簡易檢索 / 詳目顯示

回結果列表
研究生： 陳柏叡
Chen, Po-Rui
論文名稱： 基於TF-IDF及長短期記憶自編碼之智慧藥局問答對話與多媒體回饋系統
Intelligent Pharmacy Question Answering and Multimedia Feedback System based on TF-IDF and LSTM Auto-encoder
指導教授： 王駿發
Wang, Jhing-Fa
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 電機工程學系
Department of Electrical Engineering
論文出版年： 2018
畢業學年度： 106
語文別： 英文
論文頁數： 54
中文關鍵詞： 問答對話 、詞嵌入 、LSTM 、Auto-encoder 、TF-IDF 、句子相似度
外文關鍵詞： Question Answering, Word Embedding, LSTM, Auto-encoder, TF-IDF, Sentence Similarity
相關次數： 點閱：74  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 近幾年人工智慧發展迅速，機器人也成為大家開發研究的對象，在各大展場活動或是公共場所可見機器人的身影。本篇論文以機器人為平台提出一個智慧藥局系統，並分成兩個部分。一個部分是問答對話，問答對話包含商品諮詢與流感問答，而整個問答對話又分五個步驟，1.斷詞 2.詞嵌入(Word Embedding) 3.TF-IDF 4.LSTM Auto-encoder 5.結合步驟三和四再做相似度運算。先將文本斷詞，引入詞嵌入模型，詞嵌入模型能夠將具有相同性質的詞聚集一起，而他們的詞向量會非常相似，再利用TF-IDF計算每個詞的權重，這能顯現出詞對句子的重要性。另外，我們使用LSTM Auto-encoder的方式建立學習機制，並將藥局文本輸入至模型訓練，Auto-encoder能自己將輸入語句做編碼並取得其特徵向量，再經由解碼將特徵向量重建成輸入語句，輸出結果再和輸入語句做相似度的運算，最終再回應使用者。以機器學習的方式跟原本只用TF-IDF的方式來比具有學習性，本篇系統能夠學習字和字之間的關係也能夠學習語句的架構。另一個部分是多媒體回饋功能，在系統中有許多與藥局相關的功能，透過這些回饋能了解現在藥局的營運狀況或是相關活動。

    In recent years, artificial intelligence has developed rapidly, and robots have become the object of research and development. The robots can be seen in various exhibition venues or public places. This paper proposes a smart pharmacy system based on robots and is divided into two parts. One part is the question answering (QA). The QA includes product consultation and flu question and answer. The whole QA system is divided into five steps. 1. Word Segmentation 2. Word Embedding 3. TF-IDF 4. LSTM Auto-encoder 5. Combine steps 3 and 4 to calculate the sentence similarity. The corpus will first be word segmentation, introducing a word embedding model, which can aggregate words of the same nature, and their word vectors will be very similar. Then use TF-IDF to calculate the weight of each word, which can show the importance of words to sentences. And we use LSTM Auto-encoder to establish the learning model and input the pharmacy corpus into the model to train. Auto-encoder can encode the input statement and obtain its feature vector, and then reconstruct the sentence by decoding the feature vector. The output result is compared with the input statement for similarity calculation, and finally responses the answer to user. In machine learning, it learns much more learn than the original method of using only TF-IDF. This system can learn the relationship between words and words and also learn the temporal information. The other part is the multimedia feedback function, which has many functions related to the pharmacy in the system. Through these feedbacks, you can understand the current operation status of the pharmacy or related activities.

