CSL2050-Pattern Recognition and Machine Learning

Course Project : Twitter Sentiment Analysis

Project description:

A text can provide extensive insight into the sentiment conveyed by the author. Consequently, sentiment analysis derived from textual input is a well-established problem statement in the domains of machine learning and natural language processing (NLP). In this project, we aimed to address this challenge by employing a conventional machine learning methodology to analyze a dataset of over 30,000 tweets from Twitter. The objective was to implement and analyze different classifiers on the Sentiment Analysis Dataset, utilizing various preprocessing techniques (LDA & PCA) and tokenizers (BERTTokenizer and TFIDtokenizer). The project explored and thoroughly analyzed the performance of classifiers- Decision Trees, Random Forests, SVMs, Naive-Bayes, Perceptron and Logistic Regression, optimizing each to yield maximum accuracy by modifying any parameters or hyperparameter tuning. Finally, ensemble learning was employed to achieve optimum performance of the model. It was observed that TFID gave the maximum accuracy, which reached as high as 71.5

The repository contains the following files and folders:

Classifiers: This directory contains the individual classifiers' codes as python files

Dataset: This directory includes the dataset's csv files used for the project

images: This directory simply stores the images used in the project-page

webapp: This directory contains the files and code associated with the demo webapp implemented using flask.

Report.pdf : This is the report file containing detailed description and analysis of the project.

index.html : This contains the code for the project web-page.

data_preprocessing.ipynb : It contains the pre-processing code run on the raw dataset to convert it to numeric data using various vectorization methods.

master_file.ipynb : This is the main file of the project containing the pre-processing,all the classifiers and the associated results.

demo.ipynb : This notebook is a demo which takes input of the string, time and age and returns the predicted sentiment associated with the string.

presentation.pdf : It is a brief presentation of the project.

Youtube presentation link: https://youtu.be/Jw3_Ira3pJ4

Contributers:

Krishna Patil (B22CS078)

Saumitra Agarwal (B22AI054)

Harshit Goyal (B22CS024)

Mukund Gupta (B22CS086)

Aarohi Dharmadhikari (B22AI001)

Gouri Patidar (B22AI020)