About my research
My research is devoted to recognition of fine-grained emotions from textual social media streams. The aim is to automatically detect if the text author felt an emotion and to classify it in terms of categorical names. In difference with more traditional problem of opinion extraction (i.e. extraction of positive or negative attitude of the author), I focus on distinguishing the fine-grained set of emotion categories. The chosen set contains 20 categories, including such emotions as Pride, Happiness, Amusement, Love, Relief, Worry, Disappointment and Anger. Having this level of granularity would allow the analysts to distinguish the nuances of reactions in social media and perform the detailed content analysis.
In addition to high granularity, the challenge of building such emotion classifiers is the lack of high-quality annotated data. The annotation is complicated because of the subjective nature and low consensus of emotion assessment. To overcome these problems, I design the human computation techniques allowing collecting such annotations of texts and phrases with higher consistency. As another way to address the lack of annotated data, I also study how to adapt the semi-supervised methods to emotion recognition.
Another direction of my thesis is analysis and summarization of the context in which emotions appear in social media.
Emotion Recognition, Text Analysis, Data Mining, Machine Learning, Human Computation, Human Computer Interaction, Social Media Analysis
- Ongoing: Ph.D. in Computer Science. School of Computer and Communication Sciences, Swiss Federal Institute of Technology in Lausanne (EPFL). Starting from September 2011.
- M.S. in Applied Mathematics and Physics. Department of Control and Applied Mathematics at Moscow Institute of Physics and Technology (MIPT), Russia. 2009 – 2011
Specialization: Mathematics and Information Technologies.
Thesis title: The description of the function sets with finite relations on domain and range
- B.S. in Applied Mathematics and Physics. Department of Control and Applied Mathematics at Moscow Institute of Physics and Technology (MIPT), Russia. 2005 – 2009
- Valentina Sintsova, Claudiu Musat and Pearl Pu. ”Semi-Supervised Method for Multi-Category Emotion Recognition in Tweets.” In Proceedings of the 4th Workshop on the Sentiment Elicitation from Natural Text for Information Retrieval and Extraction (SENTIRE). Shenzhen, China, December 14, 2014 [access]
- Renato Kempter, Valentina Sintsova, Claudiu Musat, and Pearl Pu. “EmotionWatch: Visualizing Fine-Grained Emotions in Event-Related Tweets.” In the 8th International AAAI Conference on Weblogs and Social Media (ICWSM), 2014 [access]
- Lionel Martin, Valentina Sintsova, and Pearl Pu. ”Are Influential Writers More Objective? An Analysis of Emotionality in Review Comments.” In the 5th International Workshop on Social Recommender Systems, in conjunction with 23rd International World Wide Web Conference (WWW 2014), 799 – 804. Seoul, Korea, 2014 [access]
- Renato Kempter, Valentina Sintsova, Claudiu Musat, and Pearl Pu. “Discover Emotions in Tweets & Weibos during the 2012 Olympic Games.” In Video Track of the 23rd International Joint Conference on Articifial Intelligence (IJCAI). Beijing, China, 2013. [link]
- Valentina Sintsova, Claudiu Musat, and Pearl Pu. “Fine-Grained Emotion Recognition in Olympic Tweets Based on Human Computation.” In Proceedings of the 4th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis (WASSA), 12–20. Atlanta, Georgia: Association for Computational Linguistics, 2013 [pdf]
- Valentina Sintsova and Pearl Pu. “Importance of Emotion Awareness for Emotional Well-Being and for Improvement of Social Communications.” Position paper at the ARV Workshop on Tools and Technologies for Emotion Awareness in Computer-Mediated Collaboration and Learning, 2013