Inspiration

  • Spam waste time and resources, sometimes might be harmful.
  • Early identification can prevent potential harm and protect user’s personal information

What it does

  • Text area for message input
  • Button to check for spam
  • Reset button to clear inputs and results
  • Sends user inputs to modified AI tester
  • Display results of spam or not

How we built it

Front-end

HTML, CSS, Python, JavaScript

Back-end:

AI spam checker API, CSV Handling, Data preprocessing

Challenges we ran into

API Integration:

Searching suitable API and create API-key for it Fine tuning AI model by inputting data took much time Limited computer storage to run the AI model

Front end design:

New to front-end coding language like HTML and CSS, took time to learn and design a clear and pretty website.

Accomplishments that we're proud of

Results

  • Successfully classified messages as “Spam” or “Ham”
  • Achieved real-time feedback with minimal delay using API

Evaluation

  • Accuracy of the AI model in minimal storage usage
  • Limited response time from API

What we learned

Throughout our AI-powered SMS spam detector project, we gained valuable experience and skills across multiple domains. On the frontend, we learned to create an intuitive, responsive user interface using modern web technologies like React.js and Tailwind CSS. We also developed expertise in handling data, working with CSV files, and preprocessing data for machine learning and API integration. Collaborating under tight deadlines, we honed our problem-solving abilities, learning to handle API responses, manage errors, and ensure the smooth functioning of features like reset and clear. Overall, this project not only taught us technical skills but also the importance of effective communication, adaptability, and time management in a real-world development scenario.

What's next for Spam SMS Checker with Gen AI

  • Users could modify the AI model according to their specific situation
  • Enhance user interface (e.g. message history)
  • Develop harmful message category (e.g. Spam, Phishing )
  • Further fine tune the model to reduce memory usage – aiming to deploy it on cell phones

Built With

Share this project:

Updates