Usability data analysis platform
This disclosure describes a platform and taxonomy for analyzing the results of testing data, such as data generated through usability test of the User Experience (UX) of a software product or service. The analysis outputs actions that can be performed automatically to modify the software product or service under test. The usability test results that are analyzed can include qualitative data, such as words, sentences, phrases, comments, and or other (e.g., natural language) information that have been provided by end-users engaged in the usability testing. The platform and taxonomy described herein provided a framework through which a semantic analysis of results can be performed, with actions performed based on the semantic analysis.
1. A system comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor, and storing instructions which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
receiving input data generated through usability testing of software, the input data including comments created by end-users interacting with the software;
generating one or more clusters of the comments, each cluster including a subset of the comments that are semantically similar;
determining a role fit metric for each of the end-users comprising:
comparing characteristics of a respective end-user to characteristics of a previously determined persona, wherein the role fit metric for the respective end-user indicates a percentage degree of similarity between the characteristics of the respective end-user and the characteristics of the previously determined persona, and wherein the characteristics comprise one or more of an end-user role, a role description of responsibilities, a role-fit for a particular task, or a set of skills;
calculating, for each of the one or more clusters, an importance metric for a respective cluster, wherein the importance metric is based on a proportion of the end-users that generated references to the comments in the respective cluster, and wherein the proportion is determined based on weighting the references of each of the end-users according to a percentage of the role fit metric of the respective end-user;
generating, for each of the one or more clusters, one or more recommendations to address an issue described in the comments included in the respective cluster, wherein the one or more recommendations are generated based at least partly on the importance metric for the respective cluster; and
providing the one or more recommendations through an interface.
2. The system of claim 1 , the operations further comprising:
determining a priority of addressing the issue associated with the respective cluster, the priority determined based on the importance metric of the respective cluster,
wherein the one or more recommendations are based at least partly on the priority.
3. The system of claim 1 , the operations further comprising:
determining an impact, to the software of the issue associated with the respective cluster,
wherein the one or more recommendations are based at least partly on the impact.
4. The system of claim 3 , wherein the impact is determined through at least one machine learning algorithm.
5. The system of claim 1 , wherein providing the one or more recommendations through the interface includes presenting the one or more recommendations through a user interface.
6. The system of claim 1 , wherein providing the one or more recommendations through the interface includes providing the one or more recommendations through at least one application programming interface (API) that is accessed by one or more services.
7. The system of claim 1 , the operations further comprising:
generating a name for each of the one or more clusters by applying natural language processing to the comments in the respective cluster.
8. The system of claim 1 , the operations further comprising:
determining a usability issue type for each of the one or more clusters by applying natural language processing to the comments in the respective cluster.
9. The system of claim 8 , the operations further comprising:
determining a usability issue judgment for each usability issue type by applying the natural language processing to the comments in the respective cluster.
10. The system of claim 1 , the operations further comprising:
determining, for the respective cluster, one or more of an associated screen and an associated use case.
11. A computer-implemented method performed by at least one processor, the method comprising:
receiving input data generated through usability testing of software, the input data including comments created by end-users interacting with the software;
generating one or more clusters of the comments, each cluster including a subset of the comments that are semantically similar;
determining a role fit metric for each of the end-users comprising:
comparing characteristics of a respective end-user to characteristics of a previously determined persona, wherein the role fit metric for the respective end-user indicates a percentage degree of similarity between the characteristics of the respective end-user and the characteristics of the previously determined persona, and wherein the characteristics comprise one or more of an end-user role, a role description of responsibilities, a role-lit for a particular task, or a set of skills;
calculating, for each of the one or more clusters, an importance metric for a respective cluster, wherein the importance metric is based on a proportion of the end-users that generated references to the comments in the respective cluster, and wherein the proportion is determined based on weighting the references of each of the end-users according to a percentage of the role fit metric of the respective end-user;
generating, for each of the one or more clusters, one or more recommendations to address an issue described in the comments included in the respective cluster, wherein the one or more recommendations are generated based at least partly on the importance metric for the respective cluster; and
providing the one or more recommendations through an interface.
12. The method of claim 11 , the method further comprising:
determining a priority of addressing the issue associated with the respective cluster, the priority determined based on the importance metric of the respective cluster,
wherein the one or more recommendations are based at least partly on the priority.
13. The method of claim 11 , the method further comprising:
determining an impact, to the software, of the issue associated with the respective cluster, wherein the one or more recommendations are based at least partly on the impact.
14. The method of claim 13 , wherein the impact is determined through at least one machine learning algorithm.
15. The method of claim 11 , wherein providing the one or more recommendations through the interface includes presenting the one or more recommendations through a user interface.
16. The method of claim 11 , wherein providing the one or more recommendations through the interface includes providing the one or more recommendations through at least one application programming interface (API) that is accessed by one or more services.
17. The method of claim 11 , the method further comprising:
generating a name for each of the one or more clusters by applying natural language processing to the comments in the respective cluster.
18. The method of claim 11 , the method further comprising:
determining a usability issue type for each of the one or more clusters by applying natural language processing to the comments in the respective cluster; and
determining a usability issue judgment for each usability issue type by applying the natural language processing to the comments in the respective cluster.
19. The method of claim 11 , the method further comprising: determining, for the respective cluster, one or more of an associated screen and an associated use case.
20. One or more non-transitory computer-readable storage media storing instructions which, when executed by at least one processor, cause the at least one processor to perform operations comprising:
receiving input data generated through usability testing of software, the input data including comments created by end-users interacting with the software;
generating one or more clusters of the comments, each cluster including a subset of the comments that are semantically similar;
determining a role fit metric for each of the end-users comprising:
comparing characteristics of a respective end-user to characteristics of a previously determined persona, wherein the role fit metric for the respective end-user indicates a percentage degree of similarity between the characteristics of the respective end-user and the characteristics of the previously determined persona, and wherein the characteristics comprise one or more of an end-user role, a role description of responsibilities, a role-fit for a particular task, or a set of skills;
calculating, for each of the one or more clusters, an importance metric for a respective cluster, wherein the importance metric is based on a proportion of the end-users that generated references to the comments in the respective cluster, and wherein the proportion is determined based on weighting the references of each of the end-users according to a percentage of the role fit metric of the respective end-user;
generating, for each of the one or more clusters, one or more recommendations to address an issue described in the comments included in the respective cluster, wherein the one or more recommendations are generated based at least partly on the importance metric for the respective cluster; and
providing the one or more recommendations through an interface.