Natural language and voice output for relational data base systems

ABSTRACT

A method to output answers and reasons in English and/or voice in a question-answering system is presented. The system for which the output is designed has a relational data base. A description of &pgr;-relations is given. A &pgr;-relation is one which has multiple values for the predicate or arguments of the relation. The natural language output uses predicate and function patterns for the &pgr;-relations. Pattern construction and pattern expansion into English sentences are described. The context-sensitive translator which converts written sentences into voice output is also described. Lastly, an algorithm is given which extracts the answer and reasons from a proof. The reasons are logically grouped so the user may easily follow the deductive proof procedure. The above has been implemented as part of the MRPPS 3.0 system at the University of Maryland.

REFERENCES

• 1: Bennett, S., Answer and Reason Extraction in Theorem Proving, (Scholarly Paper 117, Univ. of Maryland, 1973).
• 2: Donald D. Chamberlin, Relational Data-Base Management Systems, ACM Computing Surveys (CSUR), v.8 n.1, p.43-66, March 1976 [doi>10.1145/356662.356665]
• 3: Chin-Liang Chang , Richard C. Lee , Richard Char-Tung Lee, Symbolic Logic and Mechanical Theorem Proving, Academic Press, Inc., Orlando, FL, 1997
• 4: Chester, D., The translation of formal proofs into English, Artificial Intelligence, 3 (1976).
• 5: E. F. Codd, A relational model of data for large shared data banks, Communications of the ACM, v.13 n.6, p.377-387, June 1970 [doi>10.1145/362384.362685]
• 6: Elovitz, H. S., Automatic Text Translations to Speech Via Letter-to-Sound Rules, (Scholarly Paper, Univ. of Maryland, 1975).
• 7: Daniel Herbert Fishman, Experiments with a resolution-based deductive question-answering system and a proposed clause representation for parallel search., 1973
• 8: Fishman, D. H. and Minker, J., II-Representation: Clause Representation for Parallel Search, Artificial Intelligence 6, 2, 102-127 (1975).
• 9: Hill, R., LUSH Resolution and its Completeness, (Univ. of Edinburgh, School of Artificial Intelligence, DCL Memo No. 78, August 1974).
• 10: Klein, M., Translating II-Clauses into English, (Scholarly Paper, Univ. of Maryland, June 1976).
• 11: Kowalski, R. and Kuehner, D., Linear resolution with selection function, Artificial Intelligence 2 (1971).
• 12: Kowalski, R., Predicate logic as programming language, Proceedings of IFIP, 569 (1974).
• 13: Luckham, D. and Nilsson, N., Extracting information from resolution proof trees, Artificial Intelligence 2 (1971).
• 14: McSkimin, J. and Minker, J. A Predicate Calculus Based Semantic Network for Question-Answering Systems, (TR-509, University of Maryland, Dept. of Computer Science, March 1977).
• 15: Minker, J. and Wilson, G. A note on answer extraction in resolution based theorem proving, International Journal of Computers and Information Sciences, 6, 3 (1977).
• 16: Minker, J., Information Storage and Retrieval - A Survey and Functional Description, (TR-369, Univ. of Maryland, Dept. of Computer Science, April 1975).
• 17: Minker, J., Search Strategy and Selection Function for an Inferential Relation System, (TR-497, Univ. of Maryland, Dept. of Computer Science, December 1976).
• 18: Minker, J., An experimental relational data base system based on logic, in Logic and Data Bases, H. Gallaire and J. Minker, eds, p. 107-148 (Plenum Press, New York, 1978).
• 19: Nils J. Nilsson, Problem-Solving Methods in Artificial Intelligence, McGraw-Hill Pub. Co., 1971
• 20: Powell, P., Answer-Reason Extraction in a Parallel Relational Data Base System, (Master's Thesis, University of Maryland, April 1977).
• 21: Powell, P., MRPPS 3.0 User's Manual, (to be published at the Univ. of Maryland, 1978).
• 22: J. A. Robinson, A Machine-Oriented Logic Based on the Resolution Principle, Journal of the ACM (JACM), v.12 n.1, p.23-41, Jan. 1965 [doi>10.1145/321250.321253]
• 23: Thompson, P. Rule Precedence within Context-Sensitive Grammars; An Extension of the Letter to Sound Rules Used in Voice Production, (Scholarly Paper, Univ. of Maryland, 1976).

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.4 Systems
          Subjects: Relational databases

Additional Classification:
  H. Information Systems
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.2 User Interfaces (D.2.2, H.1.2, I.3.6)
          Subjects: Natural language; Voice I/O

General Terms:
Languages, Management

Keywords:
Answer/reason extraction, MRPPS 3.0, Natural language output, Question-answering system, Relational data base, Voice output