    中文摘要 I Abstract II 誌謝 IV Contents V Table List VII Figure List VIII Chapter 1 Introduction 1 1.1 Background 1 1.2 Motivation 2 1.3 Thesis Objective 3 1.4 Thesis Organization 3 Chapter 2 Related Works 4 2.1 The Survey of System Platform 4 2.2 The Survey of Intelligent Pharmacy 6 2.3 Overview of Question Answering 7 2.4 The Survey of LSTM Auto-encoder 10 2.5 The Survey of Sentences Similarity 12 Chapter 3 Question Answering and multimedia feedback system on Zenbo 14 3.1 System Overview 14 3.2 Sentence Similarity for Question Answering 16 3.2.1 Question Answering System Overview 16 3.2.2 Word Segmentation 18 3.2.3 Word Embedding 19 3.2.3.1 Prediction-based 20 3.2.4 TF-IDF 23 3.2.5 LSTM Auto-encoder 27 3.2.5.1 Auto-encoder 28 3.2.5.2 LSTM 30 3.2.5.3 Auto-encoder – LSTM 34 3.2.6 Combination 36 3.3 Integrated Multimedia and Feedback System 37 3.3.1 User Interface of the System 38 3.3.2 TTS Feedback 39 3.3.3 Multimedia Feedback 39 3.3.3.1 Play Advertisement 39 3.3.3.2 Promotional Activity 40 3.3.3.3 Introduction of Pharmacy 41 3.3.4 Medicine identification Feedback 42 3.3.5 Member Recognition Feedback 42 3.4 Intelligent Pharmacy System on other platforms 43 Chapter 4 Experimental Result 46 4.1 Experiment for QA System in Pharmacy 46 4.1.1 Corpus 46 4.1.2 Evaluation methods 46 4.1.3 Experimental Result 47 4.2 Evaluation for the Proposed System 49 Chapter 5 Conclusion and Future Works 51 References 52

