Earticle

In this paper, the Self-Organizing Feature Maps (SOFM) neural network is applied to analyse the multi dimensional process data, and to diagnose the inter-relationship of the process variables in a real municipal sewage treatment plant. The data set had been collected from a sewer system in the Gwangyang city, Korea. The data had been measured in the period of 1st January, 2004 and 31th December, 2006. The data set contains daily averaged values for each of the twenty three variables. Through the component planes visualization, it is evident that the effluent is related to rainfall, flow rate, temperature, MLSS, SRT, RAS and DO. Especially, rainfall, flow rate and temperature are the most important driving force to increase in effluent levels in the Gwangyang municipal sewage treatment plant. It is concluded that the SOFM technique provides an effective analyzing and diagnosing tool to understand the system behavior and to extract knowledge contained in multi-dimensional data of a large-scale sewage treatment plant.

본 연구에서는 인공지능 프로그램인 SOM (Self Organizing Maps)를 이용하여 하천의 유량과 수질의 관계를 분석하였으며, 국내 K 연안으로 유입되는 T하천에 대해 과거 4년 자료의 유량 및 수질에 관해 조사를 하였다. 조사기간 중 K 연안으로 유입되는 T하천의 평균 유량 834,850 m3/d, 수온 16.6oC, pH 7.8, DO 10.5mg/l, BOD 1.8mg/l, COD 4.6mg/l, SS 9.0mg/l, T-N 1.356mg/l, T-P 0.071 mg/l를 나타내었다. 시계열에 따른 연도별로 유량 변화에 따라 SS, T-N, T-P 농도가 영향을 많이 받는 것으로 나타났으나 유량 변화와 하천의 전체 수질 항목간의 연관성을 일괄적으로 파악하는 것은 어려웠다. 패턴분석 결과 T하천의 경우 BOD와 COD 및 T-N과 T-P 가 서로 유사한 패턴을 보였으나, 유량과는 일관된 패턴 형태를 찾기가 어려웠다. 그러나, 하천 유량과 SS, pH, DO는 밀접한 관계를 가지고 있으며, SS와는 비슷한 패턴을 pH 및 DO 와는 상반된 패턴을 보이는 것으로 나타났다. T하천의 경우 유량이 높고, DO가 낮을 때 COD, SS, T-N, T-P 및 수온이 높은 패턴을 보인 바, 강우 시 및 여름철에 수질이 악화되는 것을 시사하였다.

The pattern analysis of correlation between the river flow and water quality using an artificial intelligence program, SOM (Self Organizing Maps) technique were analysed in this study. The past four years of data flow and water quality of the T river flowing into the domestic K coast had been collected. Average flow of the T river flowing into the K coast was 834,850 m3/d, average water temperature was 16.6oC, average pH of 7.8, DO of the 10.5 mg/l, BOD of the 1.8 mg/l, COD of the 4.6 mg/l, SS of the 9.0 mg/l, T-N of the 1.356 mg/l, T-P was 0.071 mg/l. Changes in flow rate due to the time series by year appeared to be frequently affected to the SS, T-N, and T-P concentrations, but inter-relationship between overall items of water quality and river flow changes was difficult to identify. SOFMs (Self Organizing Feature Maps) of BOD and COD in T river, and SOFMs of T-N and T-P showed a similar pattern to each other, respectively, but a relationship between flow pattern and them was difficult to find. However, river flow and SS, pH, DO has a close relationship, river flow and SS was analysed to have showed a positive correlation, but the negative correlation between river flow and pH, DO. SOFMs pattern of river flow and DO, and them of COD, SS, TN, TP, and temperature suggested that water quality in T river is deteriorating during rainfall period and summer season.

