Device-Aware Rule Recommendation for the Internet of Things

摘要: With over 34 billion IoT devices to be installed by 2020, the Internet of Things (IoT) is fundamentally changing our lives. One greatest benefits powerful automations achieved applying rules devices. For instance, a rule named "Make me cup coffee when I wake up'' automatically turns on machine sensor in bedroom detects motion morning. large numbers possible out there, recommendation system great necessity help users find they need. However, little effort has been made design model tailored for recommendation, which comes with lots new challenges compared traditional tasks. We not only need re-define "users'' and "items'' task, but also have consider type entities, devices, extra information constraints brought them. To handle these challenges, we propose novel efficient algorithm, considers implicit feedback rules, takes user-rule-device interactions match between device requirements user possessions into account. In collaboration Samsung, one leading companies this field, designed an framework evaluated algorithm real-life industry dataset. Experiments show effectiveness efficiency method.