Bernardo Nugroho Yahya

Hierarchical classification method based on selective learning of slacked hierarchy for activity recognition systems

Abstract Physical activity recognition using wearable sensors has gained significant interest from researchers working in the field of ambient intelligence and human behavior analysis. The problem of multi-class classification is an important issue in the applications which naturally has more than two classes. A well-known strategy to convert a multi-class classification problem into binary sub-problems is the error-correcting output coding (ECOC) method. Since existing methods use a single classifier with ECOC without considering the dependency among multiple classifiers, it often fails to generalize the performance and parameters in a real-life application, where different numbers of devices, sensors and sampling rates are used. To address this problem, we propose a unique hierarchical classification model based on the combination of two base binary classifiers using selective learning of slacked hierarchy and integrating the training of binary classifiers into a unified objective function. Our method maps the multi-class classification problem to multi-level classification. A multi-tier voting scheme has been introduced to provide a final classification label at each level of the solicited model. The proposed method is evaluated on two publicly available datasets and compared with independent base classifiers. Furthermore, it has also been tested on real-life sensor readings for 3 different subjects to recognize four activities i.e. Walking, Standing, Jogging and Sitting. The presented method uses same hierarchical levels and parameters to achieve better performance on all three datasets having different number of devices, sensors and sampling rates. The average accuracies on publicly available dataset and real-life sensor readings were recorded to be 95% and 85%, respectively. The experimental results validate the effectiveness and generality of the proposed method in terms of performance and parameters.

Similarity Measurement Using Ontology in Vessel Clearance Process

The demands of complicated data communications have issued a new challenge to port logistics systems. Custo- mers expect ports to handle their generated administrative data while a vessel is docked in a port. One port logistics system, known as the Vessel Clearance Process (VCP), manages large numbers of documents related to port of entry. In the VCP, information flows through many organizations such as the port authority, shipping agents, marine offices, immigration offices, and others. Therefore, for effective management of the Business Process (BP) of the VCP, a standardized method of BP modeling is essential, especially in heterogeneous system environments. In a port, according to port policy, terms and data are sued that are similar to but different from those of other logistics partners, which hinders standardized modeling of the BP. In order to avoid tedious and time-consuming document customization work, more convenient modeling of BP for VCP is essential. This paper proposes an ontology-based process similarity measurement to assist designer for process modeling in port domain, especially VCP. We expect that this methodology will use convenient and quick modeling of port business processes.

Concept of Indoor 3D-Route UAV Scheduling System

The objective of the proposed concept is to develop a methodology to support Unmanned Aerial Vehicles (UAVs) operation with a path planning and scheduling system in 3D environments. The proposed 3D path-planning and scheduling allows the system to schedule UAVs routing to perform tasks in 3D indoor environment. On top of that, the multi-source productive best-first-search concept also supports efficient real-time scheduling in response to uncertain events. Without human intervention, the proposed work provides an automatic scheduling system for UAV routing problem in 3D indoor environment.

Contextual activity based Healthcare Internet of Things, Services, and People (HIoTSP): An architectural framework for healthcare monitoring using wearable sensors

Abstract Healthcare industry is gaining a lot of attention due to its technological advancement and the miniaturization in the form of wearable sensors. IoT-driven healthcare industry has mainly focused on the integration of sensors rather than the integration of services and people. Nonetheless, the framework for IoT-driven healthcare applications are significantly lacking. In addition, the use of semantics for ontological reasoning and the integration of mobile applications into a single framework have also been ignored in many existing studies. This work presents the implementation of Healthcare Internet of Things, Services, and People (HIoTSP) framework using wearable sensor technology. It is designed to achieve the low-cost (consumer devices), the easiness to use (interface), and the pervasiveness (wearable sensors) for healthcare monitoring along with the integration of services and agents like doctors or caregivers. The proposed framework provides the functionalities for data acquisition from wearable sensors, contextual activity recognition, automatic selection of services and applications, user interface, and value-added services such as alert generation, recommendations, and visualization. We used the publicly available dataset, PAMAP2 which is a physical activity monitoring dataset, for deriving the contextual activity. Fall and stress detection services are implemented as case studies for validating the realization of the proposed framework. Experimental analysis shows that we achieve, 87.16% accuracy for low-level contextual activities and 84.06%–86.36% for high-level contextual activities, respectively. We also achieved 91.68% and 82.93% accuracies for fall and stress detection services, respectively. The result is quite satisfactory, considering that all these services have been implemented using pervasive devices with the low-sampling rate. The real-time applicability of the proposed framework is validated by performing the response time analysis for both the services. We also provide suggestions to cope with the scalability and security issues using the HIoTSP framework and we intend to implement those suggestions in our future work.

Improved behavior model based on sequential rule mining

Abstract The fourth industrial revolution leads the manufacturing companies to develop future and smart factories by merging automation and digitalization to result a more efficient production method. An evolutionary and competitive experimental approach is necessary to foster the innovation and the rapid change of the automation and digitalization. Consequently, software becomes an important component of industrial automation. One of the major challenge in Industry 4.0 is to industrialize the production of software. Software factory, as one industry with a virtual production line to produce software for manufacturing companies, offers a form of flexible employment, called as telecommuting work. Although this form brings many benefits for both employee and employers, some risks associated with telecommuting work exist. Monitoring the employee behavior is one of the employer way to see the accountability of the employee. Hence, understanding the human behavior during the production process would be an important issue for fulfilling overall operational excellence in software factory. Among approaches proposed to discover the human behavior based on the sequence activities, process mining is one of which has received attentions lately. While most recent process mining approaches in the domain of human behavior address process discovery and post-analysis, few of them have paid attentions on pre-analysis. The pre-analysis is one of the ways to produce a reliable and high-quality of event log which purposely impacts on discovering a daily common behavior and disregarding irregular sequential behavior. This study aims to propose a new way of pre-analysis using sequential rule mining. The key contributions of this research first, is to determine the potential local behaviors using sequential rule mining considering time constraint. Second, the local behaviors are used to enhance event log for discovering relevant behavior model. Third, the mined model is verified by performing conformance checking approach to check the conformity between the behavior model and the real logs based on three measurements: f-measure, ABA, and DMF. The resulting local behaviors, called as rules, can be used for guiding stakeholders to pinpoint the relevant behavior for human capital and productivity enhancement.

RT-PLG: Real Time Process Log Generator

Streaming process mining has been rising as an emergent tool to analyze industrial practices. Obviously, the advance of streaming process mining requires the availability of a suite of real-world business processes and the execution logs in the real time manner. Literally, it is hard to obtain it. This paper aims to develop a real time process log generator for the usage of streaming process mining tool. The real time process log generator (RT-PLG) is constructed in an independent tool. Afterward, the RT-PLG is utilized to generate a synthetic log for streaming process mining. The tool has been evaluated using an existing simulation model.