Transfer learning is a machine learning technique in which a model is pretrained on a large, broad dataset to encode underlying features and relationships, and then refined using a smaller dataset for a specific problem space. This technique is relevant to space biology research, where datasets typically have limited sample size and the problem space is restricted. Your challenge is: (1) to design a comprehensive database of publicly available biomedical datasets that could be used to pretrain different models for a “model zoo,” and (2) to determine relevant publicly available space biology datasets that could then be used to refine the models to investigate specific space biology questions.
Our project is a static web application designed to facilitate the exploration of databases and artificial intelligence models. It allows users to easily search for and access various resources by providing details such as name, type, size, and access links. The application streamlines the process by simply entering a keyword into the search bar to retrieve the desired data. Additionally, we integrated a chatbot powered by BioGPT to handle inquiries related to space biology. We trained BioGPT using papers from OSDR (Open Science for Life in Space - Open Science Data Repository). This project aims to assist researchers and curious individuals in finding valuable models and datasets, simplifying the knowledge contribution process. We utilized Python for the chatbot training and developed the web application using a combination of HTML, CSS, and JavaScript, with a user-friendly interface for easy navigation.