Machine-learning based prediction of defect-prone components of information technology assets

1 . An apparatus comprising:

at least one processing device comprising a processor coupled to a memory;

the at least one processing device being configured:

to determine one or more specifications for an information technology asset to be developed;

to identify, utilizing at least one machine learning model, whether at least one of the one or more specifications for the information technology asset is defect-prone, wherein a given one of the one or more specifications is identified as defect-prone responsive to at least one output of the at least one machine learning model indicating that the given specification has at least a threshold likelihood of resulting in one or more defects during development of the information technology asset;

to establish a mapping between the one or more identified defect-prone specifications for the information technology asset and one or more components of the information technology asset; and

to modify one or more development processes for the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications;

wherein modifying the one or more development processes for the one or more components of the information technology asset comprises:

determining defect categories for the one or more identified defect-prone specifications for the information technology asset;

selecting, from a test script repository, one or more test scripts associated with one or more test cases utilized for the determined defect categories for one or more historical development processes to be utilized for testing the one or more components of the information technology asset mapped to the one or more identified defect-prone specifications;

applying the selected one or more test scripts as part of one or more testing processes during the one or more development processes for the information technology asset to identify one or more defects in the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications; and

applying one or more fixes to correct the identified one or more defects in the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications.

2 . The apparatus of claim 1 wherein the information technology asset comprises a software product, and wherein the one or more components of the information technology asset comprise one or more software components of the software product.

3 . The apparatus of claim 2 wherein the one or more software components of the software product comprise one or more functions of the software product.

4 . The apparatus of claim 1 wherein the information technology asset comprises a hardware product, and wherein the one or more components of the information technology asset comprise at least one of one or more hardware components of the hardware product and one or more software components running on the one or more hardware components of the hardware product.

5 . The apparatus of claim 1 wherein the at least one machine learning model comprises at least one classification model, trained on (i) descriptions for historical information technology asset specifications in a historical information technology asset specification dataset and (ii) descriptions for historical defects in a historical defect dataset, configured to determine one or more categories for an input information technology asset specification based at least in part on keywords in a description of the input information technology asset specification.

6 . The apparatus of claim 5 wherein the at least one classification model comprises a Naïve Bayes classifier.

7 . The apparatus of claim 5 wherein the at least one processing device is further configured to merge the historical information technology asset specification dataset and the historical defect dataset to create a merged dataset, the merged dataset comprising, for each of the historical information technology asset specifications, a prone to defect value indicating whether that historical information technology asset specification is associated with one or more of the historical defects.

8 . The apparatus of claim 7 wherein the at least one machine learning model further comprises at least one prediction model, trained utilizing the merged dataset, configured to identify whether the at least one of the one or more specifications for the information technology asset is defect-prone.

9 . The apparatus of claim 1 wherein the at least one machine learning model comprises a Naïve Bayes classifier and a logistic regression model.

10 . The apparatus of claim 9 wherein the Naïve Bayes classifier accounts for a probabilistic nature of features associated with the one or more specifications for the information technology asset and independence between the features associated with the one or more specifications for the information technology asset, and wherein the logistic regression model accounts for underlying relationships between the features associated with the one or more specifications for the information technology asset.

11 . The apparatus of claim 9 wherein the at least one machine learning model comprises an ensemble model combining outputs of the Naïve Bayes classifier and the logistic regression model.

12 . The apparatus of claim 9 wherein identifying whether the at least one of the one or more specifications for the information technology asset is defect-prone is based at least in part on a weighted combination of outputs of the Naïve Bayes classifier and the logistic regression model.

13 . The apparatus of claim 1 wherein modifying the one or more development processes for the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications comprises increasing allocation of resources for development teams performing the one or more development processes for the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications for the information technology asset relative to resources allocated for development teams performing one or more development processes for one or more other components of the information technology asset that are not mapped to the one or more identified defect-prone specifications.

14 . The apparatus of claim 1 wherein the at least one processing device is further configured to deliver one or more notifications to product development teams performing the one or more development processes for the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications, the one or more notifications indicating that the one or more components of the information technology asset are mapped to the one or more identified defect-prone specifications for the information technology asset.

