Active Context
Abstract
The primary goal of our project was to build a human-machine communication system for collaborative creation of explanatory mechanistic models in biomedicine. We aimed to build a system leveraging modeling and simulation, as well as large knowledge bases (most of which are themselves machine-built) to provide an open-ended, natural language-driven platform where human and machine play equally important, yet complementary roles. The system aims to achieve major advances in both human-machine communication, and the understanding of complex molecular systems and disease. In addition, we also aimed to develop algorithms and software for reasoning about complex mechanisms operating in the natural world, explaining large-scale data, assisting humans in generating actionable, model-based hypotheses and testing these hypotheses empirically. We tackled this challenge by developing and combining fully automated algorithms for large-scale data interpretation and hypothesis generation with focused, dialogue-driven human-machine communication. Our integrated dialogue systems are designed to accelerate laboratory-based research and to assist molecular tumor boards in reasoning about anti-cancer therapies.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 12, 2021
- Accession Number
- AD1200477
Entities
People
- Peter K. Sorger
Organizations
- Harvard Medical School