Sentiment Classification in Hindi by using a HindiSentiWordNet (HSWN) Pooja Pandey*, Sharvari Govilkar** and Sagar Kulkarni*** *Department of Computer Engg. PIIT, New Panvel, India pooja.clx@gmail.com **Department of Computer Engg. PIIT, New Panvel, India sgovilkar@mes.ac.in ***Department of Computer Engg. PIIT, New Panvel, India skulkarni@mes.ac.in Abstract: Sentiment Analysis is a natural language processing task that deals with finding orientation of opinion with respect to a given topic. It deals with analyzing emotions, feelings, and the attitude of a speaker or a writer from a given piece of text or any other form of media. The target of sentiment classification system is to find opinions, identify the sentiments they express, and then classify them according to their polarity. The proposed system for sentiment classification of Hindi documents uses Hindi SentiWordNet (HSWN) to find the overall polarity of the Hindi text document where the final aggregated polarity calculated by system can be positive, negative or neutral. Existing HSWN is enhanced by adding more number of sentiment bearing words. The proposed System also handles negation and discourse relations which influence sentiment associated with a given input. Keywords: Sentiment Analysis (SA), SentiWordNet, HindiSentiWordNet (HSWN), Polarity, Synset Replacement., Natural language processing (NLP). Introduction Sentiment Analysis is a task under natural language processing which finds orientation of a person opinion or feelings over an entity [1]. It deals with analyzing personal emotions, feelings, attitude and opinion of a speaker or a writer over an object. The primary target of SA is to find the sentiments expressed by person over an information or entity [2]. Sentiment analysis helps to find sentiment associated with the given input which can be in the form of single line or paragraph or a full document about a given subject. SWN consists of words present in specific language with its associated polarity. For the given input overall polarity or sentiment can be calculated by extracting and aggregating polarity of each sentiment word in the input. There are different classification levels in SA: document-level, sentence-level and aspect-level. Document-level SA aims to classify an opinion of the whole document as expressing a positive or negative sentiment. Sentence-level SA aims to classify sentiment expressed in each sentence which involves identifying whether sentence is subjective or objective. Aspect-level SA aims to classify the sentiment with respect to the specific aspects of entities which is done by identifying the entities and their aspects. The paper presents, sentiment analysis system in Hindi language where overall sentiment is classified as positive or negative. In section 2, proposed system is discussed in detail. Working of system is mentioned in detail in section 3. Section 4 explores accuracy obtained by system. Finally, paper is concluded in section 5. Related Work In this section we cite the relevant past literature of research work done in the field of sentiment analysis for Hindi language. Namita mittal et al [1] developed an efficient approach based on negation and discourse relation to identifying the sentiments from Hindi content . They developed an annotated corpus for Hindi language and improve the existing Hindi SentiWordNet (HSWN) by incorporating more opinion words into it. Aditya Joshi and Pushpak Bhattacharyya [2] proposed a fallback strategy for Hindi language. Authors proposes use of, In- language Sentiment Analysis, Machine Translation and Resource Based Sentiment Analysis to find sentiment in Hindi text. Hindi SentiWordNet (HSWN) was developed using two lexical resources (English SentiWordNet and English-Hindi WordNet Linking .78.14% accuracy was obtained using SVM classifier for in-language sentiment analysis. Akshat Bakliwal and Piyush [6] present a method of building a subjective lexicon for Hindi. Authors discussed a method of building a subjective lexicon for Hindi. Using WordNet and Breadth First Graph traversal method, they construct the subjectivity lexicon. Main contribution of their work is developing a lexicon of adjectives and adverbs with polarity scores