The user emotion analysis has been a great deal, Here primarily, the twitter dataset is used for analysis. [6] In the current trend, twitter application usage is getting drastically increased where people share their thoughts with each other throughout the world. Generally people use social media as a platform to express their opinion to others, In the meantime Facebook and Twitter serve as the major source for them. Hence they make use of this when needed. The volume of the tweets remain quite high since it is a collection of so many tweets throughout the world, and also it consists of both positive and negative tweets, For the purpose of analysis it is necessary to find the positive and negative tweets and categorize them accordingly. In this project the input dataset consists of a compilation of a number of tweets which expresses the thoughts of various people regarding the election results and their own thoughts about different election candidates who compete in the election, correspondingly each person may have varying opinion on each candidate which will be positive or negative. The analysis of these tweets will help in recording a small survey over those different candidates. The dataset consists of more than 2000 records. The analysis is being done with the help of python programming by importing various packages and implementing a number of functions required for analysis. The results are obtained in two different categories namely positive and negative. Multinomial Naive Bayes classifier has been used for the purpose of analysis. The work is further extended and tested for the imdb movie review dataset, mobile review dataset and the results are categorized as positive and negative.
Analysis of user emotions and opinion using Multinomial Naive Bayes Classifier
2019-06-01
1048524 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Prediction of Slope Stability using Naive Bayes Classifier
Online Contents | 2018
|Prediction of Slope Stability using Naive Bayes Classifier
Springer Verlag | 2018
|Improving Naive Bayes for Classification
British Library Online Contents | 2010
|