Semiotic decision making system for responding to natural language queries and components thereof
First Claim
1. A semiotic processing module for a semiotic decision making system, wherein a training corpus of information in the form of sequential sets of elements is used to create a database which is thereafter used to make decisions relating to queries input in the same type of elements, comprising:
- a knowledge base for storing data representations of analyses of sets of lineally-related elements;
said knowledge base data representations comprising predicate records and elemental and non-elemental act records wherein;
each predicate record is associated with a class of one or more act records such that each act record is associated with only one class of act records;
each elemental act record represents a set element of said predefined sets and defines a single act class of a corresponding elemental predicate record; and
each non-elemental act record represents a sequence of a case predicate record followed by a result predicate record, such that all act records are recursively defined as representations of one or more sets of lineally-related elements and each predicate record represents analyses of the sets of lineally-related elements represented by each act record in the class of act records with which it is associated;
an input for receiving sets of lineally-related elements and storing a representation of each distinct element as an elemental act record in said knowledge base; and
a processor associated with said knowledge base which recursively processes said knowledge base records by evaluating the relationship and frequency of occurrence of individual elements and sets of elements based upon the lineal relationship of those elements as received by said input to generate predicate records and non-elemental act records based thereon which are then stored in said knowledge base.
4 Assignments
0 Petitions
Accused Products
Abstract
A semiotic decision making system operates to respond, for example, to natural language queries. The system operates independent of the type of symbolic elements in which queries are cast. A training corpus of information in the form of sequential sets of a selected type of element is input to the system and processed by a semiotic processing module to create a knowledge database based upon the lineal relationship of elements within the training corpus. The knowledge database is then used to make decisions relating to queries input in the same type of elements. Accordingly, inputting a French training corpus results in a knowledge base useful in answering French questions. Multiple semiotic processing modules may be used to enhance performance.
-
Citations
16 Claims
-
1. A semiotic processing module for a semiotic decision making system, wherein a training corpus of information in the form of sequential sets of elements is used to create a database which is thereafter used to make decisions relating to queries input in the same type of elements, comprising:
-
a knowledge base for storing data representations of analyses of sets of lineally-related elements;
said knowledge base data representations comprising predicate records and elemental and non-elemental act records wherein;
each predicate record is associated with a class of one or more act records such that each act record is associated with only one class of act records;
each elemental act record represents a set element of said predefined sets and defines a single act class of a corresponding elemental predicate record; and
each non-elemental act record represents a sequence of a case predicate record followed by a result predicate record, such that all act records are recursively defined as representations of one or more sets of lineally-related elements and each predicate record represents analyses of the sets of lineally-related elements represented by each act record in the class of act records with which it is associated;
an input for receiving sets of lineally-related elements and storing a representation of each distinct element as an elemental act record in said knowledge base; and
a processor associated with said knowledge base which recursively processes said knowledge base records by evaluating the relationship and frequency of occurrence of individual elements and sets of elements based upon the lineal relationship of those elements as received by said input to generate predicate records and non-elemental act records based thereon which are then stored in said knowledge base. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
each predicate record contains a reference to each act record in the class of act records with which it is associated; and
each non-elemental act record contains a reference to the predicate record associated with the class of acts with which it is associated, a reference to its case predicate record and a reference to its result predicate record.
-
-
3. A semiotic processing module according to claim 1 wherein the predicate and act records are stored at respective locations in a memory and wherein said references comprise the respective memory addresses of those locations.
-
4. A semiotic processing module according to claim 3 wherein each predicate record further contains the memory address of each act record for which it is a case predicate and the memory address of each act record for which it is a result predicate.
-
5. A semiotic processing module according to claim 3 wherein:
said knowledge base data representations further comprise opp stat records of a first type each stored at a respective memory location, wherein each opp stat record is associated with two predicate records and represents a relationship between the two predicate records.
-
6. A semiotic processing module according to claim 5, wherein:
-
each predicate record further contains the memory addresses of each opp stat record with which it is associated; and
each opp stat record of said first type contains the memory addresses of each of the two predicate records with which it is associated.
-
-
7. A semiotic processing module according to claim 6 wherein said first type of opp stat records comprises records for P/S opp stats and C/C opp stats.
