This is a simple project that implements text sentiment analysis. Sentiment analysis is a machine learning technique that determines whether a given text expresses a positive or negative sentiment. Based on the user's input, the model classifies the text and outputs either "positive" or "negative."
The algorithm used in this model is logistic regression, provided by the Scikit-learn library. Logistic regression is one of the simplest and most efficient algorithms for handling text sentiment analysis.
The dataset used in this project is the IMDB (Internet Movie Database) dataset, which can be obtained from the Hugging Face Datasets library.
The trained model is available in .pkl and .onnx file formats.
(Note: The ONNX version has not been tested yet, so it may contain errors or compatibility issues.)
You can find all model files in the model directory. If you're interested in understanding the model logic, check out the Jupyter Notebook in this repository or click this link 👉🏻 here. You can also customize it with your own dataset or modify the model logic to improve performance.
Tip
To open the notebook without worrying about project setup and dependencies, simply use the Open in Kaggle widget. This will launch the notebook directly in your Kaggle account.
If you want to see how the model is deployed in a real application, check out the web directory, which contains the source code for the web app. The web app is built using Flask, a lightweight Python framework.
Currently, I am working on deploying the model in a way that allows it to run directly in the client’s web browser, eliminating the need for a Flask-based backend server. Unfortunately, I haven't been able to achieve this yet due to some errors in the code.
For more details on the current issues, please check the Issues tab in this repository.