What is the problem you are seeing? Please describe.
The original, core idea of App Mining were "aiming to build an algorithmic, automated mechanism for evaluating app quality that could also distribute monthly rewards for app developers."
What we have learned is that high rank in app mining does not imply than an app is of high quality as perceived by the users and the community. Yet rank is used to showcase the top apps.
How is this problem misaligned with goals of app mining?
Users and the community are disappointed about the quality of apps. They find top apps that are perceived to be of inadequate quality.
What is the explicit recommendation you’re looking to propose?
Use other criteria than rank in app mining when recommending apps to users, such as selections of apps vetted by the community.
Describe your long term considerations in proposing this change. Please include the ways you can predict this recommendation could go wrong and possible ways mitigate.
Factoring out quality assessment allows using other targets than app quality to guide the app mining algorithm, providing flexibility moving forward.
We are not limited to "design a ranking system that rewards apps that deliver a high-quality user experience" but can design a reward system that produces quality apps without making it the ranking criteria.
What is the problem you are seeing? Please describe.
The original, core idea of App Mining were "aiming to build an algorithmic, automated mechanism for evaluating app quality that could also distribute monthly rewards for app developers."
What we have learned is that high rank in app mining does not imply than an app is of high quality as perceived by the users and the community. Yet rank is used to showcase the top apps.
How is this problem misaligned with goals of app mining?
Users and the community are disappointed about the quality of apps. They find top apps that are perceived to be of inadequate quality.
What is the explicit recommendation you’re looking to propose?
Use other criteria than rank in app mining when recommending apps to users, such as selections of apps vetted by the community.
Describe your long term considerations in proposing this change. Please include the ways you can predict this recommendation could go wrong and possible ways mitigate.
Factoring out quality assessment allows using other targets than app quality to guide the app mining algorithm, providing flexibility moving forward.
We are not limited to "design a ranking system that rewards apps that deliver a high-quality user experience" but can design a reward system that produces quality apps without making it the ranking criteria.
Additional context
223 Stakeholder analysis
219 App mining with two tiers