ScholarSeek

Inspiration

Finding the right graduate program is daunting due to dispersed information on universities' research strengths and admissions criteria. As students ourselves, navigating this has been challenging. Our app aims to streamline this process for undergraduates by matching with ideal researchers, providing comprehensive university statistics, and aiding in initial contact preparation.

What it does

ScholarSeek streamlines the grad school search by identifying relevant researchers, offering detailed stats about their institutions, and generating email drafts for contact. It utilizes Google Scholar and chatGPT APIs to process a user's interests, identify suitable researchers and institutions, and streamline the search process through a user-friendly interface.

How we built it

We built ScholarSeek using Google Scholar and chatGPT APIs for information retrieval and analysis, coupled with a detailed college database. This data is presented in a streamlined Streamlit interface, developed entirely in Python, for an engaging user experience.

Challenges we ran into

Balancing direct parsing and chatGPT's capabilities was challenging, especially with institution names' variability. We opted for a hybrid approach using TF-IDF encoding and cosine similarity for a flexible, accurate database search, overcoming the limitations of strict matching and potential chatGPT inaccuracies.

Accomplishments that we're proud of

Our proudest accomplishment is delivering an end-to-end solution that simplifies the grad school search. We're particularly proud of our innovative use of NLP for data manipulation and introducing a flexible database search method, making the search process more efficient for students.

What we learned

The project enhanced our web development skills, especially in Streamlit, and provided a deep dive into managing a comprehensive project, including navigating the complexities of collaborative coding and version control.

What's next for ScholarSeek

We aim to improve ScholarSeek by speeding up information retrieval, refining our flexible search capabilities, and expanding our database to include international schools, making the search process even more efficient and comprehensive for students worldwide.

Ethical Statement

We recognize and adhere to all Hackathon rules. All code is our own, and written during the times given by the Hackathon. All development is made with the 'TechforGood' purpose in mind.