This accelerator template leverages the power and simplicity of Node.js and Express to create an AI-driven Text Analysis Accelerator. It features sentiment analysis using a machine learning model from ml5.js, trained on movie reviews. The application provides a responsive and interactive user interface, making it easy to analyze the sentiment of various texts.
- Sentiment Analysis: Uses an ml5.js model trained on movie reviews to score text sentiment from 0 ("negative") to 10 ("positive").
- Efficient Text Processing: Capable of analyzing texts up to 200 words, focusing on the 20,000 most common words found in movie reviews.
- User-Friendly Interface: A clean and intuitive UI built with Materialize CSS, offering a straightforward way for users to input text and receive sentiment scores.
- Customizable Analysis Options: Allows users to adjust parameters like word limit for more flexible text analysis.
To run this app, ensure you have Node.js and npm installed. Set up the application with these steps:
git clone [your-repo-link]
cd [your-repo-directory]
npm install
npm start
The app will be accessible at http://localhost:80.
Add the Accelerator on VMware Tanzu Application Platform
tanzu acc create awesome-ai-text-analysis --git-repo https://github.com/fklein82/awesome-ai-text-sentiment-analysis --git-branch main --interval 5s
Follow these steps to deploy on Tanzu:
Install and configure the Tanzu CLI.
Navigate to your project directory:
cd [your-repo-directory]
Deploy using the Tanzu CLI:
tanzu apps workload create -f config/workload.yaml
Monitor deployment status:
tanzu apps workload tail text-analysis --timestamp --since 1h
Access the application via the URL provided by Tanzu Application Platform.
The index.html of this app includes:
- An input area for users to enter text.
- Display area for sentiment score results.
- Controls for adjusting analysis parameters.
- Integration with ml5.js for sentiment analysis.
Contributions are welcome to enhance the Text Analysis Accelerator. Open issues or submit pull requests with your suggestions or improvements.
The software is provided "as is", without any warranties. As the creator, I am not liable for any claims or liabilities arising from the use of this software.