SENTIMENT ANALYZER USING MACHINE LEARNING PRESENTED BY: K.NAGA LOHITHA (15071A05K3) P.VARUN KUMAR (15071A05L5)
S.YESHWANTH
(15071A05M5)
E.SNIGDHA
(15071A05N8)
PROJECT GUIDE: DR. A.BRAHMANANDA REDDY
ASSOCIATE PROFESSOR/CSE
AGENDA • ABSTRACT
• USE-CASE DIAGRAM
• INTRODUCTION
• ACTIVITY DIAGRAM
• PROBLEM DEFINITION • COMPARATIVE STUDY • SYSTEM ARCHITECTURE • DESIGN METHODOLOGY
• CLASS DIAGRAM • SEQUENCE DIAGRAM • PLAN OF ACTION
• CONCLUSION • REFERENCES
ABSTRACT Sentiment analysis is the process of determining the emotional tone behind a series of words, used to gain an understanding of the attitudes, opinions and emotions expressed within an online mention. It is a contextual mining of text which identifies and extracts subjective information in source material. Depending upon the analysis the text is classified as positive, negative or neutral. Sentiment analysis is extremely useful in social media monitoring as it allows us to gain an overview of the wider public opinion behind certain topics.
INTRODUCTION Sentiment analysis is the process of computationally identifying and categorizing opinions from piece of text and determine whether the writer's attitude towards a particular topic is positive, negative or neutral. We can determine the attitude by mining the text in a sentence or a document.
PROBLEM DEFINITION To build a machine learning model such that it analyses the emotion of a person by processing their tweets in Twitter. The emotion could be strongly positive, weakly positive, neutral, weakly negative or strongly negative.
COMPARATIVE STUDY EXISTING SYSTEM As the tweets are processed, the output obtained will be specifying the positivity, negativity and neutrality of the tweet that has been processed through the algorithm.
PROPOSED SYSTEM The result obtained would be classified as neutral, strongly positive ,positive ,weakly positive and same is the case with negative. The entire output is also represented in the form of a “pie chart” which depicts the percentage of each output.
SYSTEM ARCHITECTURE
DESIGN METHODOLOGY This project has been divided into 2 phases. • First, literature study is conducted, followed by system development. Literature study involves conducting studies on various sentiment analysis techniques and method that currently in use. • In phase 2, application requirements and functionalities are defined prior to its development. Also, architecture and interface design of the program and how it will interact are also identified. In developing the twitter sentiment analysis application, several tools are utilized, such as python shell 2.7.2 and notepad.
USE-CASE DIAGRAM System Register
Login
Input/Keywords
TweetsRetrivel
PreProcessing User
SentimentDetection
Analysis of Output
Logout
ACTIVITY DIAGRAM Register entry/register with details User Login entry/login with details
Input/Keywords do/Enter the user name of tweets
TweetsRetrivel do/Enetr the value for getting number of tweets
PreProcessing do/Remove the stopwords stemmind words
SentimentDetection do/get the polarity of given sentence
AnalysisOfOutput do/Display the results as positive,negitive,nutral
Logout exit/user logout
CLASS DIAGRAM Input/Keywords Register
TweetsRetrivel
Login
SentimentDetection
User AnalysisOfOutput
PreProcessing
Logout
SEQUENCE DIAGRAM Register
Login
Input/Keywords
TweetsRetrivel
PreProcessing
SentimentDetection
: User 1 : register with detailes() 2 : login with details() 3 : Enter the user name of tweets()
4 : Enetr the value for getting number of tweets()
5 : Remove the stopwords stemmind words()
6 : get the polarity of given sentence()
7 : Display the results as positive,negitive,nutral() 8 : user logout()
AnalysisOfOutput
Logout
PLAN OF ACTION STATUS OF WORK: CODING PHASE TIMELINE FOR FURTHER SDLC ACTIVITIES:
• CODING
28-02-2019
• TESTING
05-03-2019
• IMPLEMENTATION
10-03-2019
• DOCUMENTATION
20-03-2019
• PAPER PUBLICATION
30-03-2019
CONCLUSION Sentiment analysis can be used in blogs, articles, product reviews, social-media websites where a third person narrates his/her views. The project aims at classifying the comments based on the type of tweet made. This project, when deployed in social media platforms helps to propose a methodology through which sentiments can be analyzed.
REFERENCES [1] Maxim Stankevich, Vadim Isakov, Dmitry Devyatkin And Ivansmirnov(2017).Feature Engineering For Depression Detection In Social Media
[2]https://www.newgenapps.com/blog/the-secret-way-of-measuring-customer-emotions-social-media-sentimentanalysis[
[3]https://marcobonzanini.com/2015/05/17/mining-twitter-data-with-python-part-6-sentiment-analysis-basics
[4] https://link.springer.com/chapter/10.1007/978-3-319-49586-6_59
THANK YOU