社群問答網站是人們可以相互分享知識內容的平台。隨著行動網路的日益普遍，使得社群問答網站能夠迅速發展。由於使用者數量的增加，也伴隨著新問題數量的增加。結果是這些社群問答網站都面臨一個常見的問題，平台上存在許多問題未獲得任何答案。出現此問題的原因之一是具有潛力的答題者不太可能找到他們可以回答的問題。解決問題的一種方法是自動化識別問題並且將其路由至合適的答題者。因此在本文中，我們提出了一種有效的深度學習模型，其組成架構基於卷積神經網路而構成，可以有效地預測新問題與答題者的相關性，並且利用答題者的個人特徵(例如：專業性、活躍性特徵)來提高模型預測的準確性。最後，我們對從Stack Overflow提取的真實資料集進行實驗，Stack Overflow是一個專注於程式撰寫領域的社群問答網站，然後實驗顯示我們的方法在各種指標上均優於其他方法。

Community of Question Answering (CQA) sites are platforms where people share knowledge with each other. As the number of users is growing rapidly in these platforms, the quantity of new questions increases drastically. As a result, these CQA sites encounter a common problem that many questions are left with no answers. One of the reasons for this problem is that a potential answerer is less likely to find questions that they can answer. One solution for this problem is to automatically identify and route the questions to suitable answerers. In this paper, we propose an effective method using a Convolutional Neural Network (CNN) architecture to predict the relevance between a question and potential answerers to decide the most relevant answerer for the question. Furthermore, we utilize answerer’s profile features (i.e., authority and activeness) to increase the accuracy of the prediction. Finally, we perform experiments on a real dataset extracted from Stack Overflow, a domain-specific CQA site focusing on programming knowledge, and it shows our approach outperforms others based on various measures.

Acknowledgement......1
摘要......3
Abstract......4
Table of Contents......5
List of Figures......6
List of Tables......7
1. Introduction......8
2. Related Work......10
2.1 Related Research......10
2.2 Neural Network......11
2.2.1 Representations of Words......11
2.2.2 Convolutional Neural Network......12
3. Preliminary......13
3.1 Task Description......13
3.2 Stack Overflow......13
4. Method......15
4.1 Data Collection......15
4.2 Data Preprocessing......16
4.3 Question & Answerer Embedding......17
4.4 Answerer’s Profile Features......18
4.5 Relevant Score Model Architecture......20
4.6 Recommendation of Ranking Answerers......21
5. Experiments......23
5.1 Main Experiment......23
5.1.1 Experimental Design......23
5.1.2 Performance of the Model and Results......25
5.2 Extensive Experiment......28
5.2.1 Experimental Design......28
5.2.2 Performance of the Model......30
5.2.3 Performance of Results Evaluated......31
6. Conclusion......33
Reference ......34

[1] Amr Azzam, Neamat Tazi, and Ahmad Hany Hossny, “A Question Routing Technique Using Deep Neural Network for Communities of Question Answering,” proceedings of International Conference on Database Systems for Advanced Applications, DASFFAA, 2017, pp. 35–49.
[2] Daniel Hasan Dalip, Marcos André Gonçalves, Marco Cristo, Pável Calado, “Exploiting user feedback to learn to rank answers in q&a forums: a case study with stack overflow,” proceedings of SIGIR 2013, pp. 543–552.
[3] Van Dijk, Tsagkias, and De Rijke, “Early detection of topical expertise in community question answering,” proceedings of SIGIR 2015, pp. 995–998.
[4] Vasiliki Efstathiou, Christos Chatzilenas, and Diomidis Spinellis, “Word Embeddings for the Software Engineering Domain,” MSR '18: Proceedings of the 15th International Conference on Mining Software Repositories, 2018, pp38–41.
[5] Po-Sen Huang, Xiaodong He, Jianfeng Gao, Li Deng, Alex Acero, and Larry P. Heck, “Learning deep structured semantic models for web search using clickthrough data” In Proceedings of the 22nd ACM International Conference on Information & Knowledge Management, CIKM, 2013, pp. 2333–2338.
[6] Armand Joulin, Edouard Grave, Piotr Bojanowski, and Tomas Mikolov, “Bag of Tricks for Efficient Text Classification,” proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers, 2017, pp. 427–431.
[7] Zongcheng Ji, and Bin Wang,” Learning to rank for question routing in community question answering,” proceedings of the 22nd ACM international conference on Information & Knowledge Management, CIKM, 2013, pp. 2363–2368.

