Analysis of tweets fetched in real time from twitter based on a categorical keyword and then sentiment evaluation of new tweets based on the tweet data
- Description
- Installation
- Usage
- Contributing
- Credits
- License
There is a great spectrum of sentiments that can be associated with tweets from different domains.
This project bridges the gap between the spectrum of emotions and sentiments across tweets from various domains and at different points of time.
It provides real time sentiment analysis on categorical tweets and also providing the appropriate sentiments for new data.
It focuses on analysing real time tweets data from twitter based on a keyword and extracting sentiments from tweets on twitter to identify the kind of emotion/sentiment associated with a particular person, thing, topic, idea, etc.
The fetched tweets are used to train a NLP model which in turn produces appropriate sentiment score on any new tweet based on the real time data. The user knows the sentiment associated with that tweet before tweeting
- It is always a good idea to create a virtual environment for such projects. Inside a desired directory create a virtual environment for this project. For windows use : py -m venv env
- Activate the virtual environment using : .\env\Scripts\activate
- Clone the project using : git clone <ssh/https url>
- cd into the repository directory and run pip install -r requirements.txt to install the packages only for this virtual environment.
-
You can run the application using the analysis_gui.py
-
Run it using python analysis_gui.py
-
This opens up a gui asking you to enter a word for fetching the tweets and training the model.
-
After training of the model, you can enter a word and then model predicts its sentiment.
To contribute to the project follow the below steps
- Fork the project
- Follow the installation steps provided above on your local
- For any additional feature or bug create a new issue. Or work on an existing issue.
- Create a pull request mentioning the issue in your pull request.