SBIR Phase I: The Auto-Autodidact - A Web-Delivered Learning Environment Based on Latent Semantic Analysis (LSA)

点击翻译按钮获取中文摘要

This Small Business Innovation Research (SBIR)Phase I project will combine the Internet, electronic libraries, and a new machine learning technique that simulates human understanding of text to produce an independent learning and problem solving environment for individuals and groups. Using Latent Semantic Analysis (LSA), Auto-autodidact (autodidact: a self taught person) first learns the vocabulary and concepts of a topic by automatic training on textbooks. Then, as students study and write, and groups discuss and plan, it will continuously evaluate what they know and what they do not know, find relevant information anywhere in the electronic library, and connect participants with complementary needs and knowledge. Auto autodidact capitalize s on the motivational power of peer interaction, the instant availability of enormous textual resources, and the possibility of sharing individual knowledge over time and space. Auto autodidact will integrate LSA with Knowledge Forum, a state-of-the-art facilitator for distributed knowledge-building discussion, and newly available electronic libraries, to provide continuous embedded assessment, tutorial dialogue, and meaning-based information insertion. It will be unique in its ability to construct a learning environment for a new domain in a matter of days. Knowledge Analysis Technologies proffers a learning environment technology that has potential value for science and engineering education throughout the life cycle and for research and design organizations. The firm plans to commercialize the technology directly and through publishers, distance education providers, and educational testing organizations.