跳到主要內容

臺灣博碩士論文加值系統

(18.97.14.86) 您好!臺灣時間:2025/02/12 21:30
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:李俊民
研究生(外文):Chun-Ming Lee
論文名稱:具有詞彙二階關係向量式資訊檢索技術
論文名稱(外文):Vector Information Retrieval Technique with Word Bigram Relation Model
指導教授:楊燕珠楊燕珠引用關係
指導教授(外文):Yen-Ju Yang
學位類別:碩士
校院名稱:大同大學
系所名稱:資訊經營學系(所)
學門:商業及管理學門
學類:一般商業學類
論文種類:學術論文
論文出版年:2004
畢業學年度:93
語文別:英文
論文頁數:83
中文關鍵詞:資訊檢索向量式詞彙二階關係客戶關係管理
外文關鍵詞:customer relationship managementInformation retrievalvector modelword bigram relation model
相關次數:
  • 被引用被引用:0
  • 點閱點閱:210
  • 評分評分:
  • 下載下載:15
  • 收藏至我的研究室書目清單書目收藏:2
電話客服中心是目前企業能快速與客戶建立起良好的溝通橋樑的主要窗口,也是企業要建立客戶關係管理 (customer relationship management;CRM) 所需重要資訊的主要來源。客服人員所處理的問題大多是類似的工作,透過自動代理人程式 (agents) 可提升客服人員的服務品質及工作效率。因此為了自動代理人程式的資訊處理能力,針對一般常用的向量式檢索技術 (vector model) 進行檢索效率提升的改善研究。本研究將詞彙前後文二階關係應用在向量式檢索技術,以三種詞彙相連關係:交互訊息(Mutual Information),關聯基準(Association Norm) 以及條件機率 (Conditional Probability),加強詞彙前後文的限制,提升相似度比對。一般客服處理的問題可以整理成企業相關的FAQ,故我們首先以中華電信網站的FAQ資料為實驗對象,分別進行內部測試(inside test)以調適(tune) 出最佳參數作為外部測試(outside test)的參考,並以精確率(precision rate)、召回率(recall rate)及11等級召回率的相對精確率,評估檢索效能。研究獲得結果是,內部測試的交互訊息法在標準召回率水準下平均精確率提高41.9 %,外部測試的關聯基準法提高8.14 %,確實可以作為客服代理人的一環。
“Telephone Center” has become enterprises’ major service window and important information source for customer relationship management (CRM). Customer service operators are dealing with similar questions for most of time, and by agents could improve their service quality and efficiency. if automatic agents could apply to replace part of their complicate works, they could deal with more specialized problems and reduce manpower to improve process efficiency. Therefore, for agents’ capability to undertake information, this thesis researches the improvement of generally applied vector model’s retrieval efficiency. This research applies word bigram relation model on vector model by three methods, namely, Mutual Information, Association Norm and Conditional Probability, to strengthen words constriction and increase similarity comparison. General customer service questions could compile as enterprise’s FAQ, therefore our experiment object is the FAQ in Chunghwa Telecom’s website, and undertake inside test to tune the best parameter as outside test’s reference, and evaluate retrieval performance by precision rate, recall rate and recall at 11 levels’ corresponding precision rate. The research results is that Mutual Information’s average precision rates under standard recall level increases 41.9% in inside test, and Association Norm increases 8.14% in outside test, so it is indeed could be a segment of customer service agent.
TABLE OF CONTENTS

CHINESE ABSTRACT ii
ENGLISH ABSTRACT iii
ACKNOWLEDGEMENTS v
TABLE OF CONTENTS vi
LIST OF FIGURES ix
LIST OF TABLES x

CHAPTER I INTRODUCTION 1
1.1 Background 1
1.2 Research Motivation 2
1.3 Research Purpose 3
1.4 Research Outcome 4
1.5 Other Chapters Description 5

CHAPTER II LITERATURE REVIEW 7
2.1 Information Retrieval 7
2.2 Retrieval Model 9
2.3 The Approach to Evaluate Retrieval Effect 18

CHAPTER III RESEARCH APPROACH 26
3.1 Research Framework 26
3.2 Word bigram relation model 26
3.3 Measure precision rate by standard recall at 11 levels 32
3.4 Evaluation of Precision rate and Recall rate 33

CHAPTER IV EXPERIMENT RESULT AND ANALYSIS 35
4.1 Purpose and Content 35
4.1.1 Purpose 35
4.1.2 Content 35
4.2 Select Experiment Data 36
4.3 Establishment of Keyword Collection 36
4.4 Establishment of Weights Calculation and On-line Model 38
4.4.1 Weights Calculation Model 38
4.4.2 Calculation of Documents or Queries’ Length 40
4.4.3 Experiment Environment 42
4.4.4 On-line Model 43
4.5 Combination of Experiment Parameters 47
4.6 Getting Experiment Data 48
4.6.1 Inside test 48
4.6.2 Outside Test 67
4.7 Analysis of Experiment Data and Discussion of Relevant Questions 71

