Mechanical Intelligence: Research and Applications
Abstract
This report summarizes the results of a research project whose goal is to develop computer systems that can provide easy access for nontechnicians to large, distributed data bases of information. Our goal has been to develop mechanisms for automating many of the detailed tasks that today are normally performed by a decision maker's technical staff. These include accepting a question, in natural (not necessarily grammatical) English, in the decision maker's own terms; planning a sequence of queries to various files to gather the requested information; developing the plan into a computer program or programs in the language of the data base management system on which the data resides; transmitting the retrieval programs, and monitoring their execution; and composing the information retrieved into a suitable output format. We have focused our efforts along two mutually supporting lines of research. First, we have created a performance system, called LADDER (for Language Access to Distributed Data with Error Recovery), that carries out all of the functions listed above in at least rudimentary form. Our second line of research focuses on longer-term efforts to develop the techniques required to satisfy more fully the needs of decision makers.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1977
- Accession Number
- ADA049027
Entities
People
- Daniel Sagalowicz
- Earl D. Sacerdoti
- Gary G. Hendrix
- Jonathan Slocum
- Paul Morris
- Richard E. Fikes
Organizations
- SRI International