• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.11
• /
• pp.2839-2852
• /
• 2013

Context-awareness is an essential part of ubiquitous computing, and over the past decade video based activity recognition (VAR) has emerged as an important component to identify user's context for automatic service delivery in context-aware applications. The accuracy of VAR significantly depends on the performance of the employed human body segmentation algorithm. Previous human body segmentation algorithms often engage modeling of the human body that normally requires bulky amount of training data and cannot competently handle changes over time. Recently, active contours have emerged as a successful segmentation technique in still images. In this paper, an active contour model with the integration of Chan Vese (CV) energy and Bhattacharya distance functions are adapted for automatic human body segmentation using depth cameras for VAR. The proposed technique not only outperforms existing segmentation methods in normal scenarios but it is also more robust to noise. Moreover, it is unsupervised, i.e., no prior human body model is needed. The performance of the proposed segmentation technique is compared against conventional CV Active Contour (AC) model using a depth-camera and obtained much better performance over it.

• Journal of Korean Institute of Industrial Engineers
• /
• v.29 no.2
• /
• pp.165-171
• /
• 2003

Visual search in an unstructured search field is a fruitful research area for computational modeling. Search models that describe relationship between search time and probability of target detection have been used for prediction of human search performance and provision of ideal goals for search training. Until recently, however, most of models were focused on detecting a single target in a search field, although, in practice, a search field includes multiple targets and search models for multiple targets may differ from search models for a single target. This study proposed a random search model for multiple targets, generalizing a random search model for a single target which is the most typical search model. To test this model, human search data were collected and compared with the model. This model well predicted human performance in visual search for multiple targets. This paper also proposed how to determine optimal stopping time in multiple-target search.

• Educational Technology International
• /
• v.10 no.1
• /
• pp.107-135
• /
• 2009

The purpose of this paper is to review a variety of challenges facing the Human Performance Technology (HPT) in supporting knowledge workers' performance, and to explore possible HPT models for knowledge workers. The first section of this paper investigates the core attributes and major models of HPT as a foundation. While HPT has a lot of strengths in terms of systemic, systematic, methodologically eclectic, evidence based, and results oriented approaches, some pitfalls - which could be detected if these principles were mindlessly applied to problem areas - are explored. The second section presents some considerations such as analysis, intervention design, and leadership that HP technologists need to take in order to make HPT a better field of practice for knowledge workers. The author also suggests a tentative diagnostic model and a process model for knowledge workers, core principles of which are based on systems thinking, in particular Senge's the fifth discipline and Checkland's soft systems methodology. The importance of formative evaluations to improve these models is noted as a conclusion.

• Journal of the Ergonomics Society of Korea
• /
• v.15 no.2
• /
• pp.157-163
• /
• 1996

A single line display menu (SDM) is widely used for the user interface of many electronic consumer products, and the designers need useful guidelines applicable to the SDM. In many studies on menus, major focus has been placed on the optimal menu structure, but only a few standard menu structures, such as $64^{1},8^{2},4^{3}$,and $2^{6}$ are usually tested for optimality. In many cases, however, ill defined or asymmetric structures are suggested as design alternatives. To determine the optimal menu structure, user performance should be obtained in terms of quantitative measures. Hence, a model is needed to provide a predicted value of user performance for a given menu structure. Altough several models have been proposed for ordinary menus, none is available for the SDM yet. To solve this problem a performance model was developed in this study using the neural network approach. This model is capable of providing quantitative measures of human performance for any menu structures without conducting additional experiments, which will save much time and reduce the design cost.

