Vital text analytics system for the enhancement of requirements engineering documents and other documents

• US 9,678,949 B2
• Filed: 12/13/2013
• Issued: 06/13/2017
• Est. Priority Date: 12/16/2012
• Status: Active Grant

• Current Assignee: Cloud 9 LLC
• Original Assignee: Cloud 9 LLC
• Inventors: Monk, II, Gordon H., Miller, James, Gooding, Donald A., Cowart, Wayne, Kosner, Anthony Wing
• Primary Examiner(s): Sanchez, Michael Ortiz
• Application Number:
  US14/106,667
• Publication Number:
  US 20140172417A1
• Time in Patent Office: 1,278 Days
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A Vital Text Analytics System (VTAS), incorporating a repository of enterprise terms or concepts, is one that improves the readability and fidelity of technical specifications, instructions, training manuals requirements engineering documents and other related engineering documents, typically from a single organization or workgroup. The system stresses ontological analysis of a corpus of related documents, and applies a suite of computational tools that supports the identification and assessment of risk in evaluating the content of the documents, as well as providing statistical measures reflecting the frequency and severity of document features that threaten comprehension.

Claims

1. One or more non-transitory computer-readable media including computer instructions for performing a method comprising:

using a central processing unit (CPU) connected via a network to a remote storage device, to process requirements documents stored in the remote memory device;

identifying, using the CPU, one or more of a plurality of strings consisting of one or more words in a requirements document as corresponding to at least one of a plurality of lexico-syntactic structural patterns;

querying a first database contained in a second memory device, to obtain a list of lexico-syntactic structural patterns associated with ambiguity and interpretive uncertainty arising in the cognitive functions invoked in a reader in reading the text;

annotating, using the CPU, the plurality of strings consisting of one or more words as one of at least a pointer, a linker, and a situator as the structural elements of the lexico-syntactic structural patterns linked to ambiguity and interpretive uncertainty in the cognitive functions invoked in a reader in reading the text;

querying a second database contained in a third memory device, to obtain a set of one or more risk patterns associated with the lexico-syntactic structural patterns, the second database comprising a plurality of risk patterns having at least one of the plurality of known strings consisting of one or more words susceptible to ambiguity and misinterpretation and the plurality of lexico-syntactic structural patterns, the plurality of risk patterns being separate from the plurality of lexico-syntactic structural patterns;

creating, using the CPU, a network having a plurality of nodes corresponding to the sets of lexico-syntactic structural patterns and risk patterns and weighted links between the sets of lexico-syntactic structural patterns and risk patterns;

utilizing, using the CPU, machine learning algorithms to build risk assessors to identify lexico-syntactic structural patterns as risk elements that invoke ambiguity and interpretive uncertainty within the cognitive functions invoked in a reader in reading a text incorporating the lexico-syntactic structural patterns;

constructing, using the CPU, at least one explication that specifies alternative strings consisting of one or more words comprising salient terms from a risk elements; and

displaying, using the CPU, the explications within a user interface to be selected by a user to address ambiguities and misinterpretations within at least one requirements document.

2. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 1 comprising identifying, using the CPU, text of the risk element within the text of the entire document;

and

creating, using the CPU, a document ontology comprising the text of a plurality of risk elements.

3. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 2 comprising identifying, using the CPU, text of the risk element within the text of an entire corpus of documents;

and

creating, using the CPU, a corporate ontology comprising the text of a plurality of risk elements.

4. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 3 comprising constructing, using the CPU, a severity of risk value from the sets of lexico-syntactic structural patterns and risk patterns and weighted links between the sets of lexico-syntactic structural patterns and risk patterns.

5. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 4 comprising constructing, using the CPU, a quality index from the severity of risk value of all risk elements within one of at least the entire document and entire corpus of documents.

6. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 1 comprising:

querying a third database contained in a fourth memory device, to obtain data to construct, using the CPU, a concept library, the data comprising a series of linguistic analyses of terms from one or more requirements documents and from one of at least group sourcing, crowd sourcing, and external resources;

aggregating, using the CPU, the collected data with salient terms and phrases that are stored as an ontological list within the concept library; and

transforming, using the CPU, the structural relationship of a pointer, linker and situator of at least one risk element using the collected data from the concept library;

constructing, using the CPU, explications and alternative strings consisting of one or more words from the salient terms and phrases and data collected within the concept library.

7. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 6 comprising constructing, using the CPU, at least one explication comprising attributes of the salient terms and phrases;

identifying the attributes based on a topology of the salient terms and phrases within a domain or category of relevant terms or concepts; and

storing salient terms and phrases within the corporate ontology based on the topology.

8. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 7 comprising displaying, using the CPU, within the user interface training materials within a curriculum including at least one risk element and a transformation of the risk element displayed as at least one explication as alternative strings consisting of one or more words comprising salient terms from the lexico-syntactic structural patterns and a plurality of alternate explications and representations within the one or more requirements documents to offer alternative strings consisting of one or more words comprising salient terms from the lexico-syntactic structural patterns to replace the text of one or more risk elements identified within the one or more requirements documents.

9. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 6 comprising utilizing, using the CPU, algorithms to perform a series of linguistic analyses to include one of at least frequency of occurrence, location, concordance values, collocation, depth of occurrence, breadth of occurrence data.

10. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 6 comprising:

constructing, using the CPU, a concept map, the concept map comprising mapping of the usage, definitions, synonyms and semantic relationships of the salient terms and phrases.

11. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 10, comprising displaying, using the CPU, collected data and salient terms and phrases in the concept map in different shapes and color coded based on the severity of risk value and quality index.

12. One or more non-transitory computer-readable media including computer instructions for performing a method comprising:

using a central processing unit (CPU) connected via a network to a remote memory device, to process requirements documents stored in the remote memory device;

parsing, using the CPU, text from one or more requirements documents;

converting, using the CPU, the text to a machine readable format;

utilizing, using the CPU, algorithms to perform a plurality of linguistic analyses on the formatted text including the identification of structural components of text;

identifying, using the CPU, one or more of a plurality of strings consisting of one or more words in a requirements document as corresponding to at least one of a plurality of lexico-syntactic structural patterns;

querying a first database contained in a second memory device, to obtain a list of lexico-syntactic structural patterns associated with ambiguity and interpretive uncertainty based on the cognitive functions invoked in a reader in reading the text;

annotating, using the CPU, the structural components of strings consisting of one or more words identified within the text from at least one requirements document as one of at least a pointer, a linker, and a situator as the structural elements of lexico-syntactic structural patterns linked to ambiguity and interpretive uncertainty within the cognitive functions invoked in a reader in reading the text;

compiling, using the CPU, the results of a plurality of linguistic analyses performed and the annotations of the strings consisting of one or more words from at least one sentence into a computational object module to construct risk patterns comprising text susceptible to ambiguity and misinterpretation based on a relationship between all or a part of any text annotated as a pointer, linker, and situator and the identification of the annotated text within a structural class that invokes ambiguity and misinterpretation within the cognitive functions of a reader;

storing the risk patterns within a second database contained in a third memory device;

querying the second database contained in the third memory device, to obtain a set of one or more risk patterns associated with the lexico-syntactic structural patterns, the second database comprising associations between a plurality of risk patterns and at least one of the plurality of known strings consisting of one or more words susceptible to ambiguity and misinterpretation and the plurality of lexico-syntactic structural patterns, the plurality of risk patterns being separate from the plurality of lexico-syntactic structural patterns;

applying, using the CPU, the risk patterns through the comparison of the relationships between all or a part of the text annotated as one of at least a pointer, a linker, and a situator and other annotated text, the risk patterns capable of inducing ambiguities and misinterpretations;

identifying, using the CPU, risk elements as annotated text matching an applied risk pattern;

identifying, using the CPU, salient terms and phrases from the identified risk elements and constructing an ontological list based on the depth and breadth of meaning of a salient term in relation to other terms within the document, salient terms having greater depth present a greater opportunity for ambiguity and misinterpretation, salient terms having greater breadth present a meaning that is different from common usage;

compiling, using the CPU, within the ontological list definitions, synonyms, and semantic relationships of the salient terms to identify alternative elements of language and phrases;

transforming, using the CPU, the structural relationship of one of at least the pointer, linker, and situator of a risk element based on the salient terms and structural class to construct at least one explication as alternative text comprising the alternative elements of language and phrases; and

displaying, using the CPU, the explications within a user interface to be selected by a user to offer options to address ambiguities and misinterpretations within at least one requirements document.

13. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 12, comprising:

utilizing, using the CPU, algorithms to perform highlighting the one or more identified risk elements in the one or more requirements documents; and

creating, using the CPU, a network having a plurality of nodes corresponding to the sets of lexico-syntactic structural patterns and risk patterns and weighted links between the sets of lexico-syntactic structural patterns and risk patterns;

utilizing, using the CPU, algorithms to refine weighted links in the network to assess the one or more risk elements for severity of risk based on the quality of the matching of the risk element to the risk pattern.

14. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 12, comprising:

constructing, using the CPU, an explication template that includes attributes of the salient terms and phrases and identifying the attributes based on the topology of the salient terms and phrases within a domain or category of relevant terms or concepts.

15. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 14 comprising constructing, using the CPU, explications for the salient terms and phrases from the explication templates completed by one or more experts.

16. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 15 comprising storing, using the CPU, the one or more of the explications within a concept library database.

17. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 16 comprising creating, using the CPU, a topology from the one or more identified risk elements and explications to construct an ontology and one or more alternate ontologies.

18. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 17 comprising utilizing, using the CPU, algorithms to rank the requirements document based on the one or more identified risk elements and explications;

and

displaying, using the CPU, the ranking within the one or more requirements documents.

19. The one or more non-transitory computer-readable media including computer instructions for performing a method of claim 17 comprising displaying, using a CPU, a plurality of explications as text;

selecting, using a CPU, one of the plurality of explications to replace a risk element in the one or more requirements documents within training materials.

20. One or more non-transitory computer-readable media including computer instructions for performing a method comprising:

using a central processing unit (CPU) connected via a network to a remote storage device, to process requirements documents stored in the remote memory device;

identifying, using the CPU, one or more of a plurality of strings consisting of one or more words in a requirements document as corresponding to at least one of a plurality of lexico-syntactic structural patterns;

querying a first database contained in a second memory device, to obtain a list of lexico-syntactic structural patterns associated with ambiguity and interpretive uncertainty based on the cognitive functions invoked in a reader in reading the text;

annotating, using the CPU, structural components of the text of at least one requirements document as one of at least a pointer, a linker, and a situator as the structural elements linked to the cognitive functions invoked in a reader in reading the text;

querying a second database contained in a third memory device, to obtain a set of one or more risk patterns associated with the lexico-syntactic structural patterns, the second database comprising associations between a plurality of risk patterns and at least one of the plurality of known strings consisting of one or more words susceptible to ambiguity and misinterpretation and the plurality of lexico-syntactic structural patterns, the plurality of risk patterns being separate from the plurality of lexico-syntactic structural patterns;

utilizing, using the CPU, algorithms to perform a matching comparison of relationships between all or a part of any source text annotated as a pointer, linker, and situator to a plurality of risk patterns, the plurality of risk patterns comprising text annotated as at least one of a pointer, linker, and situator susceptible to ambiguity and misinterpretation; and

identifying, using the CPU, source text matching at least one of the plurality of risk pattern as a risk element and storing the risk element within at least one structural class of risk patterns that invoke ambiguity and misinterpretation within the cognitive functions of a reader;

transforming, using the CPU, the structural relationship of one of at least the pointer, linker, and situator of a risk element to construct at least one explication as alternative text to remove ambiguity or risk of misinterpretation from the risk element; and

displaying, using the CPU, the at least one explication within a user interface to be selected by a user to repair ambiguities and misinterpretations within at least one requirements document.