臺灣博碩士論文加值系統

搜尋(搜索)廣告是一樁由搜尋引擎、廣告主和使用者構成的三方生意。起初這只是一個互利關係，搜尋引擎提供純搜尋結果旁的位置，給特定關鍵字的廣告主使用。在搜尋引擎日趨發達之後，使用者們去搜尋常常是為了他們的興趣或購買上的選擇。因此，適時地展示適當的廣告給使用者可以促使他們去點擊這些搜索廣告。隨著廣告的快速增長，搜尋引擎產生了一個關鍵字競價的機制，讓這些廣告主對他們廣告所使用的關鍵字做競價。基於競價的關鍵字及廣告本身的品質，搜尋引擎可以排名出廣告放置的順序。廣告商就得斟酌關鍵字的用字組合來提升整個廣告的品質，進而增加廣告被點擊的機會。而在廣告品質中有一項重要因素就是廣告點擊的歷史表現。所以截至目前，如何有效地提升廣告成效來得到更多的點擊一直是一個重要的課題。
此篇研究主要著重在智慧型手機的領域並產生一個社群傾向的模型，加上廣告的相關特徵來預測未來廣告點擊率的趨勢。就社群傾向模型而言，我們藉由分析科技類論壇中的中文文章內容，得到四項特點，分別是：熱門(Hotness)、情緒(Sentiment)、優惠(Promotion)和事件(Event)。而我們的實驗結果指出，只要事先知道大眾的意見傾向或事件的發生，將有助於增強廣告的點擊預測。由社群傾向來調整廣告的競價關鍵字，這對廣告主在提升廣告成效上有相當大的助益。

Search advertising is a tripartite business between search engine, advertisers, and users (consumers). In the beginning, it is a mutually beneficial relation that search engines provide placements next to the original search results for advertisers on specific keywords. With the developed search engines, users often search for their interests or purchasing decision. Therefore timely presenting proper advertisements to users will encourage them to click on search ads. With the rapid growth of advertising, there is a bidding mechanism that advertisers need to bid keywords on their ads. Based on bidding keywords and ad’s quality, search engines rank ads for displaying position. Advertisers should carefully compose keywords for increasing ads quality in order to enhance the opportunity for their ads to be clicked. One of the important factors in ad quality is historical click performance. Until now, how to efficiently improve the ad performance to earn more clicks remains a main task.
In this paper, we focus on the scope of smart phone and produce a social intention model with advertising based features to forecast future trend on ads’ click-through rate (CTR). In terms of social intention model, we analyze Chinese text content of technology forum to derive social intentional factors which are Hotness, Sentiment, Promotion, and Event. Our results indicate that with knowing public opinions or occurring events beforehand can efficiently enhance click prediction. This will be very helpful for advertisers on adjusting bidding keywords to improve ad performance via social intention.

中文摘要 III
ABSTRACT IV
TABLE LISTING VIII
FIGURE LISTING X
1. INTRODUCTION 1
1.1 Background 1
1.2 Motivation 3
1.3 Approach abstract 7
1.4 Paper structure 9
2. RELATED WORK 10
2.1 Traditional click prediction problem 10
2.1.1 On the basis of keywords 10
2.1.2 User behavior 12
2.1.3 Similarity and semantic 14
2.2 Using social media for prediction 15
3. Method 16
3.1 Problem definition 17
3.2 Social intention-based model 17
3.2.1 Natural Language Processing (NLP) 17
3.2.2 Social intentional methods 21
3.2.3 Summary 29
3.3 Advertising-based model 29
3.4 Prediction model 31
4. EXPERIMENTS 34
4.1 Dataset and preprocessing 34
4.1.1 Technology forum dataset 34
4.1.2 Advertising dataset 36
4.1.3 Preprocessing 36
4.2 Evaluation metrics 37
4.3 Experiment process 38
4.3.1 Correlation and causality 38
4.3.2 Baseline 41
4.3.3 Criteria selection on social intentional factors 42
4.3.4 Previous seven days’ data usage 46
4.4 Results and discussion 48
4.4.1 Training with one keyword data 48
4.4.2 Training with more than one keyword data 52
5. CONCLUSION AND FUTURE WORKS 55
6. REFERENCES 56

