A hybrid approach to inferring the Internet of Things for complex activity recognition

Abstract

With the rapid development and large-scale uptake of the Internet of Things, smart home is evolving from a vision towards a realistically viable solution for assisted living. Activity recognition is one of the fundamental tasks in order to provide accurate and timely assistance and service. As daily living scenarios are full of similar activities, missing data, and noise, inferring complex activities using knowledge-driven reasoning algorithms suffers from several drawbacks, e.g., real-time raw sensor data segmentation, poor generalization, higher computational complexity, and scalability. To address these problems, this paper proposes a hybrid approach to complex daily activity recognition by merging the first-order logic and probability graphic modeling. Specifically, we develop a novel "Markov logic network" combining data-driven multi-feature and simplified rule-based modeling and inference, thus enabling and supporting the applicability and robustness of daily activity recognition. To evaluate the approach and associated methods, we design a testing scenario with a number of similar activity groups, missing data, or disturbance test datasets in a multi-modeling sensor scene. Initial results show our approach outperforms the traditional approach with a better accuracy in the situations of similar activities with missing data and noise disturbance. Experiments are also conducted to compare the Gibbs sampling and MC-SAT sampling algorithms for Markov logic network, and the results show that the Gibbs is better in our experimental settings.

Multi-resident type recognition based on ambient sensors activity

“…Because of the difference between this work and other resident recognition, this work tries to inference the resident type, not resident role which can not do the comparsion experiments. We compared the results with the [11] model (without any preferences). We can easily find the [11] model (without any preferences) just has the bad performance because of the computation of classical probability, the comparsion experiments will not be repeated anymore in the following experiments.…”

“…We compared the results with the [11] model (without any preferences). We can easily find the [11] model (without any preferences) just has the bad performance because of the computation of classical probability, the comparsion experiments will not be repeated anymore in the following experiments. The time sequence of an activity is a sequence of several ordered events, like, turning on faucet, feeling the dropping water, taking the mop, shaking the mop, and then pressing the mop.…”

“…Gayathri et al [19] have adopted Markov Logic Network (MLN) which combined the first-order logic and Markov Network to recognize the activities but set the weights manually without statistical probability which is not the real combination. In 2019, we have adopted MLN to handle the noisy and data missing problems in activity recognition which get good performance [11]. We adopt the MLN in resident recognition in this paper which will be detailed introduced in section 2.…”

Revolutionary Strategies Analysis and Proposed System for Future Infrastructure in Internet of Things

“…Luk et al [37] addressed System Security Networking (SSN), recovery and privacy safety, and authentication. By combining first-order reasoning and probability graphic simulation, Li et al [38] proposed a hybrid approach to complex everyday task identification. They created a novel "Markov logic network" that combines data-driven multi-feature modeling and inference with simpler rules-based modeling and inference, allowing and promoting the applicability and robustness of everyday activity recognition.…”

“…IoT and IPv6 integration [25,32,73] Services discovery protocols [34,38] Discuss challenges and approaches, internet protocol IoT incorporation, authentication, and authentication problems.…”