CHAPTER V CONCLUSIONS AND SUGGESTIONS 78
REFERENCES 80
VITA 83

LIST OF FIGURES
Fig. 2-1: General Information Retrieval System’s Framework 8
Fig. 2-2: Relation of Recall and Precision Rates 21
Fig. 2-3: Graph of precision rate under every recall level 22
Fig. 2-4: Graph of Recall and Precision 25
Fig.3-1: Research Framework 27
Fig.4-1: Establishment of Keyword Collection 39
Fig.4-2: Weight Calculation Process 41
Fig. 4-3:Documents or queries’ length calculation process 44
Fig. 4-4: On-line model’s process 46
Fig. 4-5: Four methods’ corresponding average precision rate under single term importance parameter C 66
Fig. 4-6: Four methods’ average precision rate under standard recall at 11 levels in outside test 70

LIST OF TABLES
Table 4-1 Corresponded precision rates under standard recall at 11 levels 51
Table 4-2 20 queries’ corresponded precision rates under standard recall level=0% 52
Table 4-3 Calculation of four methods’ average precision rate under standard recall at 11 levels 53
Table 4-4 Four methods’ average precision rate under single term parameter C=1 54
Table 4-5:Four methods’ average precision rate under single term parameter C=0.9 55
Table 4-6:Four methods’ average precision rate under single term parameter C=0.8 57
Table 4-7:Four methods’ average precision rate under single term parameter C=0.7 58
Table 4-8: Four methods’ average precision rate under single term parameter C=0.6 59
Table 4-9: Four methods’ average precision rate under single term parameter C=0.5 61
Table 4-10: Four methods’ total average precision rate under single term parameter C=1 63
Table 4-11: Four methods’ total average precision rate under single term parameter C=0.9 63
Table 4-12: Four methods’ total average precision rate under single term parameter C=0.8 64
Table 4-13: Four methods’ total average precision rate under single term parameter C=0.7 64
Table 4-14: Four methods’ total average precision rate under single term parameter C=0.6 65
Table 4-15: Four methods’ total average precision rate under single term parameter C=0.5 65
Table 4-16: Four methods’ average precision rate under outside test 69
REFERENCES

[1] Salton, G., “The SMART Retrieval System—Experiments in automatic Document
Processing.”, Prentice Hall Inc., Englewowod Cliffs, NJ, 1971.

[2] Salton, G., and Lesk, M. E., “Computer evaluation of indexing and text
processing.”, Journal of the ACM, 15(1):8-36, January 1968.

[3] Salton, G. and McGill, M. J., “Introduction to Modern Information
Retrieval.”, McGraw -Hill Book Co., New York, 1983.

[4] Salton, G. and Buckley, C., “Term-weighting approaches in automatic
retrieval.”, Information Processing & Managerment, 24(5):513-523, 1988.

[5] Korfhage, Robert., “Information Storage & Retrieval.”, John Willy & Sons,
Inc., 1997.

[6] Shaw Jr, W. M., Burgin, R., and Howell, P., “Performance standards and
evaluations in IR test collections: Cluster-based retrieval models.”.
Information Processing & Managerment, 33(1):1-14, 1997.

[7] Shaw Jr, W. M., Burgin, R., and Howell, P., “Performance standards and
evaluations in IR test collections: Vector-space and other retrieval
models.”, Information Processing & Managerment, 33(1):15-36, 1997.

[8] Van Risbergen, C. J., “Information Retrieval.”, Butterworths, 1979.

[9] Mizzaro, S., “Relevance: The whole history.”, Journal of the American
Society for Information Science, 48(9):810-832, 1997.

[10] Church, K. W., “A Stochastic Parts Program and Noun Phrase Parser for
Unrestricted Text.” Proceedings of Applied Natural Language Processing,
Austin, Texas, 1988.

[11] Church, K. W. and Hanks, P., “Word Association Norms, Mutual Information,
and Lexicography.”, in Proceedings of 27th Annual Meeting of the
Association for Computational Linguistics, pp. 76-83, 1989.

[12] Devore, Jay L., “Probability and Statistics for engineering and the
sciences.”, Duxbury Inc., 2000.

[13] Han, Jiawei. and Kamber Micheline., “Data Mining: Concept and
Techniques.”, Morgan Kaufmann Publishers, 2001

[14] Xu, R., and Yeung, D., “Experiments on the use of corpus-based Word BI-
gram in Chinese Word Segmentation.”, Systems, Man, and Cybernetics., IEEE
International Conference, pages: 4222-4227 vol.5, 1998.

[15] Sproat, R., Shin, C., Gale, W. A., and Chang, N., “A stochastic finite-
state word segmentation algorithm for Chinese.”, Computational
liguistics, 22(3), 377-404, 1996.

[16] Wang Sheng-Zhong, Hong Wen-Bin, “Research of Syntax Chinese Word
Segmentaiton”, Master Thesis, Department of CSIE, Tamkang University,
June 1994.

[17] Chinese Knowledge Information Processing Group, Analysis of Chinese Words.
Institute of Information Science, Academia Sinica. June 1993.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top