Airmen Intelligent Role-base Search Engine (A-RISE) Tactical Funding Increase (TACFI)

We propose an extension to the SBIR Phase II-Funded minimal viable product, A-RISE (Airman Role-based Intelligence Search Engine), our chatbot assistant for fusion analysts that simplifies and speeds up the information search process. A-RISE uses natural language processing (NLP), a precursor to large language models (LLMs) to show more relevant results to the user based on topic extraction and ranking of found files. This smart search system was designed to show results that account for use of synonyms and a general understanding of topics so that relevant results could be shown without using the same exact language as in the found text. Our proposed TACFI extension on A-RISE will consist of two main activities: (1) Shift to an LLM framework that can handle geospatial data and (2) train A-RISE to associate LLM-type queries with Space Force guardian roles. The first task will allow our technology to benefit from the advancements made in LLMs in the last year, which allow for a higher quality of data processing and more conversational queries with the chatbot. The second activity will enable the system to personalize results over time for users, based on their specific roles and responsibilities. This will allow Airmen and Space Force Guardians to receive more relevant and accurate information for their specific duties.