Natural language processing system and method for processing input information by predicting kind thereof
First Claim
Patent Images
1. A natural language processing system comprising:
- input means for inputting information represented in a natural language, the information including first information and second information, wherein the second information appears subsequent to the first information in a sequence of natural language information, and the second information is incomplete;
a knowledge base for storing therein knowledge concerning a domain of information to be inputted, general knowledge and linguistic knowledge;
analysis means for analyzing the first information and the second information inputted from said input means by referring to the knowledge stored in the knowledge base;
expectation means for expecting an attribute of the second information on the basis of a result of an analysis of the first information performed by said analysis means; and
expectation information storing means for storing the attribute of the second information expected by said expectation means, as expectation information,wherein, when said analysis means analyzes the second information, said analysis means compensates for the incompleteness of the second information based on the expectation information.
1 Assignment
0 Petitions
Accused Products
Abstract
A natural language processing method, by which a sequence of natural language information is analyzed so as to derive a concept represented by the information. In this method, the input natural language information is sequentially processed as word by word. At that time, the kind of a subsequent word is expected from a currently processed word by using knowledge concerning the word order of words in the natural language information. Thus the processing is performed by eliminating ambiguity in the information on the basis of such an expectation.
342 Citations
23 Claims
-
1. A natural language processing system comprising:
-
input means for inputting information represented in a natural language, the information including first information and second information, wherein the second information appears subsequent to the first information in a sequence of natural language information, and the second information is incomplete; a knowledge base for storing therein knowledge concerning a domain of information to be inputted, general knowledge and linguistic knowledge; analysis means for analyzing the first information and the second information inputted from said input means by referring to the knowledge stored in the knowledge base; expectation means for expecting an attribute of the second information on the basis of a result of an analysis of the first information performed by said analysis means; and expectation information storing means for storing the attribute of the second information expected by said expectation means, as expectation information, wherein, when said analysis means analyzes the second information, said analysis means compensates for the incompleteness of the second information based on the expectation information. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
-
-
12. A natural language processing method comprising:
-
the input step of inputting information in a natural language, the information including first information and second information, wherein the second information appears subsequent to the first information in a sequence of natural language information, and the second information is incomplete; the analysis step of analyzing the first information and the second information inputted at said input step in a sequence of natural language information inputted at said input step by referring to knowledge stored in a knowledge base which stores therein knowledge concerning a domain of information to be inputted, general knowledge and linguistic knowledge; the expectation step of expecting an attribute of second information on the basis of a result of an analysis of the first information performed in said analysis step; and the expectation information storing step of storing the attribute of the second information expected in the expectation step, in an expectation information memory as expectation information, wherein, when the second information is analyzed in the analysis step, compensation for the incompleteness of the second information is performed based on the expectation information. - View Dependent Claims (13, 14, 15, 16, 17, 18, 19, 20, 21, 22)
-
-
23. A computer-readable storage medium storing a natural language processing program for controlling a computer to process natural language information, said program comprising codes for causing the computer to perform:
-
the input step of inputting information in a natural language, the information including first information and second information, wherein the second information appears subsequent to the first information in a sequence of natural language information, and the second information is incomplete; the analysis step of analyzing the first information and the second information inputted at said input step by referring to knowledge stored in a knowledge base, which stores therein knowledge concerning a domain of information to be inputted, general knowledge and linguistic knowledge; the expectation step of expecting an attribute of second information on the basis of a result of an analysis of the first information performed in said analysis step; and the expectation information storing step of storing the attribute of the second information expected in the expectation step, in an expectation information memory as expectation information, wherein, when the second information is analyzed in the analysis step, compensation for the incompleteness of the second information is performed based on the expectation information.
-
Specification