NOVEL SYSTEM, SENTIMENT ANALYSIS ON TWITTER USING MAPREDUCE AND NAIVE BAYES

Abstract

ABSTRACT Of; THE INVENTION Today, technology plays a major role in driving global communication. Social media sites like Facebook, lnstagram, and Twitter are the most significant forums for sharing opinions about the changes that are occurring in and across the world on a daily basis.Twitter is a valuable source of information for analyzing user· sentiment. This work embodies the notion of considering user. opinions in sentiment and emotion analysis as well as in drawing conclusions on relevant topics through the use of machine learning algorithms. Using supervised learning, Naive Bayes and Support Vector Machines in machine learning are fine-tuned to produce outputs for sentiment and emotion analysis, respectively. Sentiment analysis aims to extract sentiment polarity {positive or negative), while emotion analysis uses user data to extract specific emotions {such as empty, sad, angry, etc.). This kind of study basically acts as a portal for consumer needs and generates growth.

Specification

DESCRiPTiON·
Social media platforms like Twitter have revolutionized communication by allowing
users to share opinions and experiences in real-time. This has led to the g!')neration qf va~t <!mounts of
user-generated content, presenting both opportunities and challenges for analysis. Sentiment analysis,
or opinion mining, is crucial for applications like brand monitoring and reputation management.
However, Twitter data presents challenges like limited text length, slang, misspellings, and emojis.
MapReduce and Naive Bayes are scalable and efficient approaches for sentiment analysis, enabling
the analysis of large-scale Twitter datasets to extract valuable insights.
This study aims to create a scalable sentiment analysis system using MapReduce and Naive Bayes
for Twitter data analysis.lt will implement efficient data preprocessing, feature extraction, model training, and
sentiment prediction algorithms. The system will be evaluated for scalability, accuracy, and effi~iency in
handling large-scale Twitter datasets.
Th~ project aims to analyze sentiment in Twitter data using distributed computing techniques and
machine learning algorithms. It aims to handle large-scale datasets efficiently, utilizing the MapReduce
fr(!m!')~ork for distributed processing. The project also employs the Naive Bayes classifier fpr s!'lntiment
classification. The project's main goal is to develop a scalable sentiment analysis system that can analyze
sentiment across diverse topics, time periods, and geowaphical r~ions. The system will enable real-tinie
analysis, providing actionable insights for businesses, marketers, researchers, and policymakers. The project
contributes to research in sentiment analysis methodologies? distributed computing techniques? and their
applications in social media analytics and opinion mining.
0-. --.··- zr'P.,t!=Nt t;trrrcc C:\iCNNI\I 1 2 • 4 4
I ..........
-Q)
C)
Ill
D..
Q)
-1- N
E....
0
-LL. N
M
"("0'"
CIO
0 .....
-::1'
-::1'
N
0
~
N
(0
M
1.1')
M -..... -::1'
N
0
~
>
Novelty:
• Sentiment analysis in twitter.
• {)tiliz~~ MlipReduct; frlimt;W()rk f()r effi<;i~nt pr<Jcessing ()fllirgt;"~<;lilt; Twint;r 9!!~!\~t;~s,
• Uses Naive Bayes classifier for accurate sentiment classification.
Advantages:
• Analysis of large-scale Twitter datasets to extract valuable insights into public sentiment
and ()Pi!l_iO.f.i dy_f!a!TIJcs, Qood i!"!fO!!l!atiO.J:l data bas~
• Robust system capable of accurately analyzing sentiment in real-time Twitter data
streams.
S.CLAIMS
We claim,
I. The Novel System, for sentiment analysis on twitter
2. 5 Softwares are used
3. Twitter Sentiment Analysis System
• Configuring ·development ·environment.
• Collecting and preprocessing Twitter data.
• Optimizing MapR,ed11ce and Naive ~ayes algorithms.
4. Computing techniques to process massive volumes of tweet data.

Documents