Rules-based decision support system for assessment of digital content involving natural language
A decision support system for assessing and reviewing large volume of digital content which comprise complex subject matter and providing recommendations on relevance of each content by applying context-based rules which are specific to the subject matter of interest. The rules are captured in a standardized format and the algorithm for the rules-based decision making is designed with the flexibility to select the rules based on the subject matter of interest.
1 . A method for processing natural language input to answer questions, the method comprising:
parsing content comprising the natural language input;
transforming the natural language input into context-based attributes, wherein transforming comprises:
performing a plurality of checks on the content, using the context-based attributes, wherein each context-based attribute is assigned a specific check which generates an outcome represented by an outcome value; and
converting outcome values for each context-based attribute into at least one of binary, ternary, or other discrete forms;
creating a rules grid which is a set of rules, used to process the natural language input;
applying the set of rules to make an assessment on relevance of the natural language input for a subject matter of interest through qualitative and quantitative measures, indicating relevance and ranking the content, respectively; and
enabling a response to the questions based on the set of rules applied, wherein the response comprises a summary response and a detailed response providing a basis for the assessment of the content.
2 . The method for processing natural language input as claimed in claim 1 , comprising defining the context-based attributes, wherein the context-based attributes provide contextual information related to the subject matter of interest and generate outcomes in a form of discrete output values.
3 . The method for processing natural language input as claimed in claim 1 , comprising setting rules, wherein the rules are defined by a combination of the outcome values of each context-based attribute, which are used to make the assessment of the content.
4 . The method for processing natural language input as claimed in claim 1 , wherein the context-based attributes are used to determine the relevance of the natural language content for the subject matter of interest, by evaluating the content based on the context-based attributes.
5 . The method for processing natural language input as claimed in claim 1 , wherein the plurality of checks comprises checking for one or more of a keyword, a sentiment, an association, value, a location of an alphabet or a string, and a model based prediction.
6 . The method for processing natural language input as claimed in claim 1 , wherein the outcome value comprises one or more of:
yes or no;
positive, negative, or neutral;
found or not found;
a combination of greater than, less than or equal to;
a location-based input;
categorization using meta-data; and
classification using ML/DL models.
7 . The method for processing natural language input as claimed in claim 1 , wherein a rule of the set of rules that is predefined as a distinct combination of the outcome values of each context-based attribute, represents a decision.
8 . The method for processing natural language input as claimed in claim 1 , wherein applying the set of rules comprises performing rule optimization by one or more of:
a frequency based selection;
a staging approach; and
an exclusion approach.
9 . A computer software product for use on a computer system, the computer software product comprising:
a computer readable non-transitory storage medium,
computer program code means stored on the computer readable non-transitory storage medium, the computer program code means comprising encoded instructions, wherein the encoded instructions comprise:
parsing content comprising the natural language input;
transforming the natural language input into context-based attributes, wherein transforming comprises:
performing a plurality of checks on the content, using the context-based attributes, wherein each context-based attribute is assigned a specific check which generates an outcome represented by an outcome value; and
converting outcome values for each context-based attribute into at least one of binary, ternary, or other discrete forms;
creating a rules grid which is a set of rules, used to process the natural language input;
applying the set of rules to make an assessment on relevance of the natural language input for a subject matter of interest through qualitative and quantitative measures, indicating relevance and ranking the content, respectively; and
enabling a response to the questions based on the set of rules applied, wherein the response comprises a summary response and a detailed response providing a basis for the assessment of the content.
10 . The computer software product claim of claim 9 , wherein the context-based attributes are used to determine the relevance of the natural language content for the subject matter of interest, by evaluating the content based on the context-based attributes.
11 . The computer software product claim of claim 9 , wherein the plurality of checks comprises checking for one or more of a keyword, a sentiment, association, a value, a location of an alphabet or a string, and a model based prediction.
12 . The computer software product claim of claim 9 , wherein the outcome value comprises one or more of:
yes or no;
positive, negative, or neutral;
found or not found;
a combination of greater than, less than or equal to;
a location-based input;
categorization using meta-data; and
classification using ML/DL model.
13 . The computer software product claim of claim 9 , wherein a rule of the set of rules that is predefined as a distinct combination of the outcome values of each context-based attribute, represents a decision.
14 . The computer software product claim of claim 9 , wherein applying the set of rules comprises performing rule optimization by one or more of:
a frequency based selection;
a staging approach; and
an exclusion approach.
15 . A system for processing natural language input to answer questions, the system comprising a server configured to:
parse content comprising the natural language input;
transform the natural language input into context-based attributes, wherein transform comprises:
performing a plurality of checks on the content, using the context-based attributes, wherein each context-based attribute is assigned a specific check which generates an outcome represented by an outcome value; and
converting outcome values for each context-based attribute into at least one of binary, ternary, or other discrete forms;
create a rules grid which is a set of rules, used to process the natural language input;
apply the set of rules to make an assessment on relevance of the natural language input for a subject matter of interest through qualitative and quantitative measures, indicating relevance and ranking the content, respectively; and
enable a response to the questions based on the set of rules applied, wherein the response comprises a summary response and a detailed response providing a basis for the assessment of the content.
16 . The system as claimed in claim 15 , wherein the server is configured to define the context-based attributes, which are relevant to the subject matter.
17 . The system as claimed in claim 15 , wherein the server is configured to apply the set of rules to perform rule optimization, by one or more of:
a frequency based selection;
a staging approach; and
a exclusion approach.