Project Introduction

  • Input your "SOS" message and click on Classify Your Message to see how the NLP model predict your request.
  • Or click on Classify Sample Message to let the NLP model predict a real SOS message from the sample database.

This demo aids in transforming business requirements into product prototypes by AI/ML, enhancing business intelligence, and improving decision-making processes.

This is a state-of-the-art NLP model with a 95% recall score in multi-label classification, enabling precise text analysis. NLP model can help clients analyze and predict user behaviors from reviews, posts, and messages, enhancing customer experience and supporting robust recommendation systems.
- Data Source: credit appen

  • Frontend: HTML, JavaScript, CSS, and Bootstrap.
  • Backend: Python, Flask, and Flask-RESTful.
  • Data Science & Machine Learning: NumPy, pandas, Plotly, Scikit-learn, NLTK, and MLflow.
  • DevOps: Model containerization into APIs to handle user requests. Web App was developed and deployed by Git, Docker, CI/CD, and AWS, ensuring scalability and reliability for real-time interactions, leading to intuitive and interactive user experience.
  • Security & Resilience: Well-architected framework with least privilege, access control, and auto scaling.

Overview of Training Dataset