CONVERTING PROCEDURAL TEXT TO AN ACTIONABLE KNOWLEDGE FORM

• US 20150142825A1
• Filed: 11/14/2014
• Published: 05/21/2015
• Est. Priority Date: 11/15/2013
• Status: Active Grant

• Current Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Original Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Inventors: Deshmukh, Veena S., Kelkar, Rahul Ramesh, Dhagat, Mudit, Chougule, Sandeep
• Granted Patent: US 10,922,614 B2
• Time in Patent Office: Days
• US Class Current: 707/748
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

Disclosed is method and system for converting text used to perform an operation to reusable actionable knowledge form. Procedural knowledge available in run books is converted into reusable actionable knowledge form. The method comprises structuring the text by performing merging, grouping, editing, removing statements and marking statements present in structured text into action segments, predicates, and comments, by parsing technique and rule based reasoning. Predicate comprise conditions, action segment comprise actionable statements executed upon fulfilling conditions. Actionable statement is used to perform a task of the operation. Action segments are mapped with predicates to generate predicate-action pairs, standard operators relevant to each of conditions of predicate and actionable statements of action segment are selected, and score for standard operators is determined, and a standard operator having highest score is linked with conditions and actionable statements of the predicate-action pair, thereby converting predicate-action pair in reusable actionable knowledge form.

Claims

1. A method for converting text contained in a document to an actionable knowledge form, wherein the text is used to perform an operation, the method comprising:

structuring, by a processor, the text by performing at least one of merging, grouping, editing, or removing statements present in the text to generate structured text;

marking, by the processor, one or more statements present in the structured text into one or more categories, wherein the one or more categories comprise an action segment, a predicate, and a comment, wherein the predicate comprises one or more conditions, and wherein the action segment comprises one or more actionable statements executed upon fulfilling the one or more conditions, and wherein the one or more actionable statements are used to perform one or more tasks during execution of the operation;

mapping, by the processor, the action segment with the predicate to generate a predicate-action pair; and

selecting, by the processor, one or more standard operators relevant to each of the one or more conditions of the predicate and each of the one or more actionable statements of the action segment, of the predicate-action pair;

determining, by the processor, a score for the one or more standard operators relevant to each of the one or more conditions of the predicate and each of the one or more actionable statements of the action segment, of the predicate-action pair; and

linking, by the processor, a standard operator having a highest score with at least one of the one or more conditions of the predicate and the one or more actionable statements of the action segment, of the predicate-action pair, thereby converting the predicate-action pair in an actionable knowledge form.

2. The method of claim 1, wherein the document is one of a operating procedure document, a service document, a run book, a service manual, and wherein the document is in a text form, and wherein the text comprises procedural knowledge to perform the operation, and the text is in English Language.

3. The method of claim 1, wherein the actionable knowledge form is reusable, and the actionable knowledge form is converted to computer executable form.

4. The method of claim 1, comprises linking a standard operator having a highest score, from the one or more standard operators relevant to each of the one or more conditions, and the one or more standard operators relevant to each of the one or more actionable statements, with each of the one or more conditions and each of the one or more actionable statements respectively.

5. The method of claim 1, wherein the structuring comprises:

segmenting the text into independent textual blocks; and

segmenting the textual blocks into the statements by using a boundary detection technique.

6. The method of claim 1, wherein the one or more statements are marked into the one or more categories by using a parsing technique and a rule based reasoning.

7. The method of claim 1, wherein the selecting comprises:

locating, by parsing, one or more keywords from the one or more conditions and the one or more actionable statements; and

matching the one or more keywords with one or more key markers of the one or more standard operators to select the one or more standard operators.

8. The method of claim 7, wherein the score for the one or more standard operators is determined based on a degree of matching of the one or more keywords with the one or more key markers of the one or more standard operators.

9. The method of claim 1, further comprising executing the standard operator with the highest score for executing the one or more conditions and the one or more actionable statements, associated with the one or more task, wherein the standard operator is converted to computer executable form.

10. A system for converting text contained in a document to an actionable knowledge form, wherein the text is used to perform an operation, the system comprising:

a processor; and

a memory coupled to the processor, wherein the processor is capable of executing a plurality of modules stored in the memory, and wherein the plurality of modules comprising;

a structuring module for

structuring the text by performing at least one of merging, grouping, editing, or removing statements present in the text to generate structured text;

marking one or more statements present in the structured text into one or more categories, wherein the one or more categories comprise an action segment, a predicate, and a comment, wherein the predicate comprises one or more conditions, and wherein the action segment comprises one or more actionable statements executed upon fulfilling the one or more conditions, and wherein the one or more actionable statements are used to perform one or more tasks during execution of the operation;

a mapping module for mapping the action segment with the predicate to generate a predicate-action pair; and

a selecting module for

selecting one or more standard operators relevant to each of the one or more conditions of the predicate and each of the one or more actionable statements of the action segment, of the predicate-action pair;

determining a score for the one or more standard operators relevant to each of the one or more conditions of the predicate and each of the one or more actionable statements of the action segment, of the predicate-action pair; and

linking a standard operator having a highest score with at least one of the one or more conditions of the predicate and the one or more actionable statements of the action segment, of the predicate-action pair, thereby converting the predicate-action pair in an actionable knowledge form.

11. The system of claim 10, wherein the document is one of a operating procedure document, a service document, a run book, a service manual, and wherein the document is in a text form, and wherein the text comprises procedural knowledge to perform the operation, and the text is in English Language.

12. The system of claim 10, wherein the actionable knowledge form is reusable, and the actionable knowledge form is converted to computer executable form.

13. The system of claim 10, wherein the selecting module,

locates, by parsing, one or more keywords from the one or more conditions and the one or more actionable statements;

and

matches the one or more keywords with one or more key markers of the one or more standard operators to select the one or more standard operators.

14. A computer program product having embodied thereon a computer program for converting text contained in a document to an actionable knowledge form, wherein the text is used to perform an operation, the computer program product comprising:

a program code for structuring the text by performing at least one of merging, grouping, editing, or removing statements present in the text to generate structured text;

a program code for marking one or more statements present in the structured text into one or more categories, wherein the one or more categories comprise an action segment, a predicate, and a comment, wherein the predicate comprises one or more conditions, and wherein the action segment comprises one or more actionable statements executed upon fulfilling the one or more conditions, and wherein the one or more actionable statements are used to perform one or more tasks during execution of the operation;

a program code for mapping the action segment with the predicate to generate a predicate-action pair;

a program code for selecting one or more standard operators relevant to each of the one or more conditions of the predicate and each of the one or more actionable statements of the action segment, of the predicate-action pair;

a program code for determining a score for the one or more standard operators relevant to each of the one or more conditions of the predicate and each of the one or more actionable statements of the action segment, of the predicate-action pair; and

a program code for linking a standard operator having a highest score with at least one of the one or more conditions of the predicate and the one or more actionable statements of the action segment, of the predicate-action pair, thereby converting the predicate-action pair in an actionable knowledge form.