본 논문은 이미지에서 효과적인 문자검출을 위해 색상단순화 및 윤곽선에서의 패턴 분석을 통한 문자 검출방법을 제안한다. 윤곽선 기반방법을 사용하는 문자검출 알고리즘은 단순한 배경의 이미지에서는 우수한 성능을 보이지만, 복잡한 배경의 이미지에서는 성능이 떨어지는 단점이 있다. 따라서 제안하는 방법은 복잡한 배경에서의 비문자영역을 최소화하기 위해 이미지 단순화 및 패턴분석을 통한 문자 검출 알고리즘을 제안한다. 먼저 이미지에서의 문자영역 부분을 검출하기 위하여 전처리 과정으로 K-means 군집화를 사용하여 이미지의 색상을 단순화하고, 색상 단순화 과정에서의 물체의 경계의 흐릿해짐을 개선하기 위해 고주파통과필터를 통해 물체의 경계를 강화한다. 그 후 모폴로지 기법의 팽창과 침식의 차이를 이용하여 물체의 윤곽선을 검출하고, 획득한 영역의 윤곽선 부분의 정보(높이, 너비 면적)를 구한 후 패턴분석을 통해 조건을 줌으로써 문자 후보영역을 판별하여 문자가 아닌 불필요한 영역(그림, 배경)을 제거한다. 최종 결과로 라벨링을 통해 불필요한 영역이 제거된 결과를 보여준다.

In this paper, we propose a text extraction method by pattern analysis on contour for effective text detection in image. Text extraction algorithms using edge based methods show good performance in images with simple backgrounds, The images of complex background has a poor performance shortcomings. The proposed method simplifies the color of the image by using K-means clustering in the preprocessing process to detect the character region in the image. Enhance the boundaries of the object through the High pass filter to improve the inaccuracy of the boundary of the object in the color simplification process. Then, by using the difference between the expansion and erosion of the morphology technique, the edges of the object is detected, and the character candidate region is discriminated by analyzing the pattern of the contour portion of the acquired region to remove the unnecessary region (picture, background). As a final result, we have shown that the characters included in the candidate character region are extracted by removing unnecessary regions.

기종점 데이터는 수요 분석 및 서비스 설계를 위해서 대중교통, 도로운영 등 다양한 분야에 서 저장 및 활용되고 있다. 최근 빅데이터의 활용성이 증대되면서 기종점 데이터의 분석 및 활용에 대한 수요도 함께 증가하고 있다. 기존의 일반적인 교통 정보 데이터가 수집장비 수(n) 에 비례하여 데이터양이 증가(a·n)하는 것과는 다르게, 기종점 데이터는 수집지점 수(n)의 증 가에 따라 수집 데이터의 양이 기하급수적으로 증가(a·n2)하는 경향이 있다. 이로 인하여 기종 점 데이터를 원시 데이터의 형태로 장기간 저장하고 빅데이터 분석에 활용하는 것은 대용량의 저장 공간이 필요하다는 것을 고려할 때 실용적 대안으로 여겨지지 않고 있다. 이와 함께 기종 점 데이터는 0~10 사이의 작은 수요 부분에 패턴화된 형태와 무작위 적인 형태의 데이터가 섞여있어 작은 수요가 그룹화되어 발생하는 주요 패턴을 추출하기에 어려움이 있다. 이러한 기종점 데이터의 저장용량의 한계와 패턴화 분석의 한계를 극복하고자 본 연구에서는 주성분 분석을 활용한 대중교통 기종점 데이터의 압축 및 분석 방법을 제안하였다. 본 연구에서는 서 울시와 세종시의 대중교통 이용 데이터를 활용하여 모빌리티 데이터를 분석하고, 모빌리티 기 종점 데이터에 포함된 무작위 성향이 높은 데이터를 제거하기 위해 주성분분석 기반의 데이터 압축 및 복원에 관한 연구를 수행하였다. 주성분분석으로 분해된 기종점 데이터와 원데이터를 비교하여 주요한 수요 패턴을 찾고 이를 통해 압축률과 복원율을 높일 수 있는 주성분 범위를 제안하였다. 본 연구에서 분석한 결과, 서울시 기준 1~80, 세종시 기준 1~60까지의 주성분을 사용할 경우 주요 이동 데이터의 손실 없이 기종점 데이터에 포함되어있는 노이즈를 제거하고 데이터를 압축 및 복원이 가능하였다.

Origin-destination data have been collected and utilized for demand analysis and service design in various fields such as public transportation and traffic operation. As the utilization of big data becomes important, there are increasing needs to store raw origin-destination data for big data analysis. However, it is not practical to store and analyze the raw data for a long period of time since the size of the data increases by the power of the number of the collection points. To overcome this storage limitation and long-period pattern analysis, this study proposes a methodology for compression and origin-destination data analysis with the compressed data. The proposed methodology is applied to public transit data of Sejong and Seoