• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.397-398
• /
• 2022

본 논문에서는 SK텔레콤에서 진행하는 디지털 선도기업 아카데미 FLYAI의 교육과정을 설계하고 개발한다. 이 교육과정은 Project Based Learning(272시간)과 Product Based Learning(128시간)으로 구성하여 총 400시간을 교육하도록 설계한다. 특히 Product Based Learning의 AI-Hackathon(80시간)에서는 SK텔레콤 각 부서에서 제안하는 제픔을 기획하고 개발하는 과정으로 SK텔레콤 AI 개발자들이 멘토로 참여함으로써 기업 현장의 경험을 체험할 수 있도록 개발한다.

• Journal of Multimedia Information System
• /
• v.8 no.4
• /
• pp.243-250
• /
• 2021

Instructional contents are used to demonstrate a technical process to teach and walkthrough certain procedures to carry out a task. This type of informational content is widely used for teaching and lectures in form of tutorial videos and training videos. Since there are questions and uncertainties for what could be the killer application for the novel wearables, we propose a self-instruction training application on a smart glass to utilize already-available instruction videos as well as public open data in creative ways. We design and implement a prototype application to help users train by wearing smart glasses specifically designed for two concrete and hand-constrained use cases where the user's hands need to be free to operate. To increase the efficiency and feasibility of the self-instruction training, we contribute to the development of a wearable killer application by integrating a voice-based user interface using speech recognizer, public open data APIs, and timestamp-based procedural content navigation structure into our proof-of-concept application.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.3
• /
• pp.251-258
• /
• 2022

In this paper, we propose a standard AI developer job training track based on industry needs. The characteristic of this curriculum is that it can minimize the mismatch of AI developer job competency between industries and universities. To develop an AI developer job training track, a survey will be conducted for AI developers working in industrial fields. In this survey, among the five NCS-based AI developer jobs, job analysis is conducted by deriving AI developer jobs with high demand for manpower in industrial fields. In job analysis, the core competency unit elements of the job are selected, and knowledge, skills, tools, etc. necessary to perform the core competency unit elements are derived. In addition, a standard AI developer job curriculum is developed by deriving core subjects and road-map that can educate knowledge, skills, tools, etc. In addition, we present an efficient AI developer job training method using the standard AI developer job training course proposed in this paper.

• International Journal of Computer Science & Network Security
• /
• v.21 no.3
• /
• pp.21-30
• /
• 2021

Economic transformations have led to an increase in the role of creative assets and their central role in public life. Changes in creative activity have led to a change in the organization of the work of institutes engaged in the training of specialists, in particular teachers of labor education. Methods and approaches to training determine the development of creative industries, being the basis for models of professional training of future teachers of labor training. The purpose of an article was to develop a modern model of professional training of future teachers of labor training based on the concept of creative economy. The methodology is based on the concepts of holistic craft and creative economy. Based on the integration of pedagogical learning models "Craft as design and problem-solving", "Craft as skill and knowledge building", "Craft as product-making" and "Craft as self-expression" developed and experimentally confirmed the conceptual model of professional training of future teachers of labor training. The proposed model forms a practitioner with professional, technical, digital and creative skills who is able to transfer the experience to students. The training course "Creativity and creative thinking" has been developed. The model provided for the development of a course based on the strategy of developing professional creativity, flexibility, improvisation, openness, student activity, joint practice, student-oriented approach. The practical value implies the adaptation of the developed model of professional training of future teachers of labor education during the training of teachers in higher education, which is confirmed in the experiment.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.327-338
• /
• 2022

Classification models for human action recognition require robust features and large training sets for good generalization. However, data augmentation methods are employed for imbalanced training sets to achieve higher accuracy. These samples generated using data augmentation only reflect existing samples within the training set, their feature representations are less diverse and hence, contribute to less precise classification. This paper presents new data augmentation and action representation approaches to grow training sets. The proposed approach is based on two fundamental concepts: virtual video generation for augmentation and representation of the action videos through robust features. Virtual videos are generated from the motion history templates of action videos, which are convolved using a convolutional neural network, to generate deep features. Furthermore, by observing an objective function of the genetic algorithm, the spatiotemporal features of different samples are combined, to generate the representations of the virtual videos and then classified through an extreme learning machine classifier on MuHAVi-Uncut, iXMAS, and IAVID-1 datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.4028-4042
• /
• 2021

