International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-9 Issue-1S3, December 2019 57 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: A10121291S319/2019@BEIESP DOI:10.35940/ijeat.A1012.1291S319 Abstract: Sentiment analysis, also known as Opinion Mining is one of the hottest topic Nowadays. in various social networking sites is one of the hottest topic and field nowadays. Here, we are using Twitter, the biggest web destinations for people to communicate with each other to perform the sentiment analysis and opinion mining by extracting the tweets by various users. The users can post brief text updates in twitter as it only allows 140 characters in one text message. Hashtags helps to search for tweets dealing with the specified subject. In previous researches, binary classification usually relies on the sentiment polarity(Positive , Negative and Neutral). The advantage is that multiple meaning of the same world might have different polarity, so it can be easily identified. In Multiclass classification, many tweets of one class are classified as if they belong to the others. The Neutral class presented the lowest precision in all the researches happened in this particular area. The set of tweets containing text and emoticon data will be classified into 13 classes. From each tweet, we extract different set of features using one hot encoding algorithm and use machine learning algorithms to perform classification. The entire tweets will be divided into training data sets and testing data sets. Training dataset will be pre-processed and classified using various Artificial Neural Network algorithms such as Reccurent Neural Network, Convolutional Neural Network etc. Moreover, the same procedure will be followed for the Text and Emoticon data. The developed model or system will be tested using the testing dataset. More precise and correct accuracy can be obtained or experienced using this multiclass classification of text and emoticons. 4 Key performance indicators will be used to evaluate the effectiveness of the corresponding approach. Keywords: Multiclass Sentiment Analysis, Data Pre-processing, Natural Language Processing, Feature Extraction, Classification, Emoticons, Neural Networks I. INTRODUCTION 1.1 Sentiment Analysis Sentiment Analysis of various Social Media is considered as one of the hottest topic nowadays. Here, sentiment analysis was done on the famous social media, Twitter where the data was downloaded for the analysis. Nowadays, Sentiment Analysis is an emerging trend as a lot of organizations or institutions following this procedure to understand the views and opinions of various people[2] . For example, the usage of a particular product can be analyzed by Revised Manuscript Received on December 04, 2019 * Correspondence Author Nirmal Varghese Babu, Computer Science and Engineering Department, Amal Jyothi College of Engineering, Kanjirappaly, India, nirmalvarghesebabu@cs.ajce.in Fabeela Ali Rawther, Computer Science and Engineering Department, Amal Jyothi College of Engineering, Kanjirappaly, India, fabeelaalirawther@amaljyothi.ac.in the way people respond to it[1]. Various Neural Network Algorithms were used for the classification procedure. the inner meaning of the texts and emoticons can be identified using this procedure. Multiclass Sentiment Analysis provides more precise and accurate accuracy values when compared to the existing methods like Binary and Ternary Classifications. Here, the data will be classified into more than 3 classes so that the deep meaning of each and every word will be found out. Emoticons also plays an important role in the multiclass sentiment analysis. The emoticons can also be classified as same as the texts. For that, the meaning of the emoticons must be found out. The accuracy can get reduced in the multiclass analysis as the classification is getting more intense and accurate. A single word can represents different emotions or feelings but in Multiclass Classification, this problem can be reduced. Various Hashtags were used to retrieve the required data from the various Social Media using the keyword Matching process, for example #twitter. The maximum word count of data(texts and emoticons) which can be posted online ie tweets is 140. The pre-processing[3] steps like the removal of URL‟s, Special symbols, Full stops, Stop words etc. were performed where the unimportant or not useful data will be removed[4]-[20]. Tokenization was done to tokenize or categorize the words from the available dataset. Stemming was also performed to remove the suffixes and prefixes from the tokenized words. Various Natural Language tools or packages are imported for this procedure. One Hot Encoding was used here for the Feature Extraction Procedure. Here, the data will be encoded based on the available data in the unique array created from the datasets. The data will be encoded into either 1 or 0. The encoded data will be given as input to various Neural Network Algorithms like Recurrent Neural Networks, Convolutional Neural Networks, Recurrent Convolutional Neural Networks and Convolutional Neural Networks. The created model was trained using these algorithms and was tested based on the corresponding Sentiments of each tweets. The Accuracy of each model was found out and more values were predicted. Here, the data was classified into 13 sentiment classes. II. PROPOSED SYSTEM This Particular System contains various steps which include: - Data Collection - Data Pre-Processing - Feature Extraction - Text – Emoticon Analysis - Classification - Results Evaluation Multiclass Sentiment Analysis of Social Media Data using Neural Networks Nirmal Varghese Babu, Fabeela Ali Rawther