15 . A computer program product comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device:

to determine one or more specifications for an information technology asset to be developed;

to identify, utilizing at least one machine learning model, whether at least one of the one or more specifications for the information technology asset is defect-prone, wherein a given one of the one or more specifications is identified as defect-prone responsive to at least one output of the at least one machine learning model indicating that the given specification has at least a threshold likelihood of resulting in one or more defects during development of the information technology asset;

to establish a mapping between the one or more identified defect-prone specifications for the information technology asset and one or more components of the information technology asset; and

to modify one or more development processes for the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications;

wherein modifying the one or more development processes for the one or more components of the information technology asset comprises:

determining defect categories for the one or more identified defect-prone specifications for the information technology asset;

selecting, from a test script repository, one or more test scripts associated with one or more test cases utilized for the determined defect categories for one or more historical development processes to be utilized for testing the one or more components of the information technology asset mapped to the one or more identified defect-prone specifications;

applying the selected one or more test scripts as part of one or more testing processes during the one or more development processes for the information technology asset to identify one or more defects in the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications; and

applying one or more fixes to correct the identified one or more defects in the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications.

16 . The computer program product of claim 15 wherein the at least one machine learning model comprises at least one classification model, trained on (i) descriptions for historical information technology asset specifications in a historical information technology asset specification dataset and (ii) descriptions for historical defects in a historical defect dataset, configured to determine one or more categories for an input information technology asset specification based at least in part on keywords in a description of the input information technology asset specification.

17 . The computer program product of claim 16 wherein the historical information technology asset specification dataset and the historical defect dataset are merged to create a merged dataset, the merged dataset comprising, for each of the historical information technology asset specifications, a prone to defect value indicating whether that historical information technology asset specification is associated with one or more of the historical defects, and wherein the at least one machine learning model further comprises at least one prediction model, trained utilizing the merged dataset, configured to identify whether the at least one of the one or more specifications for the information technology asset is defect-prone.

18 . A method comprising:

determining one or more specifications for an information technology asset to be developed;

identifying, utilizing at least one machine learning model, whether at least one of the one or more specifications for the information technology asset is defect-prone, wherein a given one of the one or more specifications is identified as defect-prone responsive to at least one output of the at least one machine learning model indicating that the given specification has at least a threshold likelihood of resulting in one or more defects during development of the information technology asset;

establishing a mapping between the one or more identified defect-prone specifications for the information technology asset and one or more components of the information technology asset; and

modifying one or more development processes for the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications;

wherein modifying the one or more development processes for the one or more components of the information technology asset comprises:

determining defect categories for the one or more identified defect-prone specifications for the information technology asset;

selecting, from a test script repository, one or more test scripts associated with one or more test cases utilized for the determined defect categories for one or more historical development processes to be utilized for testing the one or more components of the information technology asset mapped to the one or more identified defect-prone specifications;

applying the selected one or more test scripts as part of one or more testing processes during the one or more development processes for the information technology asset to identify one or more defects in the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications; and

applying one or more fixes to correct the identified one or more defects in the one or more components of the information technology asset that are mapped to the one or more identified defect-prone specifications; and

wherein the method is performed by at least one processing device comprising a processor coupled to a memory.

19 . The method of claim 18 wherein the at least one machine learning model comprises at least one classification model, trained on (i) descriptions for historical information technology asset specifications in a historical information technology asset specification dataset and (ii) descriptions for historical defects in a historical defect dataset, configured to determine one or more categories for an input information technology asset specification based at least in part on keywords in a description of the input information technology asset specification.

20 . The method of claim 19 wherein the historical information technology asset specification dataset and the historical defect dataset are merged to create a merged dataset, the merged dataset comprising, for each of the historical information technology asset specifications, a prone to defect value indicating whether that historical information technology asset specification is associated with one or more of the historical defects, and wherein the at least one machine learning model further comprises at least one prediction model, trained utilizing the merged dataset, configured to identify whether the at least one of the one or more specifications for the information technology asset is defect-prone.