業者欲搜尋適合其品牌或產品/服務之意見領袖時，其往往先透社群平台（如Facebook、Instagram、YouTube等）觀察具代表性討論內容，並自此些討論內容得知其點閱率、回覆數、…等資訊，進而判定討論內容受共鳴、關注之程度，從而決定潛在之意見領袖（即所發表之討論內容獲得較高之點閱率、回覆數，引發較多共鳴與關注者）。最後，業者再由多位潛在意見領袖中選定符合其產品/服務特質之意見領袖為其品牌或產品/服務進行宣傳與廣告代言。然而，業者透過各社群媒體所觀察之具代表性討論內容及其所對應之回覆數量眾多且形式不一，其判定討論內容受共鳴、關注程度之過程往往需花費大量時間；此外，業者決定潛在之意見領袖多半以點閱率、回覆數、粉絲數、…等量化數據為判斷意見領袖之基準，而忽略討論內容中回覆對象的特性所涵蓋之領域範圍（如性別、年齡層、感興趣之事物等），導致業者選定之意見領袖影響的對象與業者鎖定的目標客群未盡吻合。此外，業者選定之不同意見領袖所影響的對象亦可能有相互重疊之虞，或其選定之所有意見領袖所影響的對象未盡完整地包含業者所設定的目標客群，而使業者委託意見領袖進行宣傳與廣告代言之效益不如其所預期。
為解決上述問題，本研究乃先透過社群平台蒐集討論內容與其所對應之回覆，並解析此些討論內容與其所對應之回覆之關鍵特徵，以釐清討論內容與其所對應之回覆的關鍵特徵之表達方式。依前述作業之解析結果，本研究發展一套「領袖聚落之分析與視覺化呈現」方法論，首先，此方法乃將所蒐集之討論內容與其所對應之回覆內容依各關鍵特徵之表達方式擷取其對應的特徵值；接著，此方法利用聚合式階層分群法將相似之粉絲關注點的潛在意見領袖歸列於相同集群而形成不同影響領域的領袖聚落，以利業者清楚瀏覽各領袖聚落之相似粉絲關注點所涵蓋的領域範圍；之後，此方法乃計算各聚落中任兩潛在意見領袖所影響之粉絲的相似程度，以協助業者快速理解不同領袖聚落對應之潛在意見領袖所影響粉絲族群差異；最後，此方法乃整併各領袖聚落中各潛在意見領袖之討論內容與其所對應之回覆內容的特徵值，並以視覺化方式呈現各領袖聚落中各潛在意見領袖之粉絲重要特性的佔比以及聚落中任兩潛在意見領袖之相似程度。

When service providers want to search for opinion leaders who are suitable for their brand or product/service, they often first observe the representative discussion content and discussion content and the corresponding responses through the social media of the Internet, and the relevant discussion content from this Information such as the rate of clicks, the number of responses, etc., determine the extent to which the content of the discussion is resonated and concerned, and then determine the potential opinion leader. However, the representative discussion content that service providers observe through various social media and the number of responses they respond to are numerous and varied, and the observation process often takes a lot of time. In addition, service providers decide that potential opinion leaders are mostly The quantitative data such as the reading rate, the number of responses, the number of fans, and so on are the benchmarks for judging the opinion leaders, while ignoring the scope of the coverage of the characteristics of the responding objects in the discussion content, causing the service provider's selected opinion leaders to influence the objects and service providers locked in. The target audience is not consistent.
In order to solve the above problems, this study first collects the discussion content and the corresponding responses through the social media, and analyzes the key features of the discussion content and the corresponding responses to clarify the key features of the discussion content and the corresponding response way of expression. Then, in this study, the collected content and the corresponding reply content are extracted according to the expressions of each key feature; then, according to the discussion content, the "fan characteristics" and "points of interest" are collected. The type feature extraction results grouping potential opinion leaders to obtain a plurality of different fan characteristics and post-collection leaders. Finally, visually present the results of the leader settlements of the different influence areas divided by the above steps.

摘要 I
ABSTRACT II
目錄 III
圖目錄 IV
表目錄 V
第1章、 研究背景 1
1.1研究動機與目的 1
1.2研究步驟 6
1.3研究定位 9
第2章、 文獻回顧 12
2.1 社群討論內容之特徵擷取 12
2.2 社群討論之意見領袖推論 26
2.3 社群討論內容之視覺化呈現 39
第3章、 以社群互動內容為基礎之意見領袖配置分析 44
3.1討論內容與其所對應之回覆內容結構解析 46
3.2討論內容與其所對應之回覆內容特徵擷取 71
3.3潛在意見領袖分群 87
3.4潛在意見領袖之相似程度計算 104
3.5各潛在意見領袖對應之關鍵特徵統整與視覺化呈現 115
3.6小結 128
第4章、 績效驗證 130
4.1模式驗證方式說明 130
4.2模式驗證結果分析 134
第5章、 結論與未來展望 149
5.1論文總結 150
5.2未來發展 152
參考文獻 154
附錄A、模式驗證資料 158
附錄B、「意見領袖所影響之粉絲族群的特性判定」議題驗證之問卷設計 219

1. Jehan W., Siqi Z., Lin Z. and Venu V., 2010, “Analyzing Twitter for Social TV: Sentiment Extraction for Sports,” FutureTV Workshop, Vol. 720, pp. 140-145.
