International Conference on Computing, Communication and Automation (ICCCA2015) ISBN:978-1-4799-8890-7/15/$31.00 ©2015 IEEE 52 EmotionFinder: Detecting Emotion From Blogs and Textual Documents Shiv Naresh Shivhare School of Computing Science and Engineering Galgotias University Uttar Pradesh, India Shiv827@gmail.com Shakun Garg Department of Computer Science and Engineering BBDIT Ghaziabad Uttar Pradesh, India shakunbah@gmail.com Anitesh Mishra Amdocs Development Centre Gurgaon Haryana, India. anitesh.mishra@yahoo.co.in Abstract—Emotion Detection is one of the most emerging issues in human machine interaction. Detecting emotional state of a person from textual data is an active research field along with recognizing emotions from facial and audio information. Several methods were given to recognize emotion from text in previous years. This paper proposed a new architecture (a keyword based approach) to recognize emotions from text. In case of recognizing emotion from a piece of text document or a blog, any human can do this better than a machine only problem is he/she takes time. Proposed emotion detector system takes a text document and the emotion word ontology as inputs and produces one of the six emotion classes (i.e. love, sadness, joy, fear and surprise, anger) as the output. Every input text contains some short stories which are firstly read and assigned an emotion class manually and then that emotion class is compared to the output of the proposed system to check the accuracy of the Proposed Emotion Detector System. It is found that the Proposed Emotion Detector System produces output with the accuracy of more than 75%. Keywords—Human-Computer Interaction; Textual Emotion Recognition; Emotion Word Ontology I. INTRODUCTION Human emotion recognition by analyzing written documents appear challenging but many times essential due to the fact that most of the times textual expressions are not only direct using emotion words but also result from the interpretation of the meaning of concepts and interaction of concepts which are described in the text document. Emotion detection from text plays a key role in the human-computer interaction [1]. Human emotions may be expressed in many ways like person’s speech, face expression and written text known as speech, facial and text based emotion respectively [14]. In human computer interaction, human emotion recognition from text is becoming increasingly important from an applicative point of view. Methods being used for text based emotion detection are classified into keyword spotting technique, lexical affinity method, learning based method and hybrid approach however each method has its own limitations [19]. A proposed architecture which contains the emotion ontology and emotion detector algorithm is explained in Section 3. In section 4, algorithm is implemented and results are shown. Conclusion is given in Section 5. II. RELATED WORK The role of human computer interaction is proposed by Picard in the concept of affective computing [3]. Many researchers from computer science, biotechnology, psychology, and cognitive science are attracted by this domain. From another point of view, research in the field of emotion detection from textual data emerged to determine human emotions. Emotion detection from text can be formulated as follows: Let A be the set of all authors, T be the set of all possible representations of emotion-expressing texts, and E be the set of all emotions. Let r be a function to reflect emotion e of author a from text t, i.e., r: A x T E, then the function r would be the answer to the problem [4]. The concept of emotion recognition systems lies in fact that, although the definitions of T and E may be straightforward, the definitions of individual element, even subsets in both sets of T and E would be rather confusing. As the languages are constantly emerging new elements may add on one side, for the set T. Due to the complex nature of human minds, any emotion classifications can only be seen as “labels” annotated afterwards for different purposes, whereas on the other side, currently there are no standard classifications of “all human emotions”. Methods used for text based emotion recognition system [4], [5] are: A. Keyword Spotting Technique The keyword spotting technique can be described as the problem of finding occurrences of keywords (love, anger, joy, sadness, surprise and fear) from a given text document. Many algorithms to analyze sentiment or emotion have been suggested in the past. In the context of emotion detection this method is based on certain predefined keywords. These emotion words are categorized into keywords such as disgusted, sad, happy, angry, fearful, surprised etc. Occurrences of these keywords can be found and based on that an emotion class is assigned to the text document. B. Lexical Affinity Method Detecting emotions based on related keywords is an easy to use and straightforward method. Keyword spotting technique is extended into Lexical affinity approach which assigns a