進階搜尋


   電子論文尚未授權公開,紙本請查館藏目錄
(※如查詢不到或館藏狀況顯示「閉架不公開」,表示該本論文不在書庫,無法取用。)
系統識別號 U0026-2908201914232200
論文名稱(中文) 基於複雜任務結構與消費需求之購物機器人
論文名稱(英文) Shopping Chatbot based on Complex Task Structure and Consumption Need
校院名稱 成功大學
系所名稱(中) 資訊工程學系
系所名稱(英) Institute of Computer Science and Information Engineering
學年度 107
學期 2
出版年 108
研究生(中文) 王振安
研究生(英文) Chen-An Wang
學號 P76054274
學位類別 碩士
語文別 英文
論文頁數 39頁
口試委員 指導教授-盧文祥
口試委員-張景新
口試委員-江振宇
口試委員-楊中平
口試委員-梁勝富
中文關鍵字 CKIP斷詞  自然語言處理  詞性標記  聊天機器人  馬斯洛的需求層次理論  自然語言生成 
英文關鍵字 CKIP  Natural Language Processing  POS Tagging  Chatbot  Natural Language Generation 
學科別分類
中文摘要 近年來,隨著網際網路的成長,人們可以透過網路購物,即便不出門也能購買像要的商品。在台灣,蝦皮、露天、PC-home等都是常見知名的購物網站。一般來說,目前網路購物提供使用者不少便利的功能,有搜尋、熱門關鍵字、商品分類等,都是相當實用。然而仍有很多的功能現在的網路購物仍難以達成,比如當我們外出購物的時候,可以從店員、銷售員等得到一些建議或是相關知識。本論文為了能給消費者一些購物的意見及相關知識而實作購物機器人。
本論文提出了ATCN模型,使用購物文章來做資料的訓練。ATCN 模型是一個基於複雜任務架構,共四層的模型。ATCN模型訓練完後,會生成4個資料庫,我們利用這些資料庫來建構購物機器人。
本論文有兩個實驗,一個是評估任務抓取模組的表現,另一個是評估預測相關任務模組的表現。我們相信購物機器人在未來會更加的便利,這代表著使用者能花更少的時間跟體力在商品的購買上。
英文摘要 Nowadays, many people can buy things online without going out. In Taiwan, auction sites, such as Ruten, PC-HOME, are well-known. Many people buy what he or she wants through online channels. Some functions, like the searching bar, common searching words, products categories, …, is very useful. However, we can’t get the advices and recommends via online shopping. When we go outside to buy things, we often get advice from sellers. However, if we buy things online, we can only search data by ourselves. So, we want to create a shopping chatbot to provide users some advices when they are shopping.
We propose the ATCN model, Activity-Task-Consumption Need model, to train the data using shopping articles. ATCN model is based on complex structure, in which there are four layers. We use four database tables, which produced by ATCN model, to build the shopping chatbot.
We have two experiments, one is to evaluate the performance of task extraction and the other is to evaluate the performance of related task prediction.
We think the shopping chatbot will be more convenient soon. We can use less time and effort in shopping.