2. Wu H., Xin T., Ran T., Weidong Z., Gongjun Y. and Vasudeva A., 2017, “Application of Social Media Analytics: A Case of Analyzing Online Hotel Reviews,” Online Information Review, Vol. 41, pp. 921-935.
3. Benny, A. and Philip, M., 2015, “Keyword Based Tweet Extraction and Detection of Related Topics,” Procedia Computer Science, Vol. 46, pp. 364-371.
4. Nikfarjam, A., Sarker, A., O’Connor, K., Ginn, R. and Gonzalez, G., 2015, “Pharmacovigilance from Social Media: Mining Adverse Drug Reaction Mentions Using Sequence Labeling with Word Embedding Cluster Features,” Journal of the American Medical Informatics Association, pp. 671-681.
5. Imran, M., Elbassuoni, S., Diaz, F., Meier, P. and Castillo, C., 2013, “Practical Extraction of Disaster-Relevant Information from Social Media,” Proceedings of the 22nd International Conference on World Wide Web, pp. 1021-1024.
6. Pereira, J., Pasquali, A., Saleriro, P. and Rossetti, R., 2017, “Transportation in Social Media: An Automatic Classifier for Travel-Related Tweets,” EPIA 2017: Progress in Artificial Intelligence, pp. 355-366.
7. Bordoloi, M. and Biswas, S., 2018, “Keyword Extraction from Micro-Blogs Using Collective Weight,” Social Network Analysis and Mining, Vol. 8, pp. 1-16.
8. Rafea, A. and Mostafa, N., 2013, “Topic Extraction in Social Media,” 2013 International Conference on Collaboration Technologies and Systems (CTS), pp. 94-98.
9. Wang, X., Gerber, M. and Brown, D., 2012, “Automatic Crime Prediction Using Events Extracted from Twitter Posts,” Social Computing, Behavioral - Cultural Modeling and Prediction, pp. 231-238.
10. Jin, P., Mu, L., Zheng, L., Zhao, J., Yue, L. and Brown, D., 2017, “News Feature Extraction for Events on Social Network Platforms,” WWW '17 Companion: Proceedings of the 26th International Conference on World Wide Web Companion, pp. 69-78.
11. Raymond, Y. K., Yunqing, X. and Yunming, Y., 2014, “A Probabilistic Generative Model for Mining Cybercriminal Networks from Online Social Media,” IEEE Computational Intelligence Magazine, Vol. 9, pp. 31-43.
12. Diana, P., Mostofa A., Kunpeng Z., Kathy L., Christopher M., Yves X., Daniel H., Ankit A., Wei-keng L. and Alok C., 2012, “User-Interest based Community Extraction in Social Networks,” The 6th SNA-KDD Workshop, Vol. 12
13. Shu, K., Sliva, A., Wang, S., Tang, J. and Liu, H., 2017, “Fake News Detection on Social Media: A Data Mining Perspective,” SIGKDD Explorations, Vol. 19, No. 1, pp. 22-36.
14. Imran, M., Castillo, C., Lucas, J., Meier, P. and Vieweg, S., 2014, “AIDR: Artificial Intelligence for Disaster Response,” WWW '14 Companion: Proceedings of the 23rd International Conference on World Wide Web, pp. 159-162.
15. Maupom´e, D. and Meurs, M., 2018, “Using Topic Extraction on Social Media Content for the Early Detection of Depression,” Conference and Labs of the Evaluation Forum 2018, pp. 1-5.
16. Zhou, D., Chen, L. and He, Y., 2015, “An Unsupervised Framework of Exploring Events on Twitter: Filtering, Extraction and Categorization,” Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, pp. 2468-2474.
17. Shao, S., Tunc, C., Al-Shawi, A. and Hariri, S., 2020, “Automated Twitter Author Clustering with Unsupervised Learning for Social Media Forensics,” 2019 IEEE/ACS 16th International Conference on Computer Systems and Applications, pp. 1-8.
18. Yu, X., Wei, X. and Lin, X., 2012, “Networking Groups Opinion Leader Identification Algorithms Based on Sentiment Analysis,” Computer Science, Vol. 39, No. 2, pp. 34-37, 46.
19. Jing, L. and Lizhen, X., 2014, “Identification of Microblog Opinion Leader Based on User Feature and Interaction Network,” The 11th Web Information System and Application Conference, pp. 125-130.
