluation, from 1,200 users.

.............30 Table 2. 102 categories of research interests and learning topics

Table 3. Raw data for results of analysis

Table 4. Raw data for evaluation

Table 5. Two task sets containing five user tasks

Table 6. t-test result on accuracy of overall user tasks between VUDM and other software packages (top), and between VUDM and Excel (bottom)

Table 7. Average accuracy and confidence interval of each user tasks

Table 8. t-test result on the accuracy of each user task between VUDM and other software packages

Table 9. t-test result on overall completion time between VUDM and other software packages (top), and between VUDM and Excel (bottom)

Table 10. Average completion time and confidence interval of each user task

Table 11. t-test result on the completion time of each user task between VUDM and other software packages

Table 12. Results Interpretation

Table 13. Pros and Cons of VUDM

Table 14. The effect of academic level

Table 15. Importance and uses of information about users, user communities, and usage trends.



Pages:   || 2 | 3 | 4 | 5 |   ...   | 13 |


Similar works:

«10 YEARS MIAMI SHORTFESTIVAL FILM NOV.12TH 20TH2011 @THE COLONY THEATER ON LINCOLN ROAD Cover Photography by Mauricio Candela IT’S MIAMI. IT HAS TO BE SHORTS. PRESENTED BY WELCOME. Welcome to the Miami short Film Festival! Since the festival’s inception in 2002, the Miami short Film Festival (MsFF) has grown in size and reputation. Our passion for the art of filmmaking is palpable. With a philosophy that today’s short films lead to tomorrow’s world-renowned filmmakers, the festival has...»

«FABRICATION, CHARACTERIZATION, AND APPLICATION OF MULTIFUNCTIONAL MICROCANTILEVER HEATERS A Dissertation Presented to The Academic Faculty by Jungchul Lee In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the George W. Woodruff School of Mechanical Engineering Georgia Institute of Technology May 2007 FABRICATION, CHARACTERIZATION, AND APPLICATION OF MULTIFUNCTIONAL MICROCANTILEVER HEATERS Approved by: Dr. William P. King, Advisor Dr. Mark G. Allen George W....»

«Open Journal of Philosophy, 2016, 6, 176-183 Published Online May 2016 in SciRes. http://www.scirp.org/journal/ojpp http://dx.doi.org/10.4236/ojpp.2016.62016 China-West Interculture Kuangming Wu Philosophy Department, University of Wisconsin-Oshkosh, Oshkosh, WI, USA Received 3 March 2016; accepted 2 May 2016; published 5 May 2016 Copyright © 2016 by author and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY)....»

«Journal of Jewish Thought & Philosophy �� (�0�4) ��7-�63 brill.com/jjtp “But I Will Tell of Their Deeds”: Retelling a Hasidic Tale about the Power of Storytelling Levi Cooper Faculty of Law, Bar-Ilan University levicoops@gmail.com Abstract A famous Hasidic tale that depicts the decline of mysticism in Hasidic circles also bespeaks the power of storytelling. This study tracks the metamorphosis of this classic tale over a century of its retelling by writers—including Martin...»

«Forthcoming in Philosophy East & West 57:4 (2007) Language and Ontology in Early Chinese Thought∗ Chris Fraser Department of Philosophy Chinese University of Hong Kong July 2005 (Revised January 2007) Correspondence: Chris Fraser (方克濤) (Assistant Professor) Department of Philosophy Rm. 430, Fung King Hey Bldg. Chinese University of Hong Kong Shatin, N.T., Hong Kong Telephone: 852-9782-0560 Fax: 852-2603-5323 E-mail: cjfraser@cuhk.edu.hk Copyright © 2005, 2007 Brief Summary for Table of...»

«POOL AND FLOW BOILING OF NOVEL HEAT TRANSFER FLUIDS FROM NANOSTRUCTURED SURFACES A Dissertation Presented to The Academic Faculty by Aravind Sathyanarayana In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the George W. Woodruff School of Mechanical Engineering Georgia Institute of Technology December 2013 Copyright © 2013 by Aravind Sathyanarayana POOL AND FLOW BOILING OF NOVEL HEAT TRANSFER FLUIDS FROM NANOSTRUCTURED SURFACES Approved by: Dr. Yogendra Joshi,...»

«Osmoticand Stroke-Induced Blood-Brain Barrier Disruption Detected by Manganese-Enhanced Magnetic Resonance Imaging A dissertation submitted to the faculty of Worcester Polytechnic Institute in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biomedical Engineering by David G. Bennett June 2007 Approved: Christopher H. Sotak, Ph.D. Major Advisor Professor Department of Biomedical Engineering Worcester Polytechnic Institute George D. Pins, Ph.D. Karl G. Helmer,...»

«Making Room: The Geography of Rooming House Regulation in Toronto. by Lisa Marie Freeman A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Geography University of Toronto © Copyright by Lisa Marie Freeman 2013 Making Room: The Geography of Rooming House Regulation in Toronto Lisa Marie Freeman Doctor of Philosophy Department of Geography University of Toronto 2013 Abstract This dissertation addresses the contemporary moment of the...»

«INVESTIGATION OF OPTICAL LOSS CHANGES IN SILOXANE POLYMER WAVEGUIDES DURING THERMAL CURING AND AGING A Dissertation Presented to The Academic Faculty by Shashikant G. Hegde In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the G.W. Woodruff School of Mechanical Engineering Georgia Institute of Technology April 2008 COPYRIGHT 2008 BY SHASHIKANT G. HEGDE INVESTIGATION OF OPTICAL LOSS CHANGES IN SILOXANE POLYMER WAVEGUIDES DURING THERMAL CURING AND AGING Approved...»

«kwb.45-variantology_inhalt_neu:kwb.31-variantology_end_5.0 19.02.10 14:12 Seite 215 david link Scrambling T-R-U-T-H Rotating Letters as a Material Form of Thought “Know that the secrets of God and the objects of His science, the subtle realities and the dense realities, the things of above and the things from below, belong to two categories: there are numbers and there are letters. The secrets of the letters are in the numbers, and the epiphanies of the numbers are in the letters. The numbers...»

«QUASARS, CARBON, AND SUPERNOVAE: EXPLORING THE DISTRIBUTION OF ELEMENTS IN AN EXPANDING UNIVERSE by Shailendra Kumar Vikas Bachelor of Technology, Indian Institute of Technology, Kharagpur, 2001 Master of Science, University of Pittsburgh, 2007 Submitted to the Graduate Faculty of the Department of Physics and Astronomy in partial fulfillment of the requirements for the degree of Doctor of Philosophy University of Pittsburgh 2013 UNIVERSITY OF PITTSBURGH DEPARTMENT OF PHYSICS AND ASTRONOMY...»

«“Heidegger and Marcuse: On Reification and Concrete Philosophy,” The Bloomsbury Companion to Heidegger,” F. Raffoul and E. Nelson, eds., Bloomsbury Press, 2013, pp. 171-176 Andrew Feenberg Introduction Herbert Marcuse (1898–1979) completed his doctorate in 1922 but decided not to pursue the habilitation which would have qualified him for an academic career. Instead, he returned to Berlin where he established an antiquarian bookstore with a partner. When he read Being and Time shortly...»





 
<<  HOME   |    CONTACTS
2016 www.dissertation.xlibx.info - Dissertations, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.