-
8. A semiotic processing module according to claim 3 wherein:
said knowledge base data representations further comprise opp stat records of a second type each stored at a respective location in said memory, wherein each opp stat record is associated with two act records and represents a relationship between the two act records.
-
9. A semiotic processing module according to claim 8, wherein:
-
each act record further contains the memory addresses of each opp stat record with which it is associated; and
each opp stat record of said second type contains the memory addresses of each of the two act records with which it is associated.
-
-
10. A semiotic processing module according to claim 9 wherein said second type of opp stat records comprises records for C/A opp stats.
-
11. A semiotic processing module according to claim 3 wherein:
-
said knowledge base data representations further comprise thought sign records each stored at a respective location in said memory, each thought sign record being associated with a denotation of acts and containing the memory addresses of the act records for the acts in that denotation; and
each act record containing a first list of memory addresses for each thought sign record in which its act is a first one of the acts in the denotation for that thought sign.
-
-
12. A semiotic processing module according to claim 11 wherein:
each act record further contains a second list of memory addresses for each thought sign record in which its act is any one of the acts of the denotation for that thought sign.
-
13. A dyadic semiotic processing module according to claim 1 wherein:
-
the lineal relationship of set elements is sequential; and
each non-elemental act record represents a sequence of set elements which is a subset of a predefined set within a training corpus and which constitutes a sequence of set elements represented by its associated case predicate record directly followed by a sequence of set elements represented by its associated result predicate.
-
-
14. A triadic semiotic processing module according to claim 1 wherein:
-
the lineal relationship of set elements is such that each element represents a node of a multi-node tree form of said predefined sets having a single highest level node and a plurality n of lowest level nodes where each lowest level node is associated with a single higher level node and where each node which is not a lowest level node is associated with exactly two lower level nodes whereby the total number of nodes N is equal to 2n−
1; and
each non-elemental act record is associated with a case predicate record which represents a first subtree of set elements and a result predicate represents a second subtree of set elements having a highest level second subtree element which highest level second subtree element is directly lineally associated with a higher level first subtree element whereby said non-elemental act record represents a subtree of a tree form of a predefined set in a training corpus which includes all of the first and second subtree elements.
-
-
15. A triadic semiotic processing module according to claim 1 further comprising means for receiving tree form representations of set elements from a dyadic semiotic processing module such that the lowest level nodes of said tree forms represent individual sets of a training corpus which was processed by the dyadic semiotic processing module.
-
16. A triadic semiotic processing module for a semiotic decision making system, wherein a training corpus of information in the form of sequential sets of elements is used to create a database which is thereafter used to make decisions relating to queries input in the same type of elements, comprising:
-
a knowledge base for storing data representations of analyses of subsets of predefined sets of N lineally-related elements where each element represents a node of a multi-node tree form having a single highest level node and a plurality n of lowest level nodes where each lowest level node is associated with a single higher level node and where each node which is not a lowest level node is associated with left and right lower level nodes whereby the total number of nodes N is equal to 2n−
1;
said knowledge base data representations comprising predicates and elemental and non-elemental acts wherein;
each predicate is associated with a class of one or more acts such that each act is associated with only one class of acts;
each elemental act represents a set element and defines a single act class of a corresponding elemental predicate; and
each non-elemental act represents a sequence of a case predicate followed by a result predicate in either a left or right direction, such that all non-elemental acts are recursively defined as representations of one or more subsets of lineally-related elements of a predefined set and each predicate represents the subsets of lineally-related elements represented by each act within its associated class of acts where the result predicate of any given non-elemental act having a left direction represents a subset of elements having a highest node element which node element is the left lower node associated with one of the nodes represented by the case predicate of the given non-elemental act and where the result predicate of any given non-elemental act having a right direction represents a subset of elements having a highest node element which node element is the right lower node associated with one of the nodes represented by the case predicate of the given non-elemental acts;
an input for receiving for receiving tree form representations of set elements such that the lowest level nodes of said tree forms represent individual sets of a training corpus and storing a representation of each distinct tree form as an elemental act record in said knowledge base; and
a processor associated with said knowledge base which recursively processes said knowledge base records by evaluating the relationship and frequency of occurrence of sets of elements based upon the lineal relationship of those sets of elements as received by said input to generate predicate records and non-elemental act records based thereon which are then stored in said knowledge base.
-
Specification