• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.3
• /
• pp.346-354
• /
• 2024

Deep learning models require high computing power due to a substantial amount of computation. It is difficult to use them in devices with limited computing environments, such as coastal surveillance equipments. In this study, a lightweight model is constructed by analyzing the weight changes of the convolutional layers during the training process based on MobileNet and then pruning the layers that affects the model less. The performance comparison results show that the lightweight model maintains performance while reducing computational load, parameters, model size, and data processing speed. As a result of this study, an effective pruning method for constructing lightweight deep learning models and the possibility of using equipment resources efficiently through lightweight models in limited computing environments such as coastal surveillance equipments are presented.

• The Journal of the Korea Contents Association
• /
• v.8 no.7
• /
• pp.66-74
• /
• 2008

Criticality prediction models to identify most fault-prone modules in the system early in the software development process help in allocation of resources and foster software quality improvement. Many models for identifying fault-prone modules using design complexity metrics have been suggested, but most of them are training models that need training data set. Most organizations cannot use these models because very few organizations have their own training data. This paper builds a prediction model based on a well-known supervised learning model, error backpropagation neural net, using design metrics quantifying SDL system specifications. To solve the problem of other models, this model is trained by generated virtual training data set. Some simulation studies have been performed to investigate feasibility of this model, and the results show that suggested model can be an alternative for the organizations without real training data to predict their software qualities.

• Computers and Concrete
• /
• v.17 no.4
• /
• pp.489-498
• /
• 2016

In this paper, the flexural strength ($f_{fs}$) and splitting tensile strength ($f_{sts}$) of concrete containing different proportions of fly ash have been modeled by using gene expression programming (GEP). Two GEP models called GEP-I and GEP-II are constituted to predict the $f_{fs}$ and $f_{sts}$ values, respectively. In these models, the age of specimen, cement, water, sand, aggregate, superplasticizer and fly ash are used as independent input parameters. GEP-I model is constructed by 292 experimental data and trisected into 170, 86 and 36 data for training, testing and validating sets, respectively. Similarly, GEP-II model is constructed by 278 experimental data and trisected into 142, 70 and 66 data for training, testing and validating sets, respectively. The experimental data used in the validating set of these models are independent from the training and testing sets. The results of the statistical parameters obtained from the models indicate that the proposed empirical models have good prediction and generalization capability.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1E
• /
• pp.64-70
• /
• 1994

Predictive neural network models are powerful speech recognition models based on a nonlinear pattern prediction. But those models suffer from poor discrimination between acoustically similar words. In this paper we propose an discriminative training algorithm for predictive neural network models. This algorithm is derived from GPD (Generalized Probabilistic Descent) algorithm coupled with MCEF(Minimum Classification Error Formulation). It allows direct minimization of a recognition error rate. Evaluation of our training algoritym on ten Korean digits shows its effectiveness by 30% reduction of recognition error.

• Bulletin of the Korean Chemical Society
• /
• v.32 no.7
• /
• pp.2397-2404
• /
• 2011

Variable selection k nearest neighbor QSAR modeling approach was applied to a data set of 80 3-arylisoquinolines exhibiting cytotoxicity against human lung tumor cell line (A-549). All compounds were characterized with molecular topology descriptors calculated with the MolconnZ program. Seven compounds were randomly selected from the original dataset and used as an external validation set. The remaining subset of 73 compounds was divided into multiple training (56 to 61 compounds) and test (17 to 12 compounds) sets using a chemical diversity sampling method developed in this group. Highly predictive models characterized by the leave-one out cross-validated $R^2$ ($q^2$) values greater than 0.8 for the training sets and $R^2$ values greater than 0.7 for the test sets have been obtained. The robustness of models was confirmed by the Y-randomization test: all models built using training sets with randomly shuffled activities were characterized by low $q^2{\leq}0.26$ and $R^2{\leq}0.22$ for training and test sets, respectively. Twelve best models (with the highest values of both $q^2$ and $R^2$) predicted the activities of the external validation set of seven compounds with $R^2$ ranging from 0.71 to 0.93.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1996.10a
• /
• pp.29-48
• /
• 1996

