Tailored Discovery

Inspiration

The rapid growth of TikTok Shop presented a unique opportunity to enhance the shopping experience by providing personalized product recommendations. We were inspired by the potential to leverage TikTok's extensive user data and innovative algorithm to drive sales and improve user satisfaction through targeted suggestions and personalize the pages through a few functions.

What it does

1.New Users: Recommends products based on highest ratings and popularity. 2.Returning Users: Analyzes past shopping history to provide personalized recommendations, using advanced AI to suggest products matching user behavior and preferences. And allowing users to input their current circumstances to recommend the most suitable products based on both their viewing habits and needs.

How we built it

1.Backend Framework: Utilized Django for robust web application development.

2.AI Integration: Integrated Llama3-70b with OpenAI libraries for sophisticated AI support.

3.Data Aggregation: Combined user purchase histories, product ratings, and TikTok Shop interaction data.

4.Personalization Algorithms: Developed algorithms to analyze user behavior and provide tailored recommendations.

5.User Interface: Implemented checkbox filters and dynamic querying to enhance user control and interaction.

Challenges we ran into

1.Data Integration: Managing large and diverse datasets. 2.Algorithm Optimization: Ensuring accurate and relevant recommendations. 3.Scalability: Designing a system that efficiently handles large volumes of data and interactions. 4.User Experience: Balancing automated recommendations with user control

Accomplishments that we're proud of

1.High-Quality Recommendations: Providing accurate and personalized product suggestions. 2.Stunning User Interface: Allowing users to change themes according to their own interests. 3.Scalable Solution: Creating a system capable of handling extensive datasets and user interactions. 4.Innovative AI Use: Leveraging advanced AI models to enhance the recommendation process.

What we learned

1.Data Importance: Quality and quantity of data are crucial for effective recommendations. 2.User Behavior Analysis: Understanding user behavior is in order to choose the right recommendations and user friendly UI 3.Continuous Improvement: The system must continuously learn and adapt. 4.Balancing Automation and Control: Providing a mix of automated recommendations and user control enhances the experience.

What's next for Tailored Discovery

1.Data Expansion: Continuously expand our dataset for more comprehensive user data. 2.Advanced AI Models: Explore deeper AI integration with advanced techniques. 3.User Feedback Integration: Incorporate direct user feedback for ongoing refinement. 4.Feature Enhancements: Develop additional features for further personalization and engagement. 5.Market Expansion: Expand the recommendation system to other platforms and markets beyond TikTok Shop.