Aiming at the problem of software defect prediction difficulty caused by insufficient software defect marker samples and unbalanced classification, a semi-supervised software defect prediction model based on a tri-training algorithm was proposed by combining feature normalization, over-sampling technology, and a Tri-training algorithm. First, the feature normalization method is used to smooth the feature data to eliminate the influence of too large or too small feature values on the model's classification performance. Secondly, the oversampling method is used to expand and sample the data, which solves the unbalanced classification of labelled samples. Finally, the Tri-training algorithm performs machine learning on the training samples and establishes a defect prediction model. The novelty of this model is that it can effectively combine feature normalization, oversampling techniques, and the Tri-training algorithm to solve both the under-labelled sample and class imbalance problems. Simulation experiments using the NASA software defect prediction dataset show that the proposed method outperforms four existing supervised and semi-supervised learning in terms of Precision, Recall, and F-Measure values.

• International journal of advanced smart convergence
• /
• v.12 no.2
• /
• pp.159-166
• /
• 2023

Due to the aging workforce in the construction industry in South Korea, the accident rate has been increasing. The cognitive abilities of older workers are closely related to both safety incidents and labor productivity. Therefore, there is a need to improve cognitive abilities through personalized training based on cognitive assessment results, using cognitive training content, in order to enable safe performance in labor-intensive environments. The provided cognitive training content includes concentration, memory, oreintation, attention, and executive functions. Difficulty levels were applied to each content to enhance user engagement and interest. To stimulate interest and encourage active participation of the participants, the difficulty level was automatically adjusted based on feedback from the MMSE-DS results and content measurement data. Based on the accumulated data, individual training scenarios have been set differently to intensively improve insufficient cognitive skills, and cognitive training programs will be developed to reduce safety accidents at construction sites through measured data and research. Through such simple cognitive training, it is expected that the reduction of accidents in the aging construction workforce can lead to a decrease in the social costs associated with prolonged construction periods caused by accidents.

• International Journal of Contents
• /
• v.14 no.4
• /
• pp.24-29
• /
• 2018

EEG-based brain-computer interfaces has focused on explicitly expressed intentions to assist physically impaired patients. For EEG-based-computer interfaces to function effectively, it should be able to understand users' implicit information. Since it is hard to gather EEG signals of human brains, we do not have enough training data which are essential for proper classification performance of implicit intention. In this paper, we improve the subject independent classification of implicit intention through the generation of additional training data. In the first stage, we perform the PCA (principal component analysis) of training data in a bid to remove redundant components in the components within the input data. After the dimension reduction by PCA, we train ICA (independent component analysis) network whose outputs are statistically independent. We can get additional training data by adding Gaussian noises to ICA outputs and projecting them to input data domain. Through simulations with EEG data provided by CNSL, KAIST, we improve the classification performance from 65.05% to 66.69% with Gamma components. The proposed sample generation method can be applied to any machine learning problem with fewer samples.

• 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.

• Journal of the Korean Society of Systems Engineering
• /
• v.4 no.2
• /
• pp.27-33
• /
• 2008

As a part of development of Korean K2 main battle tank, embedded training computer to be operated in the main equipment, which makes it possible to train without a special-purposed training simulator, was adopted for tank combat training. The category of embedded training of Korean K2 main battle tank includes driving training, gunnery training, single tank combat training, platoon level combat training, and command and platoon leaders combat training. For realization unit level tank embedded training system, the virtual reality was utilized for real time image rendering, and network based real time communication system of K2 tank was utilized for sharing status information between tanks. As a result, it is possible to train themselves on their own tank for enhancing the operational skills and harmonized task with members.