• International Journal of Quality Innovation
• /
• v.8 no.3
• /
• pp.1-10
• /
• 2007

A sample of Italian manufacturing companies was selected in order to verify the abilities and effects (relationships) of the management factors human resources, leadership and strategic planning on company performance. The Partial Least Squares (PLS) estimation method was used for analyzing the data collected, where the relationships between the management factors and performance were formalized by a Structural Equation Model (SEM). The analysis of the survey data showed unexpected result regarding the non significant direct relationship between Leadership and Performance. The effect of Leadership is obtained by an indirect relationship through Human Resources. The combination of Leadership and Human Resources has hence been identified as the management factors which have the highest impact on the performance of Italian industrial companies. Another interesting and unexpected result was that there was no significant impact of Strategic Planning on Performance. It seems that the leaders of Italian industrial companies have not understood that good strategic planning is a necessary condition for achieving excellence. So another improvement area is in fact Strategic Planning. This area should have the highest priority of any top management team and the focus should include how to establish a strong relationship between strategic planning and performance. No correlation between strategic planning and performance is a strong indication that something is wrong. It is not enough that Leadership is doing Strategic Planning-Leadership is also about studying and follow up on results in order to assure impacts on performance. This link seems to be missing in Italian industrial companies.

• Journal of the Korea Management Engineers Society
• /
• v.23 no.4
• /
• pp.159-164
• /
• 2018

Human-machine serial systems must be normal in both systems. Though the failure of machine is irreducible by itself, the human errors are of recurring type. When the human performance is described quantitatively, non-homogeneous Poisson Process model of human errors can be developed. And the model parameters can be estimated by maximum likelihood estimation and numerical analysis method. System reliability is obtained by multiplying machine reliability by human reliability.

• Journal of the Ergonomics Society of Korea
• /
• v.32 no.1
• /
• pp.59-68
• /
• 2013

Objective: The purpose of this paper is to introduce a task complexity model combining task design aspects and complexity dimensions and to explain an approach to identifying and organizing task complexity factors based on the model. Background: Task complexity is a critical concept in describing and predicting human performance in complex systems such as nuclear power plants(NPPs). In order to understand the nature of task complexity, task complexity factors need to be identified and organized in a systematic manner. Although several methods have been suggested for identifying and organizing task complexity factors, it is rare to find an analytical approach based on a theoretically sound model. Method: This study regarded a task as a system to be designed. Three levels of design ion, which are functional, behavioral, and structural level of a task, characterize the design aspects of a task. The behavioral aspect is further classified into five cognitive processing activity types(information collection, information analysis, decision and action selection, action implementation, and action feedback). The complexity dimensions describe a task complexity from different perspectives that are size, variety, and order/organization. Combining the design aspects and complexity dimensions of a task, we developed a model from which meaningful task complexity factors can be identified and organized in an analytic way. Results: A model consisting of two facets, each of which is respectively concerned with design aspects and complexity dimensions, were proposed. Additionally, twenty-one task complexity factors were identified and organized based on the model. Conclusion: The model and approach introduced in this paper can be effectively used for examining human performance and human-system interface design issues in NPPs. Application: The model and approach introduced in this paper could be used for several human factors problems, including task allocation and design of information aiding, in NPPs and extended to other types of complex systems such as air traffic control systems as well.

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.59-70
• /
• 2024

We introduce a high-performance named entity recognition (NER) model for written and spoken language. To overcome challenges related to labeled data scarcity and domain shifts, we use transfer learning to leverage our previously developed KorBERT as the base model. We also adopt a meta-pseudo-label method using a teacher/student framework with labeled and unlabeled data. Our model presents two modifications. First, the student model is updated with an average loss from both human- and pseudo-labeled data. Second, the influence of noisy pseudo-labeled data is mitigated by considering feedback scores and updating the teacher model only when below a threshold (0.0005). We achieve the target NER performance in the spoken language domain and improve that in the written language domain by proposing a straightforward rollback method that reverts to the best model based on scarce human-labeled data. Further improvement is achieved by adjusting the label vector weights in the named entity dictionary.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2002.05a
• /
• pp.163-168
• /
• 2002

The road sign in dynamic traffic system is an important element which affects on human cognitive performance on driving. Web-based vision system simulator was developed to examine the cognition time of the road sign in dynamic environment. This experiment was designed in within-subject design with two factors; vehicle speed and the amount of information of the traffic sign. It measured the cognition time of the road sign through two evaluation methods; the subjective test with vision system simulator and computational cognitive model. In these two evaluations of human cognitive performance under the dynamic traffic environment, it demonstrated that subject's cognition time was affected by both the amount of information of traffic sign and driving speed.