Inspiration
In today's fast-paced digital age, the search for the ideal housing solution can prove to be a complex task. Often, apartment listings may be overpriced or fail to offer value for money. Moreover, individuals typically need to sift through multiple websites to find a property that perfectly aligns with their requirements. The need for a comprehensive application that aggregates data from across the internet and delivers results via simple prompts is evident. Such an application can save users a substantial amount of time and effort.
What it does
Our plugin allows users to interact with multiple apartment listing websites simultaneously. By catering to specific needs (such as proximity to a target address, preferred room or house type, etc.), our plugin generates the most suitable and cost-effective apartment search results for users.
How we built it
We divided the project into several modules: data gathering, data processing, prompt data generation, and plugin-backend interaction. This structured approach enabled us to focus on each component individually, ensuring optimal performance and functionality.
Challenges we ran into
The primary challenge involved extracting data from online sources, processing it into structured data, and making it accessible to users. However, our dedicated efforts led to the successful realization of this task.
Accomplishments that we're proud of
In under two days, we successfully got the plugin up and running. We devised a methodology to extract data from apartments.com and convert it into a useful data structure. By effectively leveraging GPT, we ensured accurate and relevant conversational prompts, which greatly improved user experience.
What we learned
This project provided us with valuable insights into interacting with large language models and utilizing data to generate effective search suggestions. It was a significant learning experience that honed our skills and broadened our knowledge.
What's next for ApartmentSearch Plugin
We aim to expand the plugin's capabilities to support multiple platform data sources. We plan to conduct an in-depth study of user needs to design a unique algorithm that offers the most helpful and cost-effective recommendations. We also aspire to make this plugin available in the OPENAI plugin store, if feasible. Our journey is just beginning, and we are excited about the opportunities that lie ahead.
Log in or sign up for Devpost to join the conversation.