A system and method to identify and implement applicationspecific screen rotation settings on a mobile are disclosed. The proposed method extends the operating system to use multiple signals, including sensor readings, accelerometer inputs, face detection algorithms, location based signals, or contextual signals as inputs. The rotation setting preferences specific to an application are aggregated based on the multiple signals feeding into a decision algorithm. The decision algorithm is configurable and learns rotation preferences specific to an application or application category. The decision algorithm can implement decisions based on a fixed combination of functions with manually chosen weights for the inputs. The combination of functions can also be learned using machine learning methods such as neural networks, support vector machines, or decision trees. The method further addresses the problem of erroneous rotation of screen content using a smart approach to detect the user’s preference.

Creative Commons License

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