Fuente: WIPO "tomato" A method for performing semantic inference in a system having a processor, memory, and a transceiver arranged to receive a video signal from a target device includes storing a semantic goal, ingesting user guideline content, inferring capabilities semantics, and emulating peripheral input device signals which are communicated to a target device to launch and operate one or more software applications. Data is captured from the video signal and displayed on a user interface control, with the system performing semantic inference on the data and determining whether software application capabilities are factorized with respect to semantic goals, and further performing semantic inference on the data to emulate peripheral input device signals based on the determination.