• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.8
• /
• pp.321-328
• /
• 2014

In this paper, we present a MapReduce-based mobility pattern mining system which can predict efficiently the next place of mobile users. It learns the mobility pattern model of each user, represented by Hidden Markov Models(HMM), from a large-scale trajectory dataset, and then predicts the next place for the user to visit by applying the learned models to the current trajectory. Our system consists of two parts: the back-end part, in which the mobility pattern models are learned for individual users, and the front-end part, where the next place for a certain user to visit is predicted based on the mobility pattern models. While the back-end part comprises of three distinct MapReduce modules for POI extraction, trajectory transformation, and mobility pattern model learning, the front-end part has two different modules for candidate route generation and next place prediction. Map and reduce functions of each module in our system were designed to utilize the underlying Hadoop infrastructure enough to maximize the parallel processing. We performed experiments to evaluate the performance of the proposed system by using a large-scale open benchmark dataset, GeoLife, and then could make sure of high performance of our system as results of the experiments.

• Journal of Broadcast Engineering
• /
• v.25 no.5
• /
• pp.750-757
• /
• 2020

One of the basic tasks for robots to interact with humans is to quickly and accurately grasp human behavior. Therefore, it is necessary to increase the accuracy of human pose recognition when the robot is estimating the human pose and to recognize it as quickly as possible. However, when the human pose is estimated using deep learning, which is a representative method of artificial intelligence technology, recognition accuracy and speed are not satisfied at the same time. Therefore, it is common to select one of a top-down method that has high inference accuracy or a bottom-up method that has high processing speed. In this paper, we propose two methods that complement the disadvantages while including both the advantages of the two methods mentioned above. The first is to perform parallel inference on the server using multi GPU, and the second is to mix bottom-up and One-class Classification. As a result of the experiment, both of the methods presented in this paper showed improvement in speed. If these two methods are applied to the entertainment robot, it is expected that a highly reliable interaction with the audience can be performed.

• Journal of Practical Engineering Education
• /
• v.12 no.1
• /
• pp.117-125
• /
• 2020

According to the government's smart factory promotion project for small and medium-sized enterprises, more than 10,000 intelligent factories are scheduled or already built in the country and the government-led goal is to nurture 100,000 skilled workers by 2022. Smart Factory introduces numerous types of education and training courses from the supplier's point of view, such as training institutions belonging to local governments, some universities, and public organizations, in the form of an efficient resource management system and ICT technology convergence in the automated manufacturing equipment. The lack of linkage with the NCS, the standard for training, seems to have room for rethinking and direction. Results of survey is provided for the family companies of K-University in the metropolitan area and Chungnam area, and analyzes job demands by identifying whether or not they want to introduce smart factories. Defining the practitioners who will serve as a window for the introduction of smart factory technology within the company, setting up a training goal in consideration of the career path, and including the level of training required competency units, optional competency units, and training time suitable for introducing and operating smart factories. Author would like to present an NCS-based qualification design plan.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.33-42
• /
• 2022

In this paper, we proposed a deep learning model based on multi-scale dilated convolution feature fusion for the segmentation of black ice region in road image to send black ice warning to drivers in real time. In the proposed multi-scale dilated convolution feature fusion network, different dilated ratio convolutions are connected in parallel in the encoder blocks, and different dilated ratios are used in different resolution feature maps, and multi-layer feature information are fused together. The multi-scale dilated convolution feature fusion improves the performance by diversifying and expending the receptive field of the network and by preserving detailed space information and enhancing the effectiveness of diated convolutions. The performance of the proposed network model was gradually improved with the increase of the number of dilated convolution branch. The mIoU value of the proposed method is 96.46%, which was higher than the existing networks such as U-Net, FCN, PSPNet, ENet, LinkNet. The parameter was 1,858K, which was 6 times smaller than the existing LinkNet model. From the experimental results of Jetson Nano, the FPS of the proposed method was 3.63, which can realize segmentation of black ice field in real time.

• Journal of The Korean Association For Science Education
• /
• v.27 no.4
• /
• pp.318-327
• /
• 2007

The purpose of this study is to examine the effects of small-scale chemistry(SSC) laboratory activities implemented in high school chemistry II classes on the students' inquiry process skills and science-related attitudes. For this study, 112 students in the 12th grade were chosen and divided into an experimental and a control group. Seven SSC lab programs that can replace the traditional experiments in chemistry II textbooks were selected and administered to the experimental group while the traditional textbook experiments were administered to the control group. The results showed that there was a significant difference in the enhancement of inquiry process skills between the two groups while no significant difference was found in science-related attitudes. Further analysis showed that the difference in the inquiry process skills came from the basic inquiry process skills. The experimental group students thought that the SSC experiments have many advantages compared to the traditional experiments, e.g., individual work, learning lab and theory in parallel, short experiment time, safety, environmental aspects, etc. These results suggest that the SSC lab programs are valuable in high school chemistry classes and developing and distributing various SSC lab programs is needed to replace the traditional experiments in the current textbooks.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.3
• /
• pp.157-167
• /
• 2019