[8] Nagendra Kumar, Srikanth G, Karthik Yadav Mudda, Gayam Trishal, Anand Konjengbam, Manish Singh, “Where to Post: Routing Questions to Right Community in Community Question Answering Systems,” ACM India Joint International Conference on Data Science and Management of Data, 2019, pp. 136–142.
[9] Yoon Kim, “Convolutional Neural Networks for Sentence Classification,” proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, EMNLP, 2014, pp. 1746–1751.
[10] Zhu Liu, Kan Li1, and Dacheng Qu,” Knowledge Graph Based Question Routing for Community Question Answering,” proceedings of International Conference on Neural Information Processing, ICONIP, 2017, pp 721–730.
[11] Sara Mumtaz, Carlos Rodriguez, and Boualem Benatallah, “Expert2Vec: Experts Representation in Community Question Answering for Question Routing,” International Conference on Advanced Information Systems Engineering, CAiSE, 2019, pp. 213–229.
[12] Tomas Mikolov, Ilya Sutskever, Kai Chen, Gregory S. Corrado, and Jeffrey Dean, “Distributed Representations of Words and Phrases and their Compositionality,” proceedings of Advances in Neural Information Processing Systems 26, NIPS, 2013.
[13] Adam Paszke, “Automatic differentiation in Pytorch,” 2017.
[14] Thiago Baesso Procaci, Bernardo Pereira Nunes, Terhi Nurmikko-Fuller, Sean W. M. Siqueira, “Finding Topical Experts in Question & Answer Communities,” IEEE 16th International Conference on Advanced Learning Technologies, 2016.
[15] Fatemeh Riahi, Zainab Zolaktaf, Mahdi Shafiei, and Evangelos Milios, “Finding expert users in community question answering,” proceedings of the 21st International Conference on World Wide Web, WWW, 2012, pp. 791–798.
[16] Pradeep Kumar Roy, Jyoti Prakash Singh, Amitava Nag, “Finding Active Expert Users for Question Routing in Community Question Answering Sites,” International Conference on Machine Learning and Data Mining in Pattern Recognition, 2018, pp. 440–451.
[17] Bin Shao and Jiafei Yan, “Recommending Answerers for Stack Overflow with LDA Model,” proceedings of the 12th Chinese Conference on Computer Supported Cooperative, ChineseCSCW, 2017, pp. 80-86.
[18] Xiaoxue Shen, Liyang Gu, Adele Lu Jia, “Cultivating Online: Question Routing in a Question and Answering Community for Agriculture,” 2019, arXiv:1904.08199v1. [Online]. Available: https://arxiv.org/abs/1904.08199
[19] Jian Wang, Jiqing Sun, Hongfei Lin, Hualei Dong, and Shaowu Zhang, “Predicting Best Answerers for New Questions: An Approach Leveraging Convolution Neural Networks in Community Question Answering,” Chinese National Conference on Social Media Processing, SMP, 2016, pp. 29–41.
[20] Lin Wang, Bin Wu, Juan Yang and Shuang Peng,” Personalized Recommendation for New Questions in Community Question Answering,” 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining.
[21] Baoguo Yang, and Suresh Manandhar, “Tag-based expert recommendation in community question answering,” proceedings of the IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, 2014.
[22] Tom Chao Zhou, Michael R. Lyu, and Irwin King, “A Classification-based Approach to Question Routing in Community Question Answering,” 21st International Conference on World Wide Web, WWW, 2012, pp. 783–790.
[23] Zhou Zhao, Qifan Yang, Deng Cai, Xiaofei He and Yueting Zhuang, “Expert Finding for Community-Based Question Answering via Ranking Metric Network Learning” proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI, 2016.