Machine Learning Approach to Analyze Classification Result for Twitter Sentiment
- Resource Type
- Conference
- Authors
- Kalaivani, P; Dinesh, D
- Source
- 2020 International Conference on Smart Electronics and Communication (ICOSEC) Smart Electronics and Communication (ICOSEC), 2020 International Conference on. :107-112 Sep, 2020
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Power, Energy and Industry Applications
Support vector machines
Sentiment analysis
Classification algorithms
Neural networks
Feature extraction
Twitter
Social data
Tweets
Opinion
Back Propagation
Supervised Learning
Support Vector Machine
- Language
Monitoring of social media has been growing day by day and analyzing social data helps in identifying behavior of people. News and media play a biased role in conveying a particular incident or policies. Thus the analysis of social media data such as twitter comments uses sentiment analysis that examines the opinion of people on certain government policy declared by the central government. In this paper, twitter sentiment analysis using SVM and NN has been proposed that analyzing the twitter dataset of particular policies and finding its polarity of sentiment. With a quick increase in the use of the Internet, people can raise out their opinion on social media about a particular topic. This work compares the accuracy between the two classification algorithms namely Back Propagation Neural Network (BPNN) and Support Vector Machine (SVM). These two algorithms are supervised algorithms, which mean that the model trains itself with the use of a training dataset and performs testing with the help of the testing dataset. This work proposes the ANN and SVM classification algorithm which is used to classify the emotion behind the text, which can achieve acceptable accuracy in classifying results. It is trained over five hundred epochs. The resulting confusion matrix is obtained by comparing predicted with actual labels. Experiments demonstrate that Support Vector Machine outperforms the accuracy when compared with BPNN.