跳到主要內容

臺灣博碩士論文加值系統

(3.236.84.188) 您好!臺灣時間:2021/08/05 00:44
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

我願授權國圖
: 
twitterline
研究生:林育賢
研究生(外文):LIN, YU-HSIEN
論文名稱:基於使用者行動裝置訊號之智慧生活熱區推薦系統
論文名稱(外文):Intelligent Living Hotspot Recommendation System for User Mobile Device Signals
指導教授:周智倫
指導教授(外文):CHOU, CHIH-LUN
口試委員:鄒耀東洪國鈞周智倫
口試委員(外文):TSOU, YAO-TUNGHORNG, GWO-JIUNCHOU, CHIH-LUN
口試日期:2020-07-29
學位類別:碩士
校院名稱:銘傳大學
系所名稱:電腦與通訊工程學系碩士班
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:2020
畢業學年度:108
語文別:中文
論文頁數:64
中文關鍵詞:人流資料資料分析推薦系統智慧校園
外文關鍵詞:People flowData analysisRecommender SystemSmart Campus
相關次數:
  • 被引用被引用:1
  • 點閱點閱:69
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
隨著科技快速的發展,智慧城市是近年來各世界逐漸邁向之目標,更是一個包含大數據等高度技術結合之綜合體,然而在城市中;往往在特定的時間會出現壅擠的人潮,當人潮集中在某些區域時會造成等待時間的延長而提高時間的成本,因此,藉由推薦系統因應產生,規劃使用者要前往的地點,使用等待時間的長短作為前往先後的順序,在推薦系統中除了常見的協同過濾演算法之外,RNN、GRU、LSTM等機器學習神經網路的加入使的推薦系統更加的完善。
本研究以銘傳大學桃園校區為實驗場所,使用各區域的等候時間,加以推薦使用者前往各區域的順序,以便減少等待的時間,並提出智慧生活熱區推薦系統架構,透過學生線上問券調查的方式對校內使用空間分布情形與時空行為模式進行分析與推薦,運用人群分析定義空間特徵模式,更藉由地區人數與停留時間兩大變數,藉由遞迴神經網路的使用,python作為視覺化紀錄,最後使用這些運算結果來進行查詢,推薦其等待時間跟人流分布,加以舒緩並改善人潮壅塞的時間。最後,本研究希望融合數據分析與實際規劃之方式,藉此分析與優化校園的空間使用現況並擴展其功能,創建一個基於數據的智慧服務新校園。
With the rapid development of science and technology, smart city is the goal that the world is gradually moving towards in recent years. In the city often appear indicates crowded at a specific time, when the crowds concentrated in certain areas caused by the extension of waiting time that increase the cost of time, so use recommendation system to relieve the crowds and reduce the time cost, in addition to the traditional collaborative filtering algorithms, RNN, GRU, LSTM more improved the accuracy and efficiency of learning.
This study university taoyuan campus in place, the experiment using the regional and recommend waiting time of the user to the order of each area, and propose the intelligent life hotspots recommendation system, use the questionnaire to analyze and recommend the school space, RNN was calculated by using the crowd model, the number of people in the area and the residence time, a python as a visual record, finally use the result for the query, In order to better understand the factors of campus crowd distribution and improve the accuracy of the recommended time, the weather, campus calendar were also included in the analysis, so as to better understand the causes of the crowd distribution. Finally, create a data-based smart service new campus.
摘要 i
ABSTRACT ii
致謝 iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 緒論 1
1.1. 研究動機與目的 1
1.2. 研究目的 1
第二章 文獻探討 1
2.1. 相關背景知識 3
2.2. 人流資料取得 10
2.3. 推薦系統 11
第三章 研究方法 15
3.1. 收集資料 16
3.2. 資料前處理 17
3.3. 熱區劃分 19
3.4. 等待時間預估 20
3.5. 推薦給使用者 23
第四章 實驗結果 24
4.1. 收集資料與分析工具 25
4.2. 資料前處理 28
4.2.1. 資料編碼 30
4.3. 人流分析 31
4.3.1. 以地點為主的人流分析 31
4.3.2. 以時間為主的人流分析 41
4.4. 等待時間預估 43
4.4.1. 資料集實驗 43
4.4.2. 機器學習 48
4.5. 推薦給使用者 50
第五章 結論 52
參考文獻 53
[1] S. D. Aäron van den Oord, Benjamin Schrauwen, "Deep content-based music recommendation," (in en), Advances in Neural Information Processing Systems 26 (NIPS 2013), 2013.
[2] A. K. Balázs Hidasi, Linas Baltrunas, Domonkos Tikk, "Session-based Recommendations with Recurrent Neural Networks," (in en), arXiv:1511.06939 2015.
[3] R. Burke, "Knowledge-based recommender systems," (in en), 2000.
[4] R. Burke, "Hybrid Recommender Systems: Survey and Experiments," (in en), User Modeling and User-Adapted Interaction, vol. 12, no. 4, pp. 331-370, 2002.
[5] CS231n, "Convolutional Neural Networks for Visual Recognition," (in en), https://is.gd/ICabpY, 2019.
[6] D. Feng, "Utility-based Recommender S ystems Using Implicit U tility and Genetic Algorithm," (in en), International Conference on Mechatronics, Electronic, Industrial and Control Engineering (MEIC 2015), 2015.
