International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 05 | May 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 6920 SENTIMENT ANALYSIS TEXT TYPE FEEDBACK SYSTEM USING ARTIFICIAL INTELLIGENCE Cibi Chandra, Anil Kumar PG Scholar, Dept. of Computer Science and Engineering, SVS college of Engineering Coimbatore, Tamilnadu, India Assistant Professor, Dept. of Computer Science and Engineering, SVS college of Engineering Coimbatore, Tamilnadu, India ---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - Feedback is used to analyze the performance in various organizations like academic, agricultural and many other organizations. At present most of the systems use grading technique to evaluate the performance. But this method always does not reveal the sentiments of the reviewer. For example, in students’ feedback in an academic system, if grading is used it does not reveal the student sentiments towards a particular individual. The same scenario exists in agricultural field also. The farmers can’t give their feedback about the agricultural Rice yield present in the country through grading. Thus, textual feedback is always found to be effective. In this work we propose a textual feedback to analyze the sentiments of the reviewer in agriculture as well as academic system. Sentimental analysis is a method for identifying the sentiment expressed in texts. The need of Sentiment Analysis of text has gained more importance in today’s situations faced by the people of the world. In machine learning technique, it uses unsupervised learning or supervised learning. Classification problem can be carried out using several algorithms like support vector machine, naïve Bayes, random forest. In lexicon-based method sentiment polarity of the textual content is detected using sentiment lexicon. A lexicon is a list of words with associated sentiment polarity. Hybrid approach is a combination of lexicon-based and machine learning methods. The training data set is labelled using sentiment lexicon and this is used to for the machine learning model. Then testing data is evaluated using this model. MATLAB software is used for this setup Key Words: Naïve Bayer, Support Vector Machine, 1.INTRODUCTION Social Media and Micro blogging platforms like Face book, Twitter, and Tumbler dominate in spreading encapsulated news and trending topics across the globe at a rapid Space. Large organizations take feedback about their products and services to increase their marketing Sites Twitter, Face book, Instagram, google+ offer a platform to people to voice their opinions. people quickly post there. This type of vast information on these sites can use for marketing and social studies. Therefore, sentiment analysis has wide applications and includes emotion mining, polarity, and classification and influence analysis. Feedback is the statement sent to an entity about its past behavior from which the entity can analyze the future and current behavior to achieve the expected result. Feedback plays an important role in education and learning by helping to adopt new knowledge and prevent repetitive mistakes. Feedback is a process which helps the organization to monitor, evaluate, and regulate the overall working environment. Good feedback practice provides useful information to the organization in improving the teaching and learning experience. Similarly, the agricultural field can be improved by taking regular feedback from farmers. In a more practical sense, our objective here is to take a text and produce a label (or labels) that summarize the sentiment of this text, e.g. positive, neutral, and negative. For example, if we were dealing with agricultural reviews, we would want the sentence ‘the incentives to help farmers were helpful ‘to be labeled as Positive, and the sentence ‘The amount for relief to farmers in budget is not enough ‘is‘ labeled as Negative. Depending on the feedback given by students or from famers in agricultural field it can be classified as textual or grading (Liker-scale based score) form. In Liker-scale based score questions are provided to the students and are asked to answer those questions using a rating-based scale. the sentiment of the student is known using textual feedback technique. In this textual form student are given with set of questions and they need to answer it in sentences. It is helpful to both the academic administration and instructor to overcome the issues related to their organization. In this paper, the student feedback with varied opinion is collected. The aim is to extract expressions of opinion and classify it as negative, positive or neutral using machine learning techniques. Similarly, in agricultural field the feedbacks are collected from farmers by government officials. Various necessities of farmers are identified by analyzing these feedbacks. In this work the data is collected from farmers using questioners. Methods like Artificial neural network, machine learning algorithms such as Support Vector Machine, Multinomial Naïve Bayes Classifier, and Random Forest are used for classifications. The results of classifiers are compared to find the best method for classification. The result shows that Artificial Neural Network has better performance over the other methods.