Applied artificial intelligence technology for conversational inferencing using named entity reduction
Disclosed herein is an NLP system that is able to extract meaning from a natural language message using improved parsing techniques. Such an NLP system can be used in concert with an NLG system to interactively interpret messages and generate response messages in an interactive conversational stream. The parsing can include (1) named entity recognition that contextualizes the meanings of words in a message with reference to a knowledge base of named entities understood by the NLP and NLG systems, (2) syntactically parsing the message to determine a grammatical hierarchy for the named entities within the message, (3) reduction of recognized named entities into aggregations of named entities using the determined grammatical hierarchy and reduction rules to further clarify the message's meaning, and (4) mapping the reduced aggregation of named entities to an intent or meaning, wherein this intent/meaning can be used as control instructions for an NLG process.
1. A natural language processing method that applies computer technology to perform named entity reduction with respect to a message, the method comprising:
a processor applying a plurality of reduction rules to combinations of named entities in a defined grammatical hierarchy of named entities derived from a message, wherein the applying starts from a lowest named entity in the defined hierarchy and progresses toward a top named entity in the defined hierarchy, the message comprising data representing a plurality of words arranged in natural language, wherein the named entities in the defined hierarchy have associated named entity types, wherein the reduction rules define how a plurality of different combinations of named entity types are reduced into a plurality of different aggregations of named entities, wherein each aggregation of named entities also has an associated named entity type, and wherein the named entity types include an entity, a qualifier, an attribute, a relationship type, and an entity type; and
in response to the applying, a processor reducing the named entities within the message to a reduced aggregation of named entities, wherein the reduced aggregation of named entities serves as a representation of an inferred core meaning for the message, the inferred core meaning for use by a natural language generation system to generate a natural language response to the message; and
wherein the reduction rules comprise at least two members of the group consisting of:
a first reduction rule that reduces a combination of named entities corresponding to the entity named entity type, the qualifier named entity type, and the attribute named entity type to an aggregated named entity corresponding to the entity named entity type;
a second reduction rule that reduces a combination of named entities corresponding to the relationship type named entity type and the entity named entity type to an aggregated named entity corresponding to the relationship entity type;
a third reduction rule that reduces a combination of named entities corresponding to the entity named entity type and the relationship named entity type to an aggregated named entity corresponding to the entity named entity type; and
a fourth reduction rule that reduces one or more named entities corresponding to the entity type named entity type to an aggregated named entity corresponding to the entity named entity type.
2. The method of claim 1 further comprising:
a processor syntactically parsing the message to generate data that defines the grammatical hierarchy of named entities within the message.
3. The method of claim 1 wherein the reduction rules comprise:
the first reduction rule;
the second reduction rule;
the third reduction rule; and
the fourth reduction rule.
4. A natural language processing apparatus for performing named entity reduction with respect to a message, the apparatus comprising:
a memory; and
a processor for cooperation with the memory, the processor configured to:
apply a plurality of reduction rules to combinations of named entities in a defined grammatical hierarchy of named entities derived from a message, wherein the applying starts from a lowest named entity in the defined hierarchy and progresses toward a top named entity in the defined hierarchy, the message comprising data representing a plurality of words arranged in natural language, wherein the named entities in the defined hierarchy have associated named entity types, wherein the reduction rules define how a plurality of different combinations of named entity types are reduced into a plurality of different aggregations of named entities, wherein each aggregation of named entities also has an associated named entity type, and wherein the named entity types include an entity, a qualifier, an attribute, a relationship type, and an entity type; and
in response to the applied reduction rules, reduce the named entities within the message to a reduced aggregation of named entities, wherein the reduced aggregation of named entities serves as a representation of an inferred core meaning for the message, the inferred core meaning for use by a natural language generation system to generate a natural language response to the message; and
wherein the reduction rules comprise at least two members of the group consisting of:
a first reduction rule that reduces a combination of named entities corresponding to the entity named entity type, the qualifier named entity type, and the attribute named entity type to an aggregated named entity corresponding to the entity named entity type;
a second reduction rule that reduces a combination of named entities corresponding to the relationship type named entity type and the entity named entity type to an aggregated named entity corresponding to the relationship entity type;
a third reduction rule that reduces a combination of named entities corresponding to the entity named entity type and the relationship named entity type to an aggregated named entity corresponding to the entity named entity type; and
a fourth reduction rule that reduces one or more named entities corresponding to the entity type named entity type to an aggregated named entity corresponding to the entity named entity type.
5. The apparatus of claim 4 wherein the processor is further configured to:
syntactically parse the message to generate data that defines the grammatical hierarchy of named entities within the message.
6. The apparatus of claim 4 wherein the reduction rules comprise:
the first reduction rule;
the second reduction rule;
the third reduction rule; and
the fourth reduction rule.
7. A computer program product embedded on computer transitory computer readable storage medium for natural language processing that applies computer technology to perform named entity reduction with respect to a message, the computer program product comprising:
a plurality of processor-executable instructions that are resident on a non-transitory computer readable storage medium, wherein the instructions are configured, upon execution by a processor, to cause the processor to:
apply a plurality of reduction rules to combinations of named entities in a defined grammatical hierarchy of named entities derived from a message, wherein the applying starts from a lowest named entity in the defined hierarchy and progresses toward a top named entity in the defined hierarchy, the message comprising data representing a plurality of words arranged in natural language, wherein the named entities in the defined hierarchy have associated named entity types, wherein the reduction rules define how a plurality of different combinations of named entity types are reduced into a plurality of different aggregations of named entities, wherein each aggregation of named entities also has an associated named entity type, and wherein the named entity types include an entity, a qualifier, an attribute, a relationship type, and an entity type; and
in response to the applied reduction rules, reduce the named entities within the message to a reduced aggregation of named entities, wherein the reduced aggregation of named entities serves as a representation of an inferred core meaning for the message, the inferred core meaning for use by a natural language generation system to generate a natural language response to the message; and
wherein the reduction rules comprise at least two members of the group consisting of:
a first reduction rule that reduces a combination of named entities corresponding to the entity named entity type, the qualifier named entity type, and the attribute named entity type to an aggregated named entity corresponding to the entity named entity type;
a second reduction rule that reduces a combination of named entities corresponding to the relationship type named entity type and the entity named entity type to an aggregated named entity corresponding to the relationship entity type;
a third reduction rule that reduces a combination of named entities corresponding to the entity named entity type and the relationship named entity type to an aggregated named entity corresponding to the entity named entity type; and
a fourth reduction rule that reduces a combination of named entities corresponding to the entity type named entity type to an aggregated named entity corresponding to the entity named entity type.
8. The computer program product of claim 7 wherein the instructions are further configured, upon execution by the processor, to cause the processor to:
syntactically parse the message to generate data that defines the grammatical hierarchy of named entities within the message.
9. The computer program product of claim 7 wherein the reduction rules comprise:
the first reduction rule;
the second reduction rule;
the third reduction rule; and
the fourth reduction rule.
10. The method of claim 1 wherein the reduction rules comprise at least three members of the group consisting of the first reduction rule, the second reduction rule, the third reduction rule, and the fourth reduction rule.
11. The apparatus of claim 4 wherein the reduction rules comprise at least three members of the group consisting of the first reduction rule, the second reduction rule, the third reduction rule, and the fourth reduction rule.
12. The computer program product of claim 7 wherein the reduction rules comprise at least three members of the group consisting of the first reduction rule, the second reduction rule, the third reduction rule, and the fourth reduction rule.