https://github.com/5hraddha/sentiment-analysis

An innovative system for filtering and categorizing movie reviews
https://github.com/5hraddha/sentiment-analysis

countvectorizer dummyclassifier lgbmclassifier logisticregression matplotlib minmaxscaler nltk nltk-stopwords nltk-tokenizer numpy pandas seaborn spacy tfidfvectorizer torch tqdm transformers

Last synced: 1 day ago
JSON representation

An innovative system for filtering and categorizing movie reviews

• Host: GitHub
• URL: https://github.com/5hraddha/sentiment-analysis
• Owner: 5hraddha
• Created: 2024-07-25T14:38:02.000Z (2 months ago)
• Default Branch: main
• Last Pushed: 2024-07-25T14:51:35.000Z (2 months ago)
• Last Synced: 2024-09-21T09:33:02.952Z (6 days ago)
• Topics: countvectorizer, dummyclassifier, lgbmclassifier, logisticregression, matplotlib, minmaxscaler, nltk, nltk-stopwords, nltk-tokenizer, numpy, pandas, seaborn, spacy, tfidfvectorizer, torch, tqdm, transformers
• Language: Jupyter Notebook
• Homepage:
• Size: 23.8 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Sentiment Analysis for Moview Reviews

## Introduction

In the age of digital content and streaming services, movie reviews play a pivotal role in helping viewers make informed choices about what to watch. **The "Film Junky Union" project, an exciting venture for classic movie enthusiasts, seeks to revolutionize this experience by developing an innovative system for filtering and categorizing movie reviews**. By leveraging the power of machine learning, our primary objective is to create a model capable of automatically detecting negative movie reviews, aiding film aficionados in avoiding cinematic disappointments.To achieve this, we will utilize a comprehensive dataset of IMDb movie reviews, with polarity labels indicating whether a review is positive or negative. Through data preprocessing, exploratory data analysis (EDA), model training, and rigorous testing, **we aim to construct a robust classifier capable of achieving an F1 score of at least 0.85**. The project's findings will not only provide valuable insights into sentiment analysis within the film industry but also empower movie enthusiasts to make more informed viewing decisions.

## Project Goal

Few of our main goals are:

1. **Data Preprocessing**: Clean and preprocess the IMDb movie review dataset, including handling missing values, text cleaning, and tokenization.

2. **Exploratory Data Analysis (EDA)**: Perform EDA to gain insights into the data distribution, class balance, and other characteristics of the dataset.

3. **Sentiment Analysis Model**: Develop a sentiment analysis model that can classify movie reviews as positive or negative based on their text content.

4. **F1 Score of 0.85**: Achieve a minimum F1 score of 0.85 to ensure the model's accuracy in detecting negative reviews.