Scribd
Upload
3 views

Craft Presentations

The document presents a project on sentiment analysis using machine learning and deep learning techniques, focusing on analyzing text data for sentiment understanding. It outlines the problem of unstructured data growth and the objectives of fine-tuning large language models. The findings indicate that DistilBERT outperforms traditional models like SVM, Naive Bayes, and Logistic Regression in accuracy and F1-score.

Uploaded by

sakshamdura
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views

Craft Presentations

The document presents a project on sentiment analysis using machine learning and deep learning techniques, focusing on analyzing text data for sentiment understanding. It outlines the problem of unstructured data growth and the objectives of fine-tuning large language models. The findings indicate that DistilBERT outperforms traditional models like SVM, Naive Bayes, and Logistic Regression in accuracy and F1-score.

Uploaded by

sakshamdura
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 21

SENTIMENT ANALYSIS

USING MACHINE LEARNING

SUBMITTED BY:
SAKSHAM DURA: 27444/077
SANDESH REGMI: 27447/077
RENJIL SUNAR: 27436/077
INTRODUCTION:
Sentiment analysis has become an important task in natural
language processing (NLP), finding its way into various areas like
business intelligence, social media monitoring, and customer
feedback reviews. This project aims to put machine learning and
deep learning techniques to work for understanding sentiments
in text.
PROBLEM STATEMENT
Exponential growth of unstructured data
Domain Specific complexities for analyzing text data

OBJECTIVES
Analyze text data based on sentiment
FineTune large language models
SYSTEM OVERVIEW
Data Collection
DataSet
link:https://www.kaggle.com/datasets/abhi8923shriv/senti
ment-analysis-dataset/data

Exploratory Data Analysis (EDA)


value counts
data samples
descriptive summary
statistical summary
generating wordclouds
Figure: WordCloud for positive text
Figure: WordCloud for negative text
Figure: WordCloud for neutral text
Preprocessing
normalizing casing
removing stopwords
stemming / lemmatization
tokenizer / vectorizer
drop null values
How CountVectorizer works

Figure: CountVectorizer [1].


How DistilBERT Tokenizer works

Figure: DistilBERT Tokenizer [2].


Statistical Machine Learning Models
Logistic Regression
Support Vector Machines
Multinomial Naive Bayes

Transformer Based Models


BERT
DistilBERT
Figure: Transformer vs BERT [3].
Figure: Transformer Model [4].
Figure: BERT Model [4].
Evaluation Metrics

Figure: SVM metrics Figure: Multinomial Naive Bayes


metrics
Evaluation Metrics

Figure: Logistic Regression metrics Figure: DistilBERT Model


Findings:
DistilBERT consistently outperformed traditional models like SVM,
Naive Bayes, and Logistic Regression in terms of accuracy and F1-score.
Refernces:
1. https://www.ronaldjamesgroup.com/article/grab-your-wine-its-time-to-
demystify-ml-and-nlp
2. https://www.cnblogs.com/emanlee/p/17521830.html
3. https://www.linkedin.com/pulse/overview-transformer-bert-sanjay-
kumar-mba-ms-phd/
4. https://arize.com/blog-course/unleashing-bert-transformer-model-nlp/
THANK
YOU VERY
MUCH!

You might also like