近來在社群網路中找尋具有關鍵影響的節點已在許多實際應用上成為熱門的研究領域，像是應用在廣告產業即具有明顯的價值。在學術領域上，各機構與大學的學者們時常投稿各式性質的期刊與會議，而我們有時也可以在報章雜誌上見到人們也熱衷於以各種方式來論斷其價值與排行，然而要公式化地去評定一個學術機構或一個刊物的影響並不是非常容易的，由於量化的程度或選擇的參數不同，結果也有可能相差甚遠，因此許多評分方式也因排名方式不公開而遭人議論。

評定排名的準確可以透過預測驗證，而過去相關方法多重於引用數與獎項數的多寡，因此我們試著提出一種加入情緒分析與網路聲望的量化方式
，希望能在原有的數量分析外加入更多作為人與群眾的觀感影響，用以增進原本的預測準確度並檢驗情緒分析的應用。

在本篇論文中，我們提出了一種評估基於情緒計算出研究機構的聲望造成的影響力，並結合其他基本特徵來確定它們在多個學術會議審查過程中的相關性。

Finding influential nodes or identifying existing patterns in a social network have recently been an active research field. Some practical applications such as advertising industry have benefited from this type of research. Similarly in academia, conferences or journals often consider the impact or influence in a university or research institution has when evaluating a submission. This has led to a growing interest in ways to judge this kind of value which can lead to an eventual ranking. However, there is still not a standard quantification or scoring method for this task.

In this thesis we propose a method to evaluate the impact of sentiment-based reputation of research institutions combined with other baseline features to determine their relevance in the review process of several academic conferences.

摘要 i
Abstract ii
Acknowledgement iii
List of Tables vi
List of Figures vii
1 Introduction 1
2 Related Work 4
2.1 Academic Ranking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Sentiment Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Methodology 6
3.1 Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2 Baseline Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2.1 Sum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2.2 Contribution-Separated Sum . . . . . . . . . . . . . . . . . . . . . 8
3.2.3 Affiliation Impact Factor . . . . . . . . . . . . . . . . . . . . . . . 9
3.2.4 Author Impact Factor . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.3 SentimentBoost Approach . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4 Experiment 13
4.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1.1 Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1.2 System Configuration . . . . . . . . . . . . . . . . . . . . . . . . 14
4.1.3 Evaluation Method . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2 Experimental Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.2.1 Comparison of baseline approaches . . . . . . . . . . . . . . . . . 15
4.2.2 Evaluation of each parameters in SentimentBoost . . . . . . . . . . 17
4.2.3 Evaluation between years . . . . . . . . . . . . . . . . . . . . . . 26
5 Conclusions and Future Works 31
5.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
5.2 Future Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
References 33

[1] Arnab Sinha, Zhihong Shen, Yang Song, Hao Ma, Darrin Eide, Bo-june Paul Hsu, and
Kuansan Wang. An overview of microsoft academic service (mas) and applications.
Proceedings of the 24th International Conference on World Wide Web (WWW ’15
Companion), pages 243–246, 2015.
[2] Kdd cup 2016, 2016. URL http://kddcup2016.azurewebsites.net.
[3] CJ J. Hutto and Eric Gilbert. Vader: A parsimonious rule-based model for sentiment
analysis of social media text. Eighth International Conference on Weblogs and Social
Media (ICWSM-14), pages 216–225, 2014.
[4] A. F. M. A. Al-Juboori, Y. Na, and F. Ko. University ranking and evaluation: Trend
and existing approaches. In Next Generation Information Technology (ICNIT), 2011
The 2nd International Conference on, pages 137–142, June 2011.
[5] Nian Cai Liu and Ying Cheng. The Academic Ranking of World Universities. Higher
Education in Europe, 30(2):127–136, jul 2005. ISSN 0379-7724. doi: 10.1080/
03797720500260116.
[6] Raj Kumar Pan and Santo Fortunato. Author Impact Factor: tracking the dynamics of
33
individual scientific impact. Scientific reports, 4:4880, 2014. ISSN 2045-2322. doi:
10.1038/srep04880.
[7] H.-Y. Chen M.-F. Tsai M.-Y. Yeh M.-H. Feng, K.-H. Chan and S.-D Lin. An effi-
cient solution to reinforce paper ranking using author/venue/citation information - the
winner’s solution for wsdm cup 2016. In WSDM Cup 2016, Febuary 2016.
[8] Bo Pang, Lillian Lee, and Shivakumar Vaithyanathan. Thumbs Up? Sentiment Classi-
fication Using Machine Learning Techniques. Proceedings of the ACL-02 conference
on Empirical methods in natural language processing - EMNLP ’02, 10:79–86, 2002.
ISSN 1554-0669.
[9] Gilad Mishne. Experiments with mood classification in blog posts. Proceedings
of ACM SIGIR 2005 Workshop on Stylistic Analysis of Text for Information Access,
page 19, 2005.
[10] T Wilson, J Wiebe, and P Hoffman. Recognizing contextual polarity in phrase level
sentiment analysis. Acl, 7(5):12–21, 2005. ISSN 0891-2017.
[11] Peter D Turney. Thumbs up or thumbs down? Semantic Orientation applied to Unsupervised
Classification of Reviews. Proceedings of the 40th Annual Meeting of
the Association for Computational Linguistics (ACL), pages 417–424, 2002. ISSN
0738467X.
[12] Carlos Argueta, Elvis Saravia, and Yi-Shin Chen. Unsupervised Graph-Based Patterns
Extraction for Emotion Classification. 2015 IEEE/ACM International Conference
on Advances in Social Networks Analysis and Mining, pages 336–341, 2015.
34
doi: 10.1145/2808797.2809419.
[13] F. J. RICHARDS. A flexible growth function for empirical use. Journal of Experimental
Botany, 10(2):290–301, 1959. doi: 10.1093/jxb/10.2.290.
[14] Neo4j: The world’s leading graph database, 2016. URL http://neo4j.org