Sentiment analysis for electronic products based on youtube comments | |
| Author | Vishnu, Thulasi |
| Call Number | AIT RSPR no.ICT-19-04 |
| Subject(s) | Machine learning Natural language processing (Computer science) Social networks |
| Note | A research submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Information and Communication Technologies, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Research studies project report ; no. ICT-19-04 |
| Abstract | Since the evolution of World Wide Web, Internet is reaching to more and more people every day. Website like Amazon, Flipkart, Lazada have been on rise when it comes to selling products online, while it is much easier for consumer to buy, the authenticity and quality of these products are always truly tested. While product reviews give benefit of the doubt, it can also be quite overwhelming for the user to grasp. YouTube, in this context, has been standard video sharing platform that is generally used to state one’s opinion (reviews) about the product. The comment section give the most concise review by bigger audience. The video and the comments about a particular electronic product are sometimes limited to particular populace that reflects if it is ideal fit for consumer. We consider these immense comments about a product to be standard point of view for the product and perform sentiment analysis. Since it is difficult to know what people are thinking about that video from the comments of that video. For popular videos large number of comments are present, most of them being spam and inappropriate comments. To overcome this issue, we present a natural language processing (NLP) based sentiment analysis approach on user comments. This analysis helps to find what the general opinion about the video is whether positive or negative. Here we perform sentiment analysis on comments using Classifier, which gives the best accuracy among the four classifiers (Naïve Bayes, Support Vector Machine, Logistic regression and Stochastic Gradient Descent), accuracy in terms of classifying positive or negative comments, the best classifier with high accuracy is determined. Using the best classifier comments are classified into positive or negative and tells about the general sentiment of the product. |
| Year | 2019 |
| Corresponding Series Added Entry | Asian Institute of Technology. Research studies project report ; no. ICT-19-04 |
| Type | Research Study Project Report (RSPR) |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Information and Communication Technology (ICT) |
| Chairperson(s) | Teerapat Sanguankotchakorn; |
| Examination Committee(s) | Vatcharaporn Esichaikul;Chutiporn Anutariya; |
| Scholarship Donor(s) | Asian Institute of Technology Fellowship; |
| Degree | Research Studies Project Report (M. Eng.) - Asian Institute of Technology, 2019 |