20. Khan, N. S., Ata, M. and Rajput, Q., 2015, “Identification of Opinion Leaders in Social Network,” 2015 International Conference on Information and Communication Technologies, pp. 1-6.
21. Jain, L. and Katarya, R., 2018, “Identification of Opinion Leader in Online Social Network Using Fuzzy Trust System,” 2018 IEEE 8th International Advance Computing Conference, pp. 233-239.
22. Wang, C., Du, Y. J. and Tang, M. W., 2016, “Opinion Leader Mining Algorithm in Microblog Platform Based on Topic Similarity,” 2016 2nd IEEE International Conference on Computer and Communications, pp. 160-165.
23. Shinde, M. and Girase, S., 2016, “Identification of Topic-Specific Opinion Leader Using SPEAR Algorithm in Online Knowledge Communities,” 2016 International Conference on Computing, Analytics and Security Trends, pp. 144-149.
24. Chen, Y. and Hsu, H., 2016, “A cluster-Based Opinion Leader Discovery in Social Network,” 2016 Conference on Technologies and Applications of Artificial Intelligence, pp. 78-83.
25. Cao, H., Wang, J. and Wang, Z., 2018, “Opinion Leaders Discovery in Social Networking Site Based on the Theory of Propagation Probability,” 2018 2nd IEEE Advanced Information Management, Communicates, Electronic and Automation Control Conference, pp. 700-704.
26. Jain, L., Katarya, R. and Sachdeva, S, 2019, “Opinion Leader Discovery Based on Text Analysis in Online Social Network,” The 4th International Conference on Information Systems and Computer Networks, pp. 446-450.
27. Dewi, F. K., Yudhoatmojo, S. B. and Budi, I, 2017, “Identification of Opinion Leader on Rumor Spreading in Online Social Network Twitter Using Edge Weighting and Centrality Measure Weighting,” The Twelfth International Conference on Digital Information Management, pp. 313-318.
28. Jiang, L., Ge, B., Xiao, W. and Gao, M., 2017, “BBS Opinion Leader Mining Based on an Improved PageRank Algorithm Using MapReduce,” The Chinese Automation Congress, pp. 392-396.
29. Aleahmad, A., Karisani, P. Rahgozar, M. and Oroumchian, F., 2015, “Identification of Opinion Leader on Rumor Spreading in Online Social Network Twitter Using Edge Weighting and Centrality Measure Weighting,” Journal of Information Science, pp. 1-18.
30. Gao, L., Wu, Y., Xiao, X. and Tang, J., 2019, “Discriminating Topical Influencers Based on the User Relative Emotion,” IEEE Access, Vol. 7, pp. 100120-100130.
31. Li, H., Huang, S. and Sun, G., 2015, “An Opinion Leader Perceptual Model Based on PageRank Algorithm,” 2015 International Conference on Behavioral, Economic and Socio-cultural Computing, pp. 150-155.
32. Deng, X., Li, Y., and Lin, L., 2013, “Parallel Micro Blog Crawler Construction for Effective Opinion Leader Approximation,” 2013 AASRI Procedia, Vol. 5, pp. 170-176.
33. Lin, X., and Han, W., 2013, “Opinion Leaders Discovering in Social Networks Based on Complex Network and DBSCAN Cluster,” 14th International Symposium on Distributed Computing and Applications for Business Engineering and Science, pp. 292-295.
34. Miao, Q., Zhang, S., Meng, Y., and Yu, H., 2013, “Domain-Sensitive Opinion Leader Mining from Online Review Communities,” Proceedings of the 22nd International Conference on World Wide Web 2013, pp. 187-188.
35. Wu, Y., Liu, S., Yan, K., Liu, M. and Wu, F., 2014, “OpinionFlow: Visual Analysis of Opinion Diffusion on Social Media,” IEEE Transactions on Visualization and Computer Graphics, Vol. 20, pp. 1763-1772.
36. Sun, G., Wu, Y., Liu, S., Peng, T., Zhu, J. J. H. and Liang, R., 2014, “EvoRiver: Visual Analysis of Topic Coopetition on Social Media,” IEEE Transactions on Visualization and Computer Graphics, Vol. 20, pp. 1753-1762.
37. Coppersmith, G., Leary, R., Whyne, E. and Wood, T., 2015, “Quantifying Suicidal Ideation via Language Usage on Social Media,” Proceedings of the Statistics Meeting in Statistical Computing Section, pp. 1-16.
38. Ji, X., Chun, S. A. and Geller, J., 2012, “Epidemic Outbreak and Spread Detection System Based on Twitter Data,” International Conference on Health Information Science, pp. 152-163.
39. Azmandian, M., Singh, K., Gelsey, B., Chang, Y. H. and Maheswaran, R., 2012, “Following Human Mobility Using Tweets,” International Workshop on Agents and Data Mining Interaction, pp. 139-149.