The Multifaceted Concept of Context in Sentiment Analysis

The contemporary web is about communication, collaboration, participation, and sharing. Currently, the sharing of content on the web ranges from sharing of ideas and information which includes text, photos, videos, audios, and memes to even gifs. Moreover, the language and linguistic tone of user-generated content are informal and indistinct. Analyzing explicit and clear sentiment is challenging owing to language constructs which may intensify or flip the polarity within the posts. Context-based sentiment analysis is the domain of study which deals with comprehending cues which can enhance the prediction accuracy of the generic sentiment analysis as well as facilitate fine grain analysis of varied linguistic constructs such as sarcasm, humor, or irony. This work is preliminary to understand the what, how and why of using the context in sentiment analysis. The concept of ‘context in use’ is described by exemplifying the types of context. A strength–weakness–opportunity–threat (SWOT) matrix is made to demonstrate the effectiveness of context-based sentiment analysis.

[1]  Rashedur M. Rahman,et al.  Localized twitter opinion mining using sentiment analysis , 2015, Decis. Anal..

[2]  Fangzhao Wu,et al.  Structured microblog sentiment classification via social context regularization , 2016, Neurocomputing.

[3]  Zhijian Wu,et al.  Twitter Sarcasm Detection Exploiting a Context-Based Model , 2015, WISE.

[4]  Erik Cambria,et al.  Multimodal Sentiment Analysis using Hierarchical Fusion with Context Modeling , 2018, Knowl. Based Syst..

[5]  Daling Wang,et al.  Attention based hierarchical LSTM network for context-aware microblog sentiment classification , 2018, World Wide Web.

[6]  Akshi Kumar,et al.  Sentiment analysis of multimodal twitter data , 2019, Multimedia Tools and Applications.

[7]  Michel Ballings,et al.  The added value of auxiliary data in sentiment analysis of Facebook posts , 2016, Decis. Support Syst..

[8]  Tomoaki Ohtsuki,et al.  A Pattern-Based Approach for Sarcasm Detection on Twitter , 2016, IEEE Access.

[9]  Harith Alani,et al.  Contextual semantics for sentiment analysis of Twitter , 2016, Inf. Process. Manag..

[10]  Nirmalie Wiratunga,et al.  Contextual sentiment analysis for social media genres , 2016, Knowl. Based Syst..

[11]  Ping Li,et al.  Augmented sentiment representation by learning context information , 2018, Neural Computing and Applications.

[12]  Elisabetta Fersini,et al.  Approval network: a novel approach for sentiment analysis in social networks , 2017, World Wide Web.

[13]  Akshi Kumar,et al.  Systematic literature review on context-based sentiment analysis in social multimedia , 2019, Multimedia Tools and Applications.

[14]  Huimin Zhao,et al.  Resolving Ambiguity in Sentiment Classification , 2017, ACM Trans. Manag. Inf. Syst..

[15]  Reshma Sheik,et al.  Entity Level Contextual Sentiment Detection of Topic Sensitive Influential Twitterers Using SentiCircles , 2018 .

[16]  David Bamman,et al.  Contextualized Sarcasm Detection on Twitter , 2015, ICWSM.

[17]  Preslav Nakov,et al.  Developing a successful SemEval task in sentiment analysis of Twitter and other social media texts , 2016, Language Resources and Evaluation.