論文目次 摘要 III
Abstract IV
致謝 V
Figure Index IX
Table Index X
Chapter 1 Introduction 1
1.1 Background 1
1.2 Motivation 2
1.3 Method 3
1.4 Contribution 4
1.5 Organization of this Dissertation 5
Chapter 2 Related Work 6
2.1 Complex Task Structure 6
2.2 Task Extraction & Related Task Prediction 8
2.3 Advertisement Recommendation 9
Chapter 3 Method 10
3.1 System Framework 10
3.1.1 Activity Extraction 11
3.1.2 Task Extraction 11
3.1.3 Related Task Prediction 11
3.1.4 Consumption Need Extraction 11
3.2 Activity Extraction 12
3.2.1 Text Processing 12
3.2.2 Activity Attributes 12
3.2.3 Coverage of Attributes 13
3.3 Task Extraction 14
3.3.1 Verb Selection 14
3.3.2 Task Attributes 15
3.4 Related Task Prediction 16
3.4.1 Dice Coefficient 18
3.4.2 Related Task Prediction Algorithm 18
3.4.3 Related Task Prediction Result 19
3.5 Consumption Need Extraction 22
3.5.1 Maslow's Hierarchy of Needs 22
3.5.2 Consumption Need 22
3.5.3 Consumption Need Extraction 23
3.6 Shopping Chatbot 24
3.6.1 Chat &Chat Management 24
3.6.2 Activity Extraction 24
3.6.3 Task Searching 24
3.6.4 Related Task Searching 24
3.6.5 Natural Language Generation Template 25
Chapter 4 Experiments 32
4.1 Dataset 32
4.2 Evaluation 32
4.2.1 Precision 33
4.2.2 Normalized discounted cumulative gain (NDCG) 33
4.3 Performance of Task Extraction 33
4.3.1 Experiment setup 33
4.3.2 Experiment result 34
4.4 Performance of Related Task Prediction 35
4.4.1 Experiment setup 35
4.4.2 Experiment result 35
Chapter 5 Conclusions 37
5.1 Conclusion 37
5.2 Future Work 37
Reference 38
參考文獻 [1] G.Consumer andI.Survey, “It ’ s time for a consumer-centred metric : introducing ‘ return on experience ,’” 2019.
[2] Jhih-Sheng Fan, “Chatbot Application: Event driven Task-Oriented Store Recommendation,” 2017.
[3] Sheng-Tang Chuang, “Product Recommendation BOT Based On Related Task Structure Using News Event and Yahoo Answer,” 2018.
[4] Ting-Xuan Wang, “Constructing Complex Search Task with Subtasks to Improve Web Search and Sponsored Search Advertising,” 2015.
[5] W.-Y.Ma andK.-J.Chen, “Introduction to CKIP Chinese Word Segmentation System for the First International Chinese Word Segmentation Bakeoff,” 2003.
[6] S.Mcleod, “Maslow’s Hierarchy of Needs simplypsychology.org/maslow.html,” 2018.
[7] M.Revathy andM. L.Madhavu, “Efficient author community generation on Nlp based relevance feature detection,” Proc. IEEE Int. Conf. Circuit, Power Comput. Technol. ICCPCT 2017, 2017.
[8] K.-J.Chen andM.-H.Bai, “Unknown Word Detection for Chinese by a Corpus-based Learning Method,” 1998.
[9] M.Likhar andS. L.Kasar, “Sentiment analysis using sentence minimization with natural language generation (NLG),” in Proceedings - 1st International Conference on Intelligent Systems and Information Management, ICISIM 2017, 2017.
[10] Y.-F.Tsai andK.-J.Chen, “Context-rule Model for Pos Tagging,” 2003.
[11] Y.-F.Tsai andK.-J.Chen, “Reliable and Cost-Effective PoS-Tagging,” 2003.
[12] S.Prasomphan, “Improvement of chatbot in trading system for SMEs by using deep neural network,” in 2019 IEEE 4th International Conference on Cloud Computing and Big Data Analytics, ICCCBDA 2019, 2019.
[13] M.Diaz-Mora, M.Diaz-Rodriguez, andM.Jimeno, “Definition and validation of an energy savings process for computers based on user behaviors and profiles,” in International Conference on Wireless and Mobile Computing, Networking and Communications, 2017.
[14] W.-Y. M.Keh-Jiann Chen, “Unknown Word Extraction for Chinese Documents,” 2002.
[15] K. S.Vu Tran, Minh Le Nguyen, “Building Legal Case Retrieval Systems with Lexical Matching and Summarization using A Pre-Trained Phrase Scoring Model,” ICAIL ’19 Proc. Seventeenth Int. Conf. Artif. Intell. Law, pp. 275–282, 2019.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2021-07-25起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2021-07-25起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw