Sentiment Analysis (SA) is an application of Natural Language Processing (NLP) to extract the sentiments expressed in the text. In this paper, we experimented five approaches to perform SA, namely, Lexicon based approach, Supervised Machine learning based approach, Hybrid approach, K-means with Bag of Word (BoW) approach and K-modes with BoW approach. We have experimented these approaches using five corpora with different feature representation techniques to predict the best approach to perform SA in Tamil texts. In this research we used Basic features such as word count and punctuation count in addition to traditional features such as Bag of Words (BoW) and Term Frequency-Inverse Document Frequency (TF-IDF) included to check their influence in the prediction. We have compared these approaches, features and the corpora. From the evaluation the highest accuracy of 79% is obtained for UJ_Corpus_Opinions_Nouns corpus with fastText for supervised Machine learning based approach.

Sentiment Analysis in Tamil Texts: A Study on Machine Learning Techniques and Feature Representation

Detecting face quiet have issues in managing pictures in the bare because of huge presence dissimilarities. Rather than exit look differences straightforwardly to measurable knowledge calculations, we suggest a various leveled (body-part) portion centered Essential replica to unequivocally catch them. This replica empowers part subtype alternative to deal with nearby appearance variations, for example, shut and exposed mouth, and body-portion distortion to catch the worldwide look differences, for example, posture and articulation. In recognition, applicant frame is fit to the Essential replica to surmise the portion area and portion subtype, and finding total is then processed in light of the fitted arrangement. Thusly, the impact of appearance variety is diminished. Other than the face replica, we abuse the co-event amongst body and face, which handles huge variations, for example, substantial impediments, to additionally support the face detection execution. We display an expression based portrayal for body detection, and propose an Essential setting replica to together encrypt the yields of body and face detector.

Essential-Replica for Face Detection in the Large Appearance Variations

Sentiment Analysis (SA) is an application of Natural Language Processing (NLP) to analyse the sentiments expressed in the text. It classifies into categories of qualities and opinions such as good, bad, positive, negative, neutral, etc. It employs machine learning techniques and lexicons for the classification. Nowadays, people share their opinions or feelings about movies, products, services, etc. through social media and online review sites. Analysing their opinions is beneficial to the public, business organisations, film producers and others to make decisions and improvements. SA is mostly employed in English language but rare for Indian languages including Tamil. This review paper aims to critically analyse the recent literature in the field of SA with Tamil text. Objectives, Methodologies and success rates are taken in consideration for the review. We shall conclude from the review that SVM and RNN classifiers taking TF-IDF and Word2vec features of Tamil text give better performance than grammar rules based classifications and other classifiers with presence of words, TF and BoW as features.

/pdf/review-on-sentiment-analysis-in-tamil-texts-1ia9t8fngm.pdf

Review on sentiment analysis in tamil texts

Mostly sentiment analysis employs dictionary approaches for recognizing the polarity of terms in a review. However, in sentiment analysis between different domains called domain adaptation (DA), the sentiment lexicon disappoints that leads to the feature mismatch problem. Now, many e-commerce sites try to process reviews in their native languages. In this paper, we propose an enhanced dictionary in our native language (Tamil) that aims at building contextual relationships among the terms of multi-domain datasets that tries to minimize the feature mismatch problem. The proposed dictionary employs both labeled and unlabeled data from the source domain and unlabeled data from the target domain. More precisely, the initial dictionary explores pointwise mutual information for calculating contextual weight then the final dictionary estimates the rank score based on the importance of terms among all the reviews. This work intends to classify reviews of multiple target domains in Tamil by using the unified dictionary with a large number of vocabularies. This extendible dictionary significantly improves the accuracy of DA with the other baseline methods and handles many words in multiple domains with ease.

An enhanced sentiment dictionary for domain adaptation with multi-domain dataset in Tamil language (ESD-DA)

Syllabication is essential in the preprocessing stage of speech systems. In the context of the Philippines' Cebuano-Visayan language's syllabication rules, the existing rules do not include hyphenated words although the hyphen defines the syllable boundary in a word. Hence, this study created grammar rules for hyphenated words which include sequences of a hyphen between vowel-consonant, consonant-consonant, vowel-vowel, and consonant-vowel. The test was done for the enhanced grammar rules for Cebuano-Visayan syllabication with 1,465 representative hyphenated and non-hyphenated words of varying lengths. The result further implies that the syllabication analysis for hyphenated words showed that hyphens improve the naturalness and intelligibility in the utterance of the words, thereby enhancing the understanding and comprehension of the Cebuano-Visayan discourse.

Rules for Orthographic Word Parsing of the Philippines’ Cebuano-Visayan Language Using Context-Free Grammars

The widespread of social media is growing every day where users are sharing their opinions, reviews, and comments on an item or product. The aim is to develop a model to mine user tweets collected from Twitter. In this paper, our contribution on user tweets to find the sentiments expressed by users about Tamil movies based on the grammar rule. Tamil movies domain is selected to confine our scope of the work. After preprocessing, N-gram approach is applied to classify tweets into different genres. This work intends to find the polarity of Tamil tweets in addition to genre classification. In this work, it is also shown how to collect user tweets which comes as data stream using modified N-gram approach to predict the sentiments of the users in the dataset. Results suggest that N-gram model not only remove the complexity of natural language process but also help to improve the decision-making process.

Grammar Rule-Based Sentiment Categorization Model for Tamil Tweets

The advent of social media has enabled people to easily and publicly express their ideas on a movie/product in such a way that it reaches millions of people within no time. This research aims to implement a tool that would be helpful in predicting the genre of the movies as perceived by the audience through linguistic rules and natural language processing (NLP) tool kit. This paper focuses on development of rule-based sentiment categorising tool for Tamil tweets and a tool has been developed using Python and NLP tool kit. Furthermore, a model is designed to determine the opinion along with genre classification of Tamil movies. For this work, a set of genres are selected from Tamil movies with public tweets based on sentiment analysis. We find that the tool classifies the genre of a particular movie provided by user tweets and validated our approach with domain experts and baseline models.

Nadana Ravishankar

Papers

Grammar Rule-Based Sentiment Categorization Model for Tamil Tweets

Grammar rule-based sentiment categorisation model for classification of Tamil tweets