The default user interface (UI) of software applications is the same for all users, even though users differ in terms of their needs and preferences for using the software. UI customization is typically limited to the most advanced and/or highly active users. As a result, a significant proportion of users of a software do not reap the benefits of having a UI that is personalized to them. This disclosure describes techniques to determine and present a personalized UI to each user or an application, with the user’s permission. UI personalization is performed based on analytics of user-permitted data of user interaction and other relevant information. The analysis can be performed by a suitably trained machine learning model which outputs the optimal personalized UI for each user. Model training and execution is performed on the user device, and if the user permits, on a server that trains the model based on aggregated, non-identifiable user data.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Rendely, Matthew S.; Nyquist, Tommy; Boonsiri, Thomas; Bansal, Tarun; Crouse, Michael; Gorski, Filip; Hu, Mingpu; Trainor, David; and Gupta, Akancha, "Automatic Personalization of User Interfaces based on User Interaction Analytics", Technical Disclosure Commons, (August 12, 2021)