    References
    [1] (2016). AI在人類生活八大領域的廣泛應用. Available: https://read01.com/zh-tw/D75j43.html#.Wj4DJN-WaUl
    [2] S. Amendola, L. Bianchi, and G. Marrocco, "Movement Detection of Human Body Segments: Passive radio-frequency identification and machine-learning technologies," IEEE Antennas and Propagation Magazine, vol. 57, no. 3, pp. 23-37, 2015.
    [3] F. E. Fernandes, Y. Guanci, H. M. Do, and W. Sheng, "Detection of privacy-sensitive situations for social robots in smart homes," in 2016 IEEE International Conference on Automation Science and Engineering (CASE), 2016, pp. 727-732.
    [4] (2018). Zenbo - 智慧居家好夥伴 - ASUS Zenbo. Available: https://zenbo.asus.com/tw/product/overview/
    [5] A. Abraham, "Rule‐Based expert systems," Handbook of measuring system design, 2005.
    [6] S. Aiym and A. Lyazzat, "Using intelligent systems in pharmacy," in 2016 IEEE 10th International Conference on Application of Information and Communication Technologies (AICT), 2016, pp. 1-5.
    [7] Knowledge Base. Available: http://wiki.mbalib.com/zh-tw/%E7%9F%A5%E8%AF%86%E5%BA%93
    [8] S. C. Shieh, C. C. Lin, T. F. Yang, and G. H. Tu, "Using RFID technology on clinic’s pharmacy operation management and development of intelligent medicine dispensing cabinet," in 2008 IEEE International Conference on Industrial Engineering and Engineering Management, 2008, pp. 2006-2009.
    [9] D. Jayaprakash and C. Fancy, "Using python inbuild modules for effective pharmacy billing application," in 2017 International Conference on IoT and Application (ICIOT), 2017, pp. 1-6.
    [10] R. Morante, M. Krallinger, A. Valencia, and W. Daelemans, "Machine Reading of Biomedical Texts about Alzheimer’s Disease 1," 2012.
    [11] S. Zhao, Y. Zheng, C. Zhu, T. Zhao, and S. Li, "Semantic computation in geography question answering," in 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD), 2016, pp. 1572-1576.
    [12] A. Abdi, N. Idris, and Z. Ahmad, "QAPD: an ontology-based question answering system in the physics domain," Soft Computing, vol. 22, no. 1, pp. 213-230, 2018.
    [13] A. Ansari, M. Maknojia, and A. Shaikh, "Intelligent question answering system based on Artificial Neural Network," in 2016 IEEE International Conference on Engineering and Technology (ICETECH), 2016, pp. 758-763.
    [14] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural computation, vol. 9, no. 8, pp. 1735-1780, 1997.
    [15] I. Sutskever, O. Vinyals, and Q. V. Le, "Sequence to sequence learning with neural networks," in Advances in neural information processing systems, 2014, pp. 3104-3112.
    [16] J. Li, M.-T. Luong, and D. Jurafsky, "A hierarchical neural autoencoder for paragraphs and documents," arXiv preprint arXiv:1506.01057, 2015.
    [17] O. Vinyals, Ł. Kaiser, T. Koo, S. Petrov, I. Sutskever, and G. Hinton, "Grammar as a foreign language," in Advances in Neural Information Processing Systems, 2015, pp. 2773-2781.
    [18] S. Bengio, O. Vinyals, N. Jaitly, and N. Shazeer, "Scheduled sampling for sequence prediction with recurrent neural networks," in Advances in Neural Information Processing Systems, 2015, pp. 1171-1179.
    [19] N. Srivastava, E. Mansimov, and R. Salakhudinov, "Unsupervised learning of video representations using lstms," in International conference on machine learning, 2015, pp. 843-852.
    [20] E. M. Nitish Srivastava, Ruslan Salakhutdinov, "Unsupervised Learning of Video Representations using LSTMs," 2016.
    [21] T. Wong and Z. Luo, "Recurrent Auto-Encoder Model for Multidimensional Time Series Representation," 2018.
    [22] P. Malhotra, V. TV, L. Vig, P. Agarwal, and G. Shroff, "TimeNet: Pre-trained deep recurrent neural network for time series classification," arXiv preprint arXiv:1706.08838, 2017.
    [23] N. Balasubramanian, J. Allan, and W. B. Croft, "A comparison of sentence retrieval techniques," in Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval, 2007, pp. 813-814: ACM.
    [24] D. Metzler, Y. Bernstein, W. B. Croft, A. Moffat, and J. Zobel, "Similarity measures for tracking information flow," in Proceedings of the 14th ACM international conference on Information and knowledge management, 2005, pp. 517-524: ACM.
    [25] R. Mihalcea, C. Corley, and C. Strapparava, "Corpus-based and knowledge-based measures of text semantic similarity," in AAAI, 2006, vol. 6, pp. 775-780.
    [26] B. H. Su, T. W. Kuan, S. P. Tseng, J. F. Wang, and P. H. Su, "Improved TF-IDF weight method based on sentence similarity for spoken dialogue system," in 2016 International Conference on Orange Technologies (ICOT), 2016, pp. 36-39.
    [27] Q. Le and T. Mikolov, "Distributed representations of sentences and documents," in International Conference on Machine Learning, 2014, pp. 1188-1196.
    [28] M. Kusner, Y. Sun, N. Kolkin, and K. Weinberger, "From word embeddings to document distances," in International Conference on Machine Learning, 2015, pp. 957-966.
    [29] X. Xu and F. Ye, "Sentences similarity analysis based on word embedding and syntax analysis," in 2017 IEEE 17th International Conference on Communication Technology (ICCT), 2017, pp. 1896-1900.
    [30] (2017). Google Assistant’s speech recognition API is now open to all developers. Available: https://www.androidauthority.com/google-assistant-api-now-open-developers-765038/
    [31] S. M. Liu and J.-H. Chen, "A multi-label classification based approach for sentiment classification," Expert Systems with Applications, vol. 42, no. 3, pp. 1083-1093, 2015.
    [32] Y. Goldberg, "Neural network methods for natural language processing," Synthesis Lectures on Human Language Technologies, vol. 10, no. 1, pp. 1-309, 2017.
    [33] T. Mikolov, I. Sutskever, K. Chen, G. S. Corrado, and J. Dean, "Distributed representations of words and phrases and their compositionality," in Advances in neural information processing systems, 2013, pp. 3111-3119.
    [34] (2018). Vector Representations of Words. Available: https://www.tensorflow.org/tutorials/representation/word2vec
    [35] S. Rifai, P. Vincent, X. Muller, X. Glorot, and Y. Bengio, "Contractive auto-encoders: Explicit invariance during feature extraction," in Proceedings of the 28th International Conference on International Conference on Machine Learning, 2011, pp. 833-840: Omnipress.
    [36] G. E. Hinton and R. R. Salakhutdinov, "Reducing the dimensionality of data with neural networks," science, vol. 313, no. 5786, pp. 504-507, 2006.
    [37] (2015). Understanding LSTM Networks. Available: http://colah.github.io/posts/2015-08-Understanding-LSTMs/

    無法下載圖示 校內：2023-08-31公開
    校外：不公開
    電子論文尚未授權公開，紙本請查館藏目錄
    QR CODE