[7] L. R. Francesco Ricci, Bracha Shapira, "Introduction to Recommender Systems Handbook," (in en), springer, 2010.
[8] S. Jain, A. Grover, P. S. Thakur, and S. K. Choudhary, "Trends, problems and solutions of recommender system," presented at the International Conference on Computing, Communication & Automation, 2015.
[9] R. H. J. McAuley, "VBPR: Visual Bayesian PersonalizedRanking from Implicit Feedback," (in en), AAAI'16, arXiv:1510.01784, 2015.
[10] T. B. D. B. A. McCallum, "Ask the GRU: Multi-Task Learning forDeep Text Recommendations," (in en), arXiv,arXiv:1609.02116, 2016.
[11] M. Nguyen, "Illustrated Guide to LSTM’s and GRU’s: A step by step explanation," (in en), https://towardsdatascience.com/illustrated-guide-to-lstms-and-gru-s-a-step-by-step-explanation-44e9eb85bf21, 2018.
[12] C. Olah, "Understanding LSTM Networks," (in en), https://colah.github.io/posts/2015-08-Understanding-LSTMs/, 2015.
[13] N. Rahman, "What is the benefit of using average pooling rather than max pooling?," (in en), https://www.quora.com/What-is-the-benefit-of-using-average-pooling-rather-than-max-pooling, 2017.
[14] L. Y. Shuai Zhang, Aixin Sun,Yi Tay, "Deep Learning Based Recommender System: A Survey and New Perspectives," (in en), ACM Computing Surveys (CSUR), vol. Vol. 1, No. 1, Article 1. , 2018.
[15] R. v. M. M. v. Someren, "Using Content-Based Filtering for Recommendation," (in en), 2000.
[16] S. Wang, Y. Wang, J. Tang, K. Shu, S. Ranganath, and H. Liu, "What Your Images Reveal," presented at the Proceedings of the 26th International Conference on World Wide Web - WWW '17, 2017.
[17] Wikimedia, "Galileo (satellite navigation)," (in en), https://en.wikipedia.org/wiki/Galileo_(satellite_navigation), 2020.
[18] N. D. Yichen Wang, Rakshit Trivedi, Le Song, "Coevolutionary Latent Feature Processes forContinuous-Time User-Item Interactions," (in en), Neural Information Processing Systems 29 (NIPS 2016), 2016.
[19] L. Zheng, V. Noroozi, and P. S. Yu, "Joint Deep Modeling of Users and Items Using Reviews for Recommendation," presented at the Proceedings of the Tenth ACM International Conference on Web Search and Data Mining - WSDM '17, 2017.
[20] 李宏毅, "Structured Learning: Recurrent Neural Network," (in en), http://speech.ee.ntu.edu.tw/~tlkagk/courses_ML16.html, 2016.
[21] M. Copeland, "人工智慧、機器學習與深度學習間有什麼區別?," (in ch), https://blogs.nvidia.com.tw/2016/07/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/, 2016.
[22] eCloudvalley, "三分鐘了解機器學習的四個學習方式," (in ch), https://www.ecloudvalley.com/zh-hant/machine-learning/, 2019.
[23] mikechenx, "處理影像的利器 -- 卷積神經網路(Convolutional Neural Network)," (in ch), https://ithelp.ithome.com.tw/articles/10191820, 2017.
[24] NTUST, "無線網路Wi-Fi," (in ch), https://is.gd/wLeHS1, 2004.
[25] read01, "手機是如何通過wifi來定位的?怎麼防止手機通過wifi被定位?," (in ch), https://read01.com/Bz6Jdg.html, 2016.
[26] 劉.-C. L. 游. P.-H. Yu, "GIS於住宅竊盜犯罪熱區空間分析應用之研究-以台北市大安區為例," (in ch), Journal of Criminology, 14(1), 2011 P.97-138, 2011.
[27] 打不死的小强, "循環神經網路– Recurrent Neural Network | RNN," (in ch), https://easyai.tech/ai-definition/rnn/, 2019.
[28] 曲建仲, "機器是如何學習與進步?人工智慧的核心技術與未來," (in ch), http://scimonth.blogspot.com/2018/03/blog-post_56.html, 2019.
[29] 李欣宜, "Beacon微定位商機崛起," (in ch), https://www.bnext.com.tw/article/34960/BN-ARTICLE-34960, 2015.
[30] 溫在弘,劉擇昌,林民浩, "犯罪地圖繪製與熱區分析方法及其應用:1998-2007年台北市住宅竊盜例犯罪為例," (in ch), Journal of Geographical Research No.52, May 2010 P.43-63, 2010.
[31] 董冠麟, "台灣大選/政黨票票倉到底在哪裡?空間自相關分析告訴你," (in ch), https://www.geog-daily.org/geogforum/14, 2016.
[32] 魯亭儀, "用於求職網站的混合式回饋推薦系統," (in ch), 2017.
[33] 賴桑, "【自學AI#6】像人腦般靈活?最受歡迎的應用:RNN," (in ch), https://makerpro.cc/2019/07/self-learning-series6-of-ai-rnn/, 2019.
電子全文 電子全文(網際網路公開日期:20250826)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top