# US 2019/0171545 A1 · Method and apparatus for test modeling > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0171545 A1 - Document ID: US20190171545A1 - Status: Published application - Kind: A1 - Application number: 15/832,783 - Filed: 2017-12-06 - Published: 2019-06-06 - Earliest claimed: 2017-12-06 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Howard Hess; Eitan Farchi; Orna Raz; Rachel Tzoref-Brill; Aviad Zlotnick - Web page: https://patentgrep.com/patent/US20190171545A1 ## CPC - G06F 11/3608: {using formal methods, e.g. model checking, abstract interpretation (theorem proving G06N5/013)} - G06F 11/3698: Environments for analysis, debugging or testing of software - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A method, system and computer program product, the method comprising: receiving a user document describing at least a part of a system to be tested; computing a similarity measurement between the user document and documents in a document corpus, each describing at least part of a system and associated with a combinatorial model representing the at least part of the system, wherein the combinatorial model comprises a set of attributes, a respective domain for each attribute defining possible values for the attribute, and restrictions, wherein each restriction comprises at least one attribute, a respective value and a logical operator, based upon the similarity measurement, selecting selected documents from the document corpus; obtaining one or more combinatorial models corresponding to the selected documents; selecting elements from the combinatorial models; generating an initial combinatorial model for the user document, wherein said generating comprises adding the elements to the initial combinatorial model. ## Claims ### Claim 1 A computer-implemented method, comprising: receiving a user document describing at least a part of a system to be tested; computing a similarity measurement between the user document and documents in a document corpus, wherein each document in the document corpus describes at least part of a system and is associated with a combinatorial model representing the at least part of the system, wherein the combinatorial model defines a test space, the combinatorial model comprising a set of attributes, a respective domain for each attribute defining possible values for the attribute, and restrictions, wherein each restriction of the restrictions comprises at least one attribute, a respective value thereof and a logical operator; based upon the similarity measurement, selecting one or more selected documents from the document corpus; obtaining one or more combinatorial models corresponding to the one or more selected documents; selecting one or more elements from the one or more combinatorial models; and generating an initial combinatorial model for the user document, wherein said generating comprises adding the one or more elements to the initial combinatorial model. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the restriction comprises an illegal combination of attribute values. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the similarity measurement is determined in accordance with words common to the user document and each document of the corpus of documents. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the similarity measurement is determined based on a computed representation of the user document and signatures of each document of the set of documents. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein the one or more selected documents are selected in accordance with similarity among the one or more selected documents. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein said selecting the one or more elements comprises: selecting common elements appearing in two or more models of the one or more combinatorial models. ### Claim 7 (depends on 6) The computer-implemented method of claim 6, wherein said selecting common elements comprises selecting an attribute appearing in the two or more models. ### Claim 8 (depends on 7) The computer-implemented method of claim 7, wherein said selecting further comprises selecting values from the two or more models for the attribute. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein said selecting the one or more elements comprises: mapping elements of the one or more combinatorial models to features in corresponding documents, wherein the selection is performed based on similarity measurement of the features with features of the user document. ### Claim 10 (depends on 1) The computer-implemented method of claim 1, further comprising outputting a suggestion to a user to add the one or more elements to an initial combinatorial model, wherein adding the one or more elements to the initial combinatorial model subject to user confirmation. ### Claim 11 (depends on 1) The computer-implemented method of claim 1, further comprising: computing a second similarity measurement between the model and each model in a model corpus of existing combinatorial models, wherein the second similarity measurement is computed based on similarity of attributes of the models and respective domains thereof; selecting one or more selected models from the model corpus in accordance with the second similarity measurement computed for models of the existing combinatorial models; determining one or more second elements from the one or more selected models to be provided as a modification suggestion for modifying the partial description; and outputting the modification suggestion to modify the partial description based on the one or more elements, whereby the modification suggestion is useful in completing the partial description to create the combinatorial model. ### Claim 12 A system having a processor, the processor being adapted to perform the steps of: receiving a user document describing at least a part of a system to be tested; computing a similarity measurement between the user document and documents in a document corpus, wherein each document in the document corpus describes at least part of a system and is associated with a combinatorial model representing the at least part of the system, wherein the combinatorial model defines a test space, the combinatorial model comprising a set of attributes, a respective domain for each attribute defining possible values for the attribute, and restrictions, wherein each restriction of the restrictions comprises at least one attribute, a respective value thereof and a logical operator; based upon the similarity measurement, selecting one or more selected documents from the document corpus; obtaining one or more combinatorial models corresponding to the one or more selected documents; selecting one or more elements from the one or more combinatorial models; and generating an initial combinatorial model for the user document, wherein said generating comprises adding the one or more elements to the initial combinatorial model. ### Claim 13 (depends on 12) The system of claim 12, wherein the restriction comprises an illegal combination of attribute values. ### Claim 14 (depends on 12) The system of claim 12, wherein the similarity measurement is determined in accordance with words common to the user document and each document of the corpus of documents. ### Claim 15 (depends on 12) The system of claim 12, wherein said selecting the one or more elements comprises: selecting common elements appearing in two or more models of the one or more combinatorial models. ### Claim 16 (depends on 15) The system of claim 15, wherein said selecting common elements comprises selecting an attribute appearing in the two or more models and selecting values from the two or more models for the attribute. ### Claim 17 (depends on 12) The system of claim 12, wherein said selecting the one or more elements comprises: mapping elements of the one or more combinatorial models to features in corresponding documents, wherein the selection is performed based on similarity measurement of the features with features of the user document. ### Claim 18 (depends on 12) The system of claim 12 wherein the processor is further adapted to output a suggestion to a user to add the one or more elements to an initial combinatorial model, wherein adding the one or more elements to the initial combinatorial model subject to user confirmation. ### Claim 19 (depends on 12) The system of claim 12 wherein the processor is further adapted to: compute a second similarity measurement between the model and each model in a model corpus of existing combinatorial models, wherein the second similarity measurement is computed based on similarity of attributes of the models and respective domains thereof; select one or more selected models from the model corpus in accordance with the second similarity measurement computed for models of the existing combinatorial models; determine one or more second elements from the one or more selected models to be provided as a modification suggestion for modifying the partial description; and output the modification suggestion to modify the partial description based on the one or more elements, whereby the modification suggestion is useful in completing the partial description to create the combinatorial model. ### Claim 20 A computer program product comprising a non-transitory computer readable medium retaining program instructions, which instructions when read by a processor, cause the processor to perform a method comprising: receiving a user document describing at least a part of a system to be tested; computing a similarity measurement between the user document and documents in a document corpus, wherein each document in the document corpus describes at least part of a system and is associated with a combinatorial model representing the at least part of the system, wherein the combinatorial model defines a test space, the combinatorial model comprising a set of attributes, a respective domain for each attribute defining possible values for the attribute, and restrictions, wherein each restriction of the restrictions comprises at least one attribute, a respective value thereof and a logical operator; based upon the similarity measurement, selecting one or more selected documents from the document corpus; obtaining one or more combinatorial models corresponding to the one or more selected documents; selecting one or more elements from the one or more combinatorial models; and generating an initial combinatorial model for the user document, wherein said generating comprises adding the one or more elements to the initial combinatorial model.