About

As the Web rapidly evolves, people are becoming increasingly enthusiastic about interacting, sharing, and collaborating through social networks, online communities, blogs, wikis, and the like. In recent years, this collective intelligence has spread to many different areas, with particular focus on fields related to everyday life such as commerce, tourism, education, and health, causing the size of the social Web to expand exponentially.

To identify the emotions (e.g. sentiment polarity, sadness, happiness, anger, irony, sarcasm, etc.) and the modality (e.g. doubt, certainty, obligation, liability, desire, etc.) expressed in this continuously growing content is critical to enable the correct interpretation of the opinions expressed or reported about social events, political movements, company strategies, marketing campaigns, product preferences, etc.

This has raised growing interest both within the scientific community, by providing it with new research challenges, as well as in the business world, as applications such as marketing and financial prediction would gain remarkable benefits.

One of the main application tasks in this context is opinion mining [1], which is addressed by a significant number of Natural Language Processing techniques, e.g. for distinguishing objective from subjective statements [2], as well as for more fine-grained analysis of sentiment, such as polarity and emotions [8]. Recently, this has been extended to the detection of irony, humor, and other forms of figurative language [3]. In practice, this has led to the organisation of a series of shared tasks on sentiment analysis, including irony and figurative language detection (SemEval 2013, 2014, 2015), with the production of annotated data and development of running systems.

However, existing solutions still have many limitations leaving the challenge of emotions and modality analysis still open. For example, there is the need for building/enriching semantic/cognitive resources for supporting emotion and modality recognition and analysis. Additionally, the joint treatment of modality and emotion is, computationally, trailing behind, and therefore the focus of ongoing, current research. Also, while we can produce rather robust deep semantic analysis of natural language, we still need to tune this analysis towards the processing of sentiment and modalities, which cannot be addressed by means of statistical models only, currently the prevailing approaches to sentiment analysis in NLP. The hybridization of NLP techniques with Semantic Web technologies is therefore a direction worth exploring, as recently shown in [4, 5, 6, 7].

Based on the lessons learnt from the first edition, this year the scope of the workshop is a bit broader (although still focusing on a very specific domain) and accepted submissions will include abstracts and position papers in addition to full papers. The workshops main focus will be discussion rather than presentations, which are seen as seeds for boosting discussion topics, and an expected result will be a joint manifesto and a research roadmap that will provide the Semantic Web community with inspiring research challenges.

The Workshop will be connected to the ESWC 2017 Fine-Grained Sentiment Analysis Challenge. Both the Workshop and the Challenge can benefit from a Google Group, called Semantic Sentiment Analysis Initiative. Please post messages related to the Workshop under the discussion “ESWC 2017 Workshop on Emotions, Modality, Sentiment Analysis and the Semantic Web.”

REFERENCES

[1]  Bo, P., and Lee, L. (2008). Opinion mining and sentiment analysis. Foundations and Trends in Information Retrieval , 2 (1-2), 1-135.

[2]  Wiebe, J., and Ellen, R. (2005). Creating Subjective and Objective Sentence Classifiers from Unannotated Texts. Computational Linguistics and Intelligent Text Processing 6th International Conference, CICLing (pp. 486-497). Mexico City: Springer.

[3]  Paula, C., Sarmento, L., Silva, M. J., and de Oliveira, E. (2009). Clues for detecting irony in user-generated contents: oh…!! it’s so easy;-). Proceedings of the 1st international CIKM workshop on Topic-sentiment analysis for mass opinion (pp. 53-56). ACM.

[4]  Reforgiato Recupero, D., Presutti, V., Consoli, S., and Gangemi, A. (2014). Sentilo: Frame-Based Sentiment Analysis. Cognitive Computation , 1-15.

[5]  Saif, H., He, Y., and Alani, H. (2012). Semantic sentiment analysis of Twitter. 11th International Semantic Web Conference (ISWC 2012) (pp. 508-524). Springer.

[6]  Gangemi, A., Presutti, V., and Reforgiato Recupero, D. (2014). Frame- based detection of opinion holders and topics: a model and a tool. IEEE Computational Intelligence , 9 (1), 20-30.

[7]  Cambria, E., and Hussain, A. (2012). Sentic Computing: Techniques, Tools, and Applications. Springer.

[8]  Liu, B. (2012). Sentiment Analysis and Opinion Mining. Synthesis Lectures on Human Language Technologies. Chicago: Morgan & Claypool Publishers.