This is a computer program designed to answer questions. It works by taking a question, then searching through various kinds of information sources, including structured question-and-answer pairs, regular text, and linked data (RDF). It then evaluates the potential answers it finds to give you a final response. The method specifically requires searching at least these three types of databases.
Why it matters: Filed before the widespread adoption of large language models (LLMs), which have since revolutionized natural language understanding, semantic search, and the ability to integrate and process diverse data types for question answering. Modern NLP tools make parsing questions and identifying properties far more robust than in 2013.
AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.
Reinvent this with AI