This disclosure describes techniques for the adaptive presentation of tasks on user interfaces. With user permission and express consent, information regarding user interaction with different tasks on a computing device is obtained along with context information. Machine learning techniques are utilized to determine the likelihood of a specific task to be activated in the current context, e.g., based on the similarity of the current context to the set of past contexts when the task was active. The current context and the set of previously active contexts constitute a set of input features for the machine learning model, and the probability that a task is active in the current context constitutes an output label from the machine learning model. The prominence of displayed tasks on the user interface is adjusted based on user context. Tasks that are likely to be of interest of the user in a current context are displayed more prominently.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.