• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1045-1055
• /
• 2007

Latent topic models are statistical models which automatically captures salient patterns or correlation among features underlying a data collection in a probabilistic way. They are gaining an increased popularity as an effective tool in the application of automatic semantic feature extraction from text corpus, multimedia data analysis including image data, and bioinformatics. Among the important issues for the effectiveness in the application of latent topic models to the massive data set is the efficient learning of the model. The paper proposes an accelerated learning technique for PLSA model, one of the popular latent topic models, by an incremental EM algorithm instead of conventional EM algorithm. The incremental EM algorithm can be characterized by the employment of a series of partial E-steps that are performed on the corresponding subsets of the entire data collection, unlike in the conventional EM algorithm where one batch E-step is done for the whole data set. By the replacement of a single batch E-M step with a series of partial E-steps and M-steps, the inference result for the previous data subset can be directly reflected to the next inference process, which can enhance the learning speed for the entire data set. The algorithm is advantageous also in that it is guaranteed to converge to a local maximum solution and can be easily implemented just with slight modification of the existing algorithm based on the conventional EM. We present the basic application of the incremental EM algorithm to the learning of PLSA and empirically evaluate the acceleration performance with several possible data partitioning methods for the practical application. The experimental results on a real-world news data set show that the proposed approach can accomplish a meaningful enhancement of the convergence rate in the learning of latent topic model. Additionally, we present an interesting result which supports a possible synergistic effect of the combination of incremental EM algorithm with parallel computing.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.10
• /
• pp.330-338
• /
• 2018

The proportional-integral-derivative(PID) controller has been widely used in the industry because of its robust performance and simple structure in a wide range of operating conditions. However, the AVR(Automatic Voltage Regulator) as a control system is not robust to variations of the power system parameters. Therefore, it is necessary to use PID controller to increase the stability and performance of the AVR system. In this paper, a novel design method for determining the optimal PID controller parameters of an AVR system using the particle swarm optimization(PSO) algorithm is presented. The proposed approach has superior features, including easy implementation, stable convergence characteristic and good computational efficiency. In order to assist estimating the performance of the proposed PSO-PID controller, a new performance criterion function is also defined. This evaluation function is intended to reflect when the maximum percentage overshoot, the settling time are given as design specifications. The ITAE evaluation function should impose a penalty if the design specifications are violated, so that the PSO algorithm satisfies the specifications when searching for the PID controller parameter. Finally, through the computer simulations, the proposed PSO-PID controller not only satisfies the given design specifications for the terminal voltage step response, but also shows better control performance than other similar recent studies.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.12
• /
• pp.6143-6150
• /
• 2012

The current society, namely information society is now moving to a specific topic which is SMART. In this sense, recently a variety of social areas including communications and SOC domains are moving fast to this topic. In Korea, The electric power area has been doing a pioneering job relatively successfully and the water supply area is just now taking the first step. The purpose of this research is to develop a technical Framework for Smart Water Grid Service. Related researches has been studied and the 4 constituting technical element areas were defined first. For each of the four areas, a framework modeling was fulfilled and as a result, a TRM(Technical Road Map) was suggested for each of the area. Finally, an Enterprise TRM covering all of the 4 areas was described. Furthermore, the currently suggested Framework model was compared to a related model and it was found that the integration of the models is desirable to wholly cover from Macro to Micro level applications and services. It is expected that the current approach contribute, more or less, to the smart implementation in the areas of water management.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.125-140
• /
• 2018

As one of the processes in the manufacturing industry, quality inspection inspects the intermediate products or final products to separate the good-quality goods that meet the quality management standard and the defective goods that do not. The manual inspection of quality in a mass production system may result in low consistency and efficiency. Therefore, the quality inspection of mass-produced products involves automatic checking and classifying by the machines in many processes. Although there are many preceding studies on improving or optimizing the process using the data generated in the production process, there have been many constraints with regard to actual implementation due to the technical limitations of processing a large volume of data in real time. The recent research studies on big data have improved the data processing technology and enabled collecting, processing, and analyzing process data in real time. This paper aims to propose the process and details of applying big data for quality inspection and examine the applicability of the proposed method to the dairy industry. We review the previous studies and propose a big data analysis procedure that is applicable to the manufacturing sector. To assess the feasibility of the proposed method, we applied two methods to one of the quality inspection processes in the dairy industry: convolutional neural network and random forest. We collected, processed, and analyzed the images of caps and straws in real time, and then determined whether the products were defective or not. The result confirmed that there was a drastic increase in classification accuracy compared to the quality inspection performed in the past.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.309-313
• /
• 2018

Maker Activity based Design-Thinking 5 Step is a better experience for providing the public with a better approach to design by sympathizing with human needs and exploring the public's potential desires. This paper studied the teaching method by applying the value of the Maker Training that emphasizes the importance of sharing and opening results with others after creating them based on the ideas of the learners using different tools, the basic mind of Design-Thinking with the observation and empathy about human beings and the derivation of user-centric results to the SW coding curriculum. For this purpose, we studied teaching methods by Design-thinking based on sections that require coding in the information classes of the secondary school curriculum. We demonstrated the class to a group of education experts by applying the syllabus of instruction. Then we collected evaluations and opinions of the experts and we modified the syllabus to reflect the collected assessment. Under the Design-thinking 5 step foundation, classes were demonstrated with a group of education experts, opinions collected, and suggestions collected were reflected on the assessment outcome. I think that It is an effective training opportunity to practice a widespread idea that enables infinite thinking and a convergence of thinking that brings out the best decision from a widespread idea.

