System and method for querying questions and answers
A system and method for information retrieval are presented. A client computer receives a natural language query comprising an array of tokens. A query processing server analyzes the natural language query (interpreted as a question) to identify a plurality of terms and a relationship between one or more pairs of the terms according to a knowledge model defining interrelationships between a plurality of entities. A set of assertions is constructed using the relationship between the pair of terms, and a query is executed against a knowledge base of frequently asked questions, corresponding answers, documents and/or data using the set of assertions to generate a set of results. The knowledge base identifies a plurality of items, each of the plurality of items is associated with at least one annotation identifying at least one of the entities in the knowledge model. The set of results are transmitted to the client computer.
1. A system comprising a server, comprising a hardware computing device coupled to the network and including at least one processor executing instructions within a memory, which, when executed, cause the server to:
receive, from a client computer, an array of tokens;
identify, within the array of tokens:
at least one token indicating a question; and
a plurality of tokens common to a model stored in a data repository coupled to a network and defining:
a plurality of entities including abstract concepts and concrete instances; and
at least one interrelationship between the plurality of entities;
execute a query of the data repository, wherein the data repository further comprises:
at least one answer to the question; and
a digital information, associated in the data repository with the at least one answer, comprising the plurality of entities and the at least one interrelationship;
return a set of results comprising the at least one answer;
apply different weights, using at least one semantic similarity or distance between the array of tokens and the digital information, balancing the relationship between the question and the at least one answer;
maintain a log of partial, mismatching, or partially matching queries to identify at least one frequently asked question (FAQ) to be added to an FAQ database within the data repository; and
transmit the at least one answer to the client computer for display.
2. The system of claim 1 , wherein the array of tokens is a natural language query.
3. The system of claim 1 , wherein the at least one token indicating a question includes:
at least one term, including what, when, where, why, or how;
a question mark;
at least one regular expression; or
a result of an analysis performed using part-of-speech tagging.
4. The system of claim 1 , wherein the model is an ontological schema for a knowledge domain storing the plurality of entities as ontological components.
5. The system of claim 1 , wherein the instructions further cause the server to: locate at least one atomic element within the array of tokens that belongs to a predefined set of categories, according to an underlying ontological model.
6. The system of claim 1 , wherein the data repository is a database built for storage and retrieval of semantic data in the form of triple statements.
7. The system of claim 6 , wherein the triple statements include a subject, a predicate, and an object.
8. The system of claim 1 , wherein the digital information comprises:
a metadata associated with the at least one answer; or
a content within a file including the at least one answer.
9. The system of claim 1 , wherein the at least one answer is stored within: a database record; an electronic document; a spreadsheet; an email; an electronic video file; a plain text file; a marked up text file; an image; an audio file; an executable file; or a raw data file.
10. A method comprising the steps of:
receiving, by a server, comprising a hardware computing device coupled to the network and including at least one processor executing instructions within a memory, from a client computer, an array of tokens;
identifying, by the server, within the array of tokens:
at least one token indicating a question; and
a plurality of tokens common to a model stored in a data repository coupled to a network and defining:
a plurality of entities including abstract concepts and concrete instances; and
at least one interrelationship between the plurality of entities;
executing, by the server, a query of the data repository, wherein the data repository further comprises:
at least one answer to the question; and
a digital information, associated in the data repository with the at least one answer, comprising the plurality of entities and the at least one interrelationship;
returning, by the server, a set of results comprising the at least one answer;
applying, by the server, different weights, using at least one semantic similarity or distance between the array of tokens and the digital information, balancing the relationship between the question and the at least one answer;
maintaining, by the server, a log of partial, mismatching, or partially matching queries to identify at least one frequently asked question (FAQ) to be added to an FAQ database within the data repository; and
transmitting, by the server, the at least one answer to the client computer for display.
11. The method of claim 10 , wherein the array of tokens is a natural language query.
12. The method of claim 10 , wherein the at least one token indicating a question includes:
at least one term, including what, when, where, why, or how;
a question mark;
at least one regular expression; or
a result of an analysis performed using part-of-speech tagging.
13. The method of claim 10 , wherein the model is an ontological schema for a knowledge domain storing the plurality of entities as ontological components.
14. The method of claim 10 , further comprising the step of: locating, by the server, at least one atomic element within the array of tokens that belongs to a predefined set of categories, according to an underlying ontological model.
15. The method of claim 10 , wherein the data repository is a database built for storage and retrieval of semantic data in the form of triple statements.
16. The method of claim 15 , wherein the triple statements include a subject, a predicate, and an object.
17. The method of claim 10 , wherein the digital information comprises:
a metadata associated with the at least one answer; or
a content within a file including the at least one answer.
18. The method of claim 10 , wherein the at least one answer is stored within: a database record; an electronic document; a spreadsheet; an email; an electronic video file; a plain text file; a marked up text file; an image; an audio file; an executable file; or a raw data file.