查看文章

With fast-evolving IoT solutions, the number of smart devices in homes have soared, expected to reach 5 billion by 2025 [13]. The emergence of cloud platforms also allows IoT sensors and actuators to better assist users in various home living activities. In this process, users’ use of devices, ie, device controls, can reflect their behavioral habits and intents. Exploiting the relationship between user behavior habits, intents and their use of devices bring about opportunities from various perspectives. For service providers, such as vendors, knowing users’ living habits through their device usage histories can offer insights for improving user experience. From the perspective of device intelligence, prediction of user behavior can help intelligent platforms recommend actions that users may like to perform, or recommend automation rules that users may be interested in, such as lock the back door when the front door is closed outside. From the perspective of user behavior analysis, accurate user behavior prediction can be used for abnormal user behavior identification, elderly/disabled care. Motivated by these, some prior studies have analyzed user behaviors at home and adopted user behavior prediction technologies for different purposes.[11, 20] infer the users’ behavioral intents from their sequential operations on the device, and generate alternative automation rules and next behavior recommendations.[1, 15, 27] analyze the behavior patterns of users at home, and observe the status of connected smart home entities (sensors and devices) through