• Korean Journal of Optics and Photonics
• /
• v.23 no.4
• /
• pp.147-153
• /
• 2012

Multi-view 3D displays based on a limited number of pixels have the problem that the stereo-scopic image has a low resolution because of increasing view number. To solve the problem of low resolution, we propose an optical approaching method that focuses the width of a unit pixel by using a commercial spherical shape lenticular lens sheet and increases the effective resolution by increasing the number of sources of light in the multi-view 3D display system based on projection type. The method was performed in such an order that several main derivable parameters were defined, and, through the theoretical and experimental result, the value of the contractible unit pixel width and the scalable effective resolution was derived in a given system environment. As a result, for the case that the ray of light from the projector transmitted the 25 LPI lenticular lens sheet which has the pitch size 1.016 mm, the focused unit pixel width was 0.19 mm and the scalable effective resolution was, at most, 5 times wider than the original one. In addition, the range of depth of focus was 1.496 mm, which shows us the range of thickness tolerances of commercial spherical shape lenticular lens sheet and sufficient optical alignment tolerances.

• Journal of Digital Convergence
• /
• v.12 no.7
• /
• pp.327-336
• /
• 2014

This study aims to identify factors to affect regular utilization status of medical care in cardio-cerebrovascular patients. The research selected 770 cardio-cerebrovascular patients among surveyees from the Korea Health Panel 2010. We analyzed states of medical care utilization using descriptive statistics. Logistic regression analysis was used to examine the main factors associated with regular utilization status of medical care in cardio-cerebrovascular patients. In result, the significant factors associated with regular utilization status of medical care in cardio-cerebrovascular patients were age, education level, household income level. CCI, presence or absence of high risk drinking, and presence or absence of obesity. There's a high probability that patients aged between 60 and 69, equal to and higher than those of high school graduate in education level, upper middle class in household income, the higher CCI, absence of high risk drinking, presence of obesity utilize medical care services more regularly. Therefore, it is necessary to develop effective program and individualized approach for patients using lesser periodical medical care and patients with high risk drinking problem. In the future, these findings can be used an important data for healthcare policy and assessment.

• Journal of Digital Convergence
• /
• v.11 no.4
• /
• pp.251-258
• /
• 2013

Recently companies are confronted with a much more sophisticated business environment than before and at the same time have to be able to adapt to rapid changes. Accordingly, the need for selecting among alternatives and managing systematic decision-making has been steadily increasing to respond to a more diverse customer needs and keep up with the fierce competition. In this study, we propose a framework that consist of problem solving procedures and techniques and knowledge structure built on processes to support strategic decision making. and discuss how to utilize simulation tools as the knowledge-based problem solving tools. In addition we discuss how to build and advance the knowledge structure to implement the proposed architecture. Management decision support systems architecture consist of three key factors. The first is Problem Solving Approach which is used as reference. The second is knowledge structure on business processes that includes standard and reference business processes. The third is simulators that are able to generate and analyze alternatives using problem solving techniques and knowledge base. In sum, the proposed framework of decision-making support systems facilitates knowledge-intensive consulting processes to promote the development and application of consulting knowledge and techniques and increase the efficiency of consulting firms and industry.

• Journal of Digital Convergence
• /
• v.15 no.10
• /
• pp.327-337
• /
• 2017

The purpose of this study is to investigate patient satisfaction for the spacial environment of the rehabilitative facilities, and by this to examine areas of patients' needs to develop patient-friendly improvement in spacial environment. 308 patients were surveyed at physical therapy centers of 6 hospitals or rehabilitative institutions in Daejeon with patient satisfaction questionnaire. The spatial environment satisfaction was categorized into 7 constructs and the data were statistically analyzed and compared by t-test and ANOVA according to the patient characteristics. As a result, inter-floor transfer was found to be the area with the least satisfaction, especially in the in-patient group. Locating floor level was the second lowest satisfied construct. Comfort was relatively highly satisfied construct overall, but patients with neurological disease had less satisfaction than with musculoskeletal disease. Patients in 20's were less satisfied in wayfinding than 60's. Satisfaction for interior design was lower in subject in 50's than 70's. These findings can be used as evidences to improve spacial environment design for rehabilitative facilities to the better patient satisfaction.

• Korean Journal of Environment and Ecology
• /
• v.31 no.3
• /
• pp.329-336
• /
• 2017

We analyzed the influence of living characteristics as environment related behavior by questionnaire survey analysis. Survey was conducted to 569 people living in Gimhae city, Gyongsangnam-do, Korea. By the results of correlation analysis, items with high levels of practice were related to legal regulations. However, group who are less interested in preserving the environment have lower performance in all aspects of their activities. In the relationship between social characteristics and environment related behavior, residents were generally aware of the practice of detached house dweller group rather than of apartment dweller. As the age increased, have more willing to practice. While the higher levels of education group do not make environment related behave better. However, it seemed that the student group's distinct characteristics was reflected in the education level. As this study, improvement of environmental awareness is needed to improve practice of environment related behavior rather than regal approach. Policy for environment related behavior is necessary to vary the methodology based on analysis of social characteristics, such as gender, residential type and education level rather than generalization.