Guided recommender

What it does

Our chatbot offers features aimed at delivering a comfortable user experience. At its core, it serves as an electric Mercedes car recommender system, employing algorithms to match users with their ideal vehicle based on an analysis of preferences, budget constraints, age, and lifestyle considerations. Furthermore, our platform incorporates profile analysis capabilities, allowing us to gain deeper insights into users' needs and preferences, thereby enabling us to provide tailored recommendations that resonate with each individual. Moreover, by integrating sentiment analysis into our system, we can effectively estimate user satisfaction levels and refine our recommendations accordingly, ensuring a seamless and personalized interaction. Through these integrated functionalities, we provide a refined and intuitive platform for users to discover their perfect electric Mercedes car with confidence and precision.

How we built it

In constructing our chatbot, we employed a systematic approach centered around three key components. First, we utilized LangChain technology to establish communication with ChatGPT, enabling the retrieval of pertinent information from a comprehensive vector database enriched with accurate gathered data on companies and their car offerings. This facilitated a robust foundation for informed decision-making and personalized recommendations. Second, we developed a logistic regression recommender specifically tailored for car matching, leveraging sophisticated algorithms to analyze user preferences and seamlessly align them with the most suitable electric Mercedes vehicle options available. Finally, we carefully engineered the control flow and guidance mechanisms within the conversation to ensure a smooth and intuitive user experience, fostering seamless interaction and engagement. Through the integration of these elements, we have created a sophisticated chatbot which provides users with tailored suggestions and a seamless browsing experience.

Challenges we ran into

Our biggest challenge was the lack of data to train our system successfully so we had to write it ourselfes.

What we learned

In our journey, we have gained invaluable insights into the development of recommendation systems and the utilization of LangChain technology. Creating a recommendation system required a deep dive into understanding user preferences and behavior, along with the intricacies of algorithmic analysis. We learned how to effectively leverage data to create personalized recommendations, enhancing user experience and satisfaction. Additionally, our exploration of LangChain provided us with a powerful tool for facilitating communication with ChatGPT and accessing vast databases of information. This experience has broadened our understanding of AI-driven systems and equipped us with valuable skills applicable across various domains.