[1].Bollen, J., H. Mao, and X.-J. Zeng, Twitter mood predicts the stock market. Journal of Computational Science, Volume 2, Issue 1, March 2011, Pages 1-8.
[2].Bollen, J., A. Pepe, and H. Mao, Modeling public mood and emotion: Twitter sentiment and socio-economic phenomena. CoRR, 2009. abs/0911.1583.
[3].Buscher, G., S.T. Dumais, and E. Cutrell, The good, the bad, and the random: an eye-tracking study of ad quality in web search, in Proceedings of the 33rd international ACM SIGIR conference on Research and development in information retrieval. 2010, ACM: Geneva, Switzerland. p. 42-49.
[4].Chang, C.-C. and C.-J. Lin, LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2011. 2: p. 27:1-27:27.
[5].Chen, Y. and T.W. Yan, Position-normalized click prediction in search advertising, in Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. 2012, ACM: Beijing, China. p. 795-803.
[6].Cheng, H. and E. Cantú-Paz, Personalized click prediction in sponsored search, in Proceedings of the third ACM international conference on Web search and data mining. 2010, ACM: New York, New York, USA. p. 351-360.
[7].Dave, K.S. and V. Varma, Learning the click-through rate for rare/new ads from similar ads, in Proceedings of the 33rd international ACM SIGIR conference on Research and development in information retrieval. 2010, ACM: Geneva, Switzerland. p. 897-898.
[8].Guo, Q., E. Agichtein, C.L.A. Clarke, and A. Ashkan, In the Mood to Click? Towards Inferring Receptiveness to Search Advertising, in Proceedings of the 2009 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Volume 01. 2009, IEEE Computer Society. p. 319-324.
[9].Hillard, D., S. Schroedl, E. Manavoglu, H. Raghavan, and C. Leggetter, Improving ad relevance in sponsored search, in Proceedings of the third ACM international conference on Web search and data mining. 2010, ACM: New York, New York, USA. p. 361-370.
[10].Kolesnikov, A., Y. Logachev, and V. Topinskiy, Predicting CTR of new ads via click prediction, in Proceedings of the 21st ACM international conference on Information and knowledge management. 2012, ACM: Maui, Hawaii, USA. p. 2547-2550.
[11].Ku, L.-W. and H.-H. Chen, Mining opinions from the Web: Beyond relevance retrieval. J. Am. Soc. Inf. Sci. Technol., 2007. 58(12): p. 1838-1850.
[12].Mokhlis, S.Y., Azizul Yadi, Consumer Choice Criteria in Mobile Phone Selection: An Investigation of Malaysian University Students. International Review of Social Sciences ＆ Humanities, 2012. 2(2): p. 203-212.
[13].Neshat, H.S., Ranking of new sponsored online ads, in Fourth International Conference on the Applications of Digital Information and Web Technologies. 2011, IEEE: Stevens Point, Wisconsin, USA. p. 7-12.
[14].O. Stitelman, C.P., B. Dalessandro, F. Provost, Estimating The Effect Of Online Display Advertising On Browser Conversion, in The Fifth International Workshop on Data Mining and Audience Intelligence for Online Advertising at SIGKDD 2011: San Diego, California USA.
[15].Perlich, C., B. Dalessandro, R. Hook, O. Stitelman, T. Raeder, and F. Provost, Bid optimizing and inventory scoring in targeted online advertising, in Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. 2012, ACM: Beijing, China. p. 804-812.
[16].Piwowarski, B. and H. Zaragoza, Predictive user click models based on click-through history, in Proceedings of the sixteenth ACM conference on Conference on information and knowledge management. 2007, ACM: Lisbon, Portugal. p. 175-182.
[17].Regelson, M. and D.C. Fain, Predicting Click-Through Rate Using Keyword Clusters, in 06th ACM Conference on Electronic Commerce. 2006.
[18].Richardson, M., E. Dominowska, and R. Ragno, Predicting clicks: Estimating the click-through rate for new ads, in In Proceedings of the 16th International World Wide Web Conference (WWW-07. 2007, ACM Press. p. 521-530.
[19].Selçuk, B. and Ö. Özlük, Optimal keyword bidding in search-based advertising with target exposure levels. European Journal of Operational Research, 2013. 226(1): p. 163-172.
[20].Xiong, C., T. Wang, W. Ding, Y. Shen, and T.-Y. Liu, Relational click prediction for sponsored search, in Proceedings of the fifth ACM international conference on Web search and data mining. 2012, ACM: Seattle, Washington, USA. p. 493-502.
[21].Yan, J., N. Liu, G. Wang, W. Zhang, Y. Jiang, and Z. Chen, How much can behavioral targeting help online advertising?, in Proceedings of the 18th international conference on World wide web. 2009, ACM: Madrid, Spain. p. 261-270.