Scribd
Upload
71 views17 pages

Project On Sentimental Analysis: Submitted by

The document summarizes a project on sentiment analysis submitted by Akshay Bhoju kothari, Dhanush shetty, and H K Nakul. It discusses machine learning and sentiment analysis, how sentiment analysis works by collecting data, analyzing it using algorithms, indexing sentiments, and delivering results. It also describes problems faced like topic differences and sarcasm undermining accuracy, and future applications in recommender systems and improving products based on customer reviews.

Uploaded by

Nakul hk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
71 views17 pages

Project On Sentimental Analysis: Submitted by

The document summarizes a project on sentiment analysis submitted by Akshay Bhoju kothari, Dhanush shetty, and H K Nakul. It discusses machine learning and sentiment analysis, how sentiment analysis works by collecting data, analyzing it using algorithms, indexing sentiments, and delivering results. It also describes problems faced like topic differences and sarcasm undermining accuracy, and future applications in recommender systems and improving products based on customer reviews.

Uploaded by

Nakul hk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 17

Project on

Sentimental analysis

submitted by-
Akshay Bhoju kothari
Dhanush shetty
H K Nakul
Agenda:-
1.Machine Learning
2.Sentimental analysis
3.Project implementation
4.Problem faced
5.Future scope
6.Reference
Machine learning
➢ Machine learning is a method of data analysis that automates
analytical model building

➢ focuses on the development of computer programs that can


access data and use it to learn for themselves.
Eg:Prediction,Recommendation
Sentiment Analysis
Using machines to analyze big data and produce incites for proper
decisions.
What is sentiment analysis ?
The process of identifying and categorizing opinions expressed in a piece of
text, especially in order to determine whether the writer's attitude towards a
particular topic, product, etc. is positive, negative, or neutral.

Traditionally, most of the research in sentiment analysis has been aimed at


larger pieces of text, like movie reviews, or product reviews.
Ex:we get emojis for every posts on facebook ..
Sentiment Analysis:
How it Works-

Collect Data Analyze Data Indexing Delivery

Algorithms process Algorithms tag Provides an


Social Media, Blogs,
the data and sentences based on output reporting
Twitter, News,
performs sentence polarity and the outcome of
Product Reviews
splitting intensity sentiments the analysis

Compiled by author from sources: researchgate.net


Sentiment Analysis:
How it Works-

 Data Collection

 Public sentiments from consumers expressed on public


forums and on social network are collected

 Opinions and feelings are expressed in different way, with


different vocabulary, context of writing, usage of short
forms and slang, makes data huge and disorganized
Sentiment Analysis:
How it Works-

 Analyze Data

 Text Preparation
 Data is extracted and filtered before analysis
 Non-textual content and content is identified and eliminated if it is
irrelevant

 Sentiment Detection
 Each sentence and opinion is examined for subjectivity
 Sentences with subjective expressions are retained and ones that convey
objective expressions are discarded
Sentiment Analysis:
How it Works-
 Indexing (Sentiment Classification)

Sentiments can be broadly classified into two groups, positive and


negative
Each subjective sentence is classified into positive, negative,
good, bad, like, dislike

 Delivery (Presentation of Output)

The result of converted unstructured text into meaningful


information
Usually displayed as graphs for easy interpretation
Design and implementation of the
model
➢ Initially we spent more time in data preprocessing

➢ After getting the proper data we’ve worked and did a lot of research in
finding the proper features.

➢ For classifying the sentiments also we have used many algorithms


Sentiment Analysis:
What Can it do for Brands?

Resolve Customer Experience


Increase Customer Retention Pain Points Optimize Customer Service

Measure Social Media ROI


Optimize Pricing

m
Sentiment Analysis:
What Can it do for Brands?

Sentiment Analysis can help


answer these questions:

How do the company’s


policies or external
What do customers Are customers happy What do customers like
events or its employees
think of the products with the services they about the brand’s
impact customers’
and brand? receive? competitors?
perception of the
brand?

Compiled by author from sources: analyticstraining.com


Sentiment Analysis:
Problems-
Emerging
Terms
Topic
Categories Negation
Differences

Sentence
Completeness Issues Sarcasm

Undermining
Accuracy:
Lack of Word Roots

Context (Stemming)

Semantic Term
Frequency
Ambiguity
Sentiment Analysis:
Problems Faced-

➢ Reducing 1.6M tweets to 20K tweets was little difficult


➢ Removing the tags punctuation websites stopwords also took
more time.
➢ Training algorithm selection was also a time consuming phase
➢ While creating a dashboard, we had faced a problem because
we were not familiar with HTML
Future scope

➢ Online shopping brands can use this model to improve their


products based on the customers reviews.
➢ It will be very useful in recommender system like movie,
restaurant.
Reference

1)
https://drive.google.com/file/d/13U_bQUY0vwxoRiLjCVfp_wKqn8zCgQOy/v
iew?usp=sharing

2)https://ieeexplore.ieee.org/document/7066632

3)https://www.kaggle.com/datasets

4)https://youtu.be/1gQ6uG5Ujiw
THANK YOU

You might also like