IDA: An Intelligent Data Access Program

This paper is concerned with one of the components of LADDER Language Access to Distributed Data with Error Recovery, a database access system currently being developed at SRI. The ultimate goal of this system is to provide decision makers with easy access to information stored in multiple computers, under various database management systems DBMSs. The particular application for LADDER is as an aid to Navy decision makers, but the techniques being developed are likely to be applicable to a wide range of decision-making activities. The first component of LADDER is INLAND Informal Natural Language Access to Navy Data, which allows the user to ask questions in English about information contained in databases similar to those currently used by the Navy. It translates the query into a formal high-level query that contains no mention of the database structure. This query is then passed to the next component of the system Intelligent Data Access IDA. IDA was developed at SRI to allow a casual user to retrieve information from a database knowing the fields present in the database, but not the structure of the database itself. IDAs input is a very simple, formal query language that is essentially a list of restrictions on fields and queries about fields, with no mention of database structure. It produces a series of DBMS queries that are transmitted over the ARPA network. The results of these queries are combined by IDA to provide the answer to the users query. In this paper, the author defines the input language, and give examples of IDAs behavior. He also presents his representation of the structural schema, which is the information needed by IDA to know how the database is actually organized. He gives an idea of some of the heuristics that are used to produce a program in the language of the DBMS. Finally, he discusses the limitations of this approach, as well as future research areas.