跳到主要內容

臺灣博碩士論文加值系統

(35.172.223.30) 您好!臺灣時間:2021/07/25 11:53
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:吳佳樺
研究生(外文):Chia-Hua Wu
論文名稱:以使用者每日活動資訊進行分類的可行性分析
論文名稱(外文):Feasibility of Routine Based Analysis for User Classification
指導教授:黃仁竑黃仁竑引用關係
指導教授(外文):Ren-Hung Hwang
口試委員:張本杰黃啟富賴槿峰
口試委員(外文):Ben-Jye ChangChi-Fu HuangChin-Feng Lai
口試日期:2013-11-25
學位類別:碩士
校院名稱:國立中正大學
系所名稱:資訊工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2013
畢業學年度:102
語文別:英文
論文頁數:35
中文關鍵詞:使用者分類
外文關鍵詞:routine baseduser classification
相關次數:
  • 被引用被引用:0
  • 點閱點閱:236
  • 評分評分:
  • 下載下載:21
  • 收藏至我的研究室書目清單書目收藏:0
With the increasing number of LBS applications, many users adopt LBS applications to record and track trajectories and daily routines. This recorded information can be analyzed to discover some interesting and meaningful information about the users; however, in the process of gathering data from users, inevitably there will be no signal or phone is off so that the collected data is incomplete or missing. In this thesis, we study the Routine Based Classification (RBC) approach for classifying users into different groups. For comparing two routines, we modify the Smith-Waterman alignment algorithm to increase the accuracy of similarity calculation. Term Frequency-Inverse Document Frequency (TF/IDF) is then used for classifying users based on their routines. In addition, we believe that there shall be some common or similar routines among the users of the same group. Thus, we propose the concept of group routine patterns which are defined as common routines appeared more frequently than other routines within a group of users. Furthermore, user classification can be based on group routine patterns. For performance comparison, we also adopt Support Vector Machines (SVM), a machine learning classification method, to classify users based on their routines. In this thesis, we perform two-stage experiments. In the first stage, we demonstrate that incomplete and missing data cannot be used to classify users accurately. In the second stage experiments, we show that the proposed mechanism without group routine pattern yields competitive performance as compared to SVM and an existing work in the literature. Moreover, the experimental results show that using the group routine pattern concept to classify users yields a higher accuracy.
Chapter 1 : INTRODUCTION 1
Chapter 2 : RELATED WORK 3
Chapter 3 : DATA PREPROCESSING 5
3.1 Data preprocessing 5
3.2 Data preprocessing in our work 7
Chapter 4 : CLASSIFICATION METHODS 12
4.1 The RBC algorithm 12
4.2 The Smith-Waterman algorithm 15
4.3 Support Vector Machines (SVM) 18
4.4 Group pattern 21
Chapter 5 : EXPERIMENTS 23
Chapter 6 : CONCLUSIONS and FUTURE WORK 30
REFERENCES 31

[1] NoniGPSPlot. http://aeguerre.free.fr/Public/PocketPC/NoniGPSPlot/EN/
[2] Endomondo. http://www.endomondo.com/
[3] aTimeLogger. http://www.atimelogger.com/
[4]Q. Li, et al., “Mining user similarity based on location history,” GIS’08, Santa Ana, CA, Nov. 2008, pp. 298-307.
[5]J. J.-C. Ying, E. H.-C. Lu, W.-C. Lee, T.-C. Weng, and V. S. Tseng, “Mining User Similarity from Semantic Trajectories,” ACM SIGSPATIAL International Workshop on Location Based Social Networks (LBSN' 10), San Jose, California, USA, November 2, 2010.
[6]A. Monreale, F. Pinelli, R. Trasarti, and F. Giannotti, “Wherenext: a location predictor on trajectory pattern mining,” 15th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD 2009), 2009, pp. 637-646.
[7]Y. Xiong and H. Lin, “Routine Based Analysis for User Classification and Location Prediction,” UIC/ATC 2012, pp. 96-103.
[8] T. F. Smith and M. S. Waterman, “Identification of Common Molecular Subsequences," Journal of Molecular Biology, Vol. 147(1), 1981, pp. 195–197.
[9]V. VAPNIK, Statistical Learning Theory, Wiley, New York, 1998.
[10]Y. He and X. Wei, “A study of spatial data mining architecture and technology,” 2nd IEEE International Conference on Computer Science and Information Technology (ICCSIT 2009), 2009, pp. 163 – 166
[11] J.-D. Ren, J. Bao, and H.-Y. Huang, “The Research On Spatio-temporal Data Model And Related Data Mining,” the Second International Conference On Machine Learning And Cybernetics, November 2003, pp. 37-40.
[12] C. Zhou, N. Bhatnagar, S. Shekhar, and L. Terveen, “Mining personally important places from gps tracks: a hybrid approach,” the 23rd International Conference on Data Engineering Workshops, ICDE 2007, Istambul, Turkey, April 2007, pp. 517-526.
[13] Y. Chen, K. Jiang, Y. Zheng, C. Li, and N. Yu, “Trajectory Simplification Method for Location-based Social Networking Services,” International Workshop on Location Based Social Network, Seattle, USA, November 2009.
[14] X. Xiao, Y. Zheng, Q. Luo, and X. Xi, “Finding similar users using category-based location history,” the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems, 2010, pp. 442–445.
[15] C.-C. Chang and C.-J., Lin, LIBSVM: a library for support vector machines, 2001.
Available on-line: http://www.csie.ntu.edu.tw/~cjlin/papers/libsvm.pdf.
[16] N. Eagle, A. Pentland, and D. Lazer, ”Inferring social network structure using mobile phone data,” the National Academy of Sciences, Vol. 106(36), 2009, pp. 15274-15278.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top