Previous studies on user training have largely focused on assessing models which describe the determinants of information technology usage or examined theeffects of training, on user satisfaction, productivity, performance and so on. Scant research efforts have been made, however, to examine those effects of training by using theoretical models. This study presented a conceptural model to predict intention to use information technology and conducted an experimentto understand how training for computer skill acquisition affects primary variables of the model. The data were obtained from 32 student subjects of an experimental group and 31 students of a control group, and the information technology employed for this study was a university's electronic mail system. The study results revealed that attitude toward usage and perceived behavioral control helped to predict user intentions ; outcome expectations were positively related to attitude toward usage ; and self-efficacy was positively related to perceived behavioral control. Thd hands-on training for the experimental group led to increases in perceived ease of use, self-efficacy and perceived behaviroal control. The changes in those variables suggest more causal effects of user training than other survey studies.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.11
• /
• pp.51-59
• /
• 2019

Due to the exponential growth of access information on the web, the need for predicting web users' next access has increased. Various models such as markov models, deep neural networks, support vector machines, and fuzzy inference models were proposed to handle web access prediction. For deep learning based on neural network models, training time on large-scale web usage data is very huge. To address this problem, deep neural network models are trained on cluster of computers in parallel. In this paper, we investigated impact of several important spark parameters related to data partitions, shuffling, compression, and locality (basic spark parameters) for training Multi-Layer Perceptron model on Spark standalone cluster. Then based on the investigation, we tuned basic spark parameters for training Multi-Layer Perceptron model and used it for tuning Spark when training Multi-Layer Perceptron model for web access prediction. Through experiments, we showed the accuracy of web access prediction based on our proposed web access prediction model. In addition, we also showed performance improvement in training time based on our spark basic parameters tuning for training Multi-Layer Perceptron model over default spark parameters configuration.

• Archives of Plastic Surgery
• /
• v.41 no.3
• /
• pp.218-224
• /
• 2014

Plastic surgery training worldwide has seen a thorough restructuring over the past decade, with the introduction of formal training curricula and work-based assessment tools. Part of this process has been the introduction of revalidation and a greater use of simulation in training delivery. Simulation is an increasingly important tool for educators because it provides a way to reduce risks to both trainees and patients, whilst facilitating improved technical proficiency. Current microsurgery training interventions are often predicated on theories of skill acquisition and development that follow a 'practice makes perfect' model. Given the changing landscape of surgical training and advances in educational theories related to skill development, research is needed to assess the potential benefits of alternative models, particularly cross-training, a model now widely used in non-medical areas with significant benefits. Furthermore, with the proliferation of microsurgery training interventions and therefore diversity in length, cost, content and models used, appropriate standardisation will be an important factor to ensure that courses deliver consistent and effective training that achieves appropriate levels of competency. Key research requirements should be gathered and used in directing further research in these areas to achieve on-going improvement of microsurgery training.

• The Journal of Korea Robotics Society
• /
• v.14 no.2
• /
• pp.114-121
• /
• 2019

This paper presents a new benchmark system for visual odometry (VO) and monocular depth estimation (MDE). As deep learning has become a key technology in computer vision, many researchers are trying to apply deep learning to VO and MDE. Just a couple of years ago, they were independently studied in a supervised way, but now they are coupled and trained together in an unsupervised way. However, before designing fancy models and losses, we have to customize datasets to use them for training and testing. After training, the model has to be compared with the existing models, which is also a huge burden. The benchmark provides input dataset ready-to-use for VO and MDE research in 'tfrecords' format and output dataset that includes model checkpoints and inference results of the existing models. It also provides various tools for data formatting, training, and evaluation. In the experiments, the exsiting models were evaluated to verify their performances presented in the corresponding papers and we found that the evaluation result is inferior to the presented performances.

• Journal of Korean Neurosurgical Society
• /
• v.65 no.4
• /
• pp.489-498
• /
• 2022

Spatial insight into intracranial pathology and structure is important for neurosurgeons to perform safe and successful surgeries. Three-dimensional (3D) printing technology in the medical field has made it possible to produce intuitive models that can help with spatial perception. Recent advances in 3D-printed disease models have removed barriers to entering the clinical field and medical market, such as precision and texture reality, speed of production, and cost. The 3D-printed disease model is now ready to be actively applied to daily clinical practice in neurosurgical planning, simulation, and training. In this review, the development of 3D-printed neurosurgical disease models and their application are summarized and discussed.