Search

Human activity recognition (HAR) using inertial motion data has gained a lot of momentum in recent years both in research and industrial applications. From the abstract perspective, this has been driven by the rapid dynamics for building intelligent, smart environments, and ubiquitous systems that cover all aspects of human life including healthcare, sports, manufacturing, commerce, etc., which necessitate and subsume activity recognition aiming at recognizing the actions, characteristics, and goals of one or more agent(s) from a temporal series of observations streamed from one or more sensors. From a more concrete and seemingly orthogonal perspective, such momentum has been driven by the ubiquity of inertial motion sensors on-board mobile and wearable devices including smartphones, smartwatches, etc. In this paper we give an introductory and a comprehensive survey to the subject from a given perspective. We focus on a subset of topics, that we think are major, that will have significant and influential impacts on the future research and industrial-scale deployment of HAR systems. These include: (1) a comprehensive and detailed description of the inertial motion benchmark datasets that are publicly available and/or accessible, (2) feature selection and extraction techniques and the corresponding learning methods used to build workable HAR systems; we survey classical handcrafted datasets as well as data-oriented automatic representation learning approach to the subject, (3) transfer learning as a way to overcome many hurdles in actual deployments of HAR systems on a large scale, (4) embedded implementations of HAR systems on mobile and/or wearable devices, and finally (5) we touch on adversarial attacks, a topic that is essentially related to the security and privacy of HAR systems. As the field is very huge and diverse, this article is by no means comprehensive; it is though meant to provide a logically and conceptually rather complete picture to advanced practitioners, as well as to present a readable guided introduction to newcomers. Our logical and conceptual perspectives mimic the typical data science pipeline for state-of-the-art AI-based systems. [ABSTRACT FROM AUTHOR]

Feature extraction, feature clustering, feature selection are suitable to enhance learning performance, reduce computational complexity, create better generalizable models, and reduce required storage. Although there are several feature reduction techniques, still it remains one of the research hot spots in the field of data mining and machine learning groups. Owl search algorithm (OSA) is one of the recent metaheuristic optimization algorithms that mimic the hunting mechanism of owls in dark. However, OSA suffers from the same problem faced by many other optimization algorithms and tends to fall in local optima and premature convergence. To overcome these problems, two improvements for OSA algorithm are proposed in this paper. The first improvement uses a fuzzy system which is responsible to tune a control parameter in the updating phase of OSA. The second improvement includes the fuzzy system along with modifying the updating equation of OSA to enhance the exploration activity. In addition, this paper presents a new feature extraction technique for regression problems based on the improved OSA, called Fuzzy Owl Clustering Dimension Reduction (FOCDR). We apply a method that uses three weighting methods (i.e., soft, hard, and mixed) to extract new features based on the generated clusters. The experiment is divided into two parts. In the first part, the performance of the OSA algorithm and two improvement versions are analyzed with ten benchmark functions. The results show that the proposed versions on average can improve the convergence rate by 6.75% and 14.2% compared to OSA in solving complex problems. The second part is conducted to show FOCDR's ability for feature selection problems. The effectiveness of FOCDR has been evaluated using four benchmark datasets and a real-world case study. [ABSTRACT FROM AUTHOR]

Spike trains (STs) induced by external stimuli are complex and challenging to decode the embedded spatiotemporal information. Although various methods have been developed to characterize STs, few applications have been reported in tactile discrimination applications. In this paper, a neurocomputing method based on reference spike train (RST) is proposed to establish a neural computation scheme, on which existing ST metrics could be fed into various traditional models directly. Moreover, existing metrics in the field of statistics and vector measurement are introduced together to extract more discriminative features. With the binning technique and feature selection algorithm applied, the neural computation scheme is aimed at taking advantage of as maximal as possible information contained in tactile signals. Based on our designed artificial fingertip, the effect is validated by improving the recognition accuracy from 77.6% to 83.4% when it is applied to the discrimination of eight roughness surfaces. Furthermore, properties of RST, such as spike intervals and distributions, are evaluated and it is found that RSTs with uniform distribution perform the best for tactile discrimination. [ABSTRACT FROM AUTHOR]