This study empirically analyzed the effects of the use of a single human resource development system in the enterprise on corporate performance using the Human Capital Enterprise Panel (HCCP) data. The results of the hierarchical regression analysis on the sales per log of job analysis use, The use of job analysis confirms that $R^2=.294$ and ${\beta}=.165$ can have a positive effect on sales per log, and Hypothesis 1 is supported. The participation in the work parallelism participation was negatively influenced by the sales per log in $R^2=.283$ and ${\beta}=-.129$, and Hypothesis 2 was rejected. This is attributed to the lack of data of 66, and it was judged that there were 45 new companies entering the company. In addition, we conducted a hierarchical regression analysis that confirms the moderating effect of the training and training obligation by using interaction variables of job analysis use and education and training obligation. It was confirmed that the use of job analysis could have a negative impact on the sales per log, and Hypothesis 3 was rejected. As the labor productivity increases, firms have supported the previous study that productivity effect is not significant because they do not want to invest in education and training. In addition, it was confirmed that the participation of the training system in the job training system could strengthen the positive sales (+). Therefore, Hypothesis 4 was supported. In order to reflect the effective aspects of job analysis, the voluntary activation of enterprises should be premised. In addition, if employing talented people with diverse backgrounds such as academic backgrounds, gender, religion, nationality, etc. and investing in human resources development through education and training focused on job analysis, recruitment of learning workers in parallel with work- It will be possible to contribute to the creation of performance.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.135-149
• /
• 2020

Artificial intelligences are changing world. Financial market is also not an exception. Robo-Advisor is actively being developed, making up the weakness of traditional asset allocation methods and replacing the parts that are difficult for the traditional methods. It makes automated investment decisions with artificial intelligence algorithms and is used with various asset allocation models such as mean-variance model, Black-Litterman model and risk parity model. Risk parity model is a typical risk-based asset allocation model which is focused on the volatility of assets. It avoids investment risk structurally. So it has stability in the management of large size fund and it has been widely used in financial field. XGBoost model is a parallel tree-boosting method. It is an optimized gradient boosting model designed to be highly efficient and flexible. It not only makes billions of examples in limited memory environments but is also very fast to learn compared to traditional boosting methods. It is frequently used in various fields of data analysis and has a lot of advantages. So in this study, we propose a new asset allocation model that combines risk parity model and XGBoost machine learning model. This model uses XGBoost to predict the risk of assets and applies the predictive risk to the process of covariance estimation. There are estimated errors between the estimation period and the actual investment period because the optimized asset allocation model estimates the proportion of investments based on historical data. these estimated errors adversely affect the optimized portfolio performance. This study aims to improve the stability and portfolio performance of the model by predicting the volatility of the next investment period and reducing estimated errors of optimized asset allocation model. As a result, it narrows the gap between theory and practice and proposes a more advanced asset allocation model. In this study, we used the Korean stock market price data for a total of 17 years from 2003 to 2019 for the empirical test of the suggested model. The data sets are specifically composed of energy, finance, IT, industrial, material, telecommunication, utility, consumer, health care and staple sectors. We accumulated the value of prediction using moving-window method by 1,000 in-sample and 20 out-of-sample, so we produced a total of 154 rebalancing back-testing results. We analyzed portfolio performance in terms of cumulative rate of return and got a lot of sample data because of long period results. Comparing with traditional risk parity model, this experiment recorded improvements in both cumulative yield and reduction of estimated errors. The total cumulative return is 45.748%, about 5% higher than that of risk parity model and also the estimated errors are reduced in 9 out of 10 industry sectors. The reduction of estimated errors increases stability of the model and makes it easy to apply in practical investment. The results of the experiment showed improvement of portfolio performance by reducing the estimated errors of the optimized asset allocation model. Many financial models and asset allocation models are limited in practical investment because of the most fundamental question of whether the past characteristics of assets will continue into the future in the changing financial market. However, this study not only takes advantage of traditional asset allocation models, but also supplements the limitations of traditional methods and increases stability by predicting the risks of assets with the latest algorithm. There are various studies on parametric estimation methods to reduce the estimated errors in the portfolio optimization. We also suggested a new method to reduce estimated errors in optimized asset allocation model using machine learning. So this study is meaningful in that it proposes an advanced artificial intelligence asset allocation model for the fast-developing financial markets.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.10
• /
• pp.2028-2042
• /
• 1994

Properly defining signal and noise in a self-organizing system like ART(Adaptive Resonance Theory) neural network model raises a number of subtle issues. Pattern context must enter the definition so that input features, treated as irrelevant noise when they are embedded in a given input pattern, may be treated as informative signals when they are embedded in a different input pattern. The ATR automatically self-scales their computational units to embody context and learning dependent definitions of a signal and noise and there is no problem in categorizing input pattern that have features similar in nature. However, when we have imput patterns that have features that are different in size and nature, the use of only one vigilance parameter is not enough to differentiate a signal from noise for a good categorization. For example, if the value fo vigilance parameter is large, then noise may be processed as an informative signal and unnecessary categories are generated: and if the value of vigilance parameter is small, an informative signal may be ignored and treated as noise. Hence it is no easy to achieve a good pattern categorization. To overcome such problems, a Coupled-ART neural network capable of modularized categorization of patterns is proposed. The Coupled-ART has two layer of tightly coupled modules. the upper and the lower. The lower layer processes the global features of a pattern and the structural features, separately in parallel. The upper layer combines the categorized outputs from the lower layer and categorizes the combined output, Hence, due to the modularized categorization of patterns, the Coupled-ART classifies patterns more efficiently than the ART1 model.