A PSL-based Approach to Human Activity Recognition in Smart Home Environments

Human activity recognition is widely used in smart cities, public safety and other fields, especially home systems where it has a pivotal role. The study addresses the shortcomings of Markov logic networks for human proposes method scenarios - an framework based on Probabilistic Soft Logic (PSL). able to deal with logical uncertainty problems provides expression inference mechanisms data this basis. utilizes Deng entropy evidence theory provide evaluation sensor event uncertainty, combines calculus modeling. Comparing PSL three common methods, Ontology, Hidden Model (HMM), network, dataset, was found that much better ability handle than algorithms. average rates ADL ADL-E sub datasets were 82.87% 80.33%, respectively. In experiments verify temporal complexity, showed least significant decrease rate maintained 81.02% presence concurrent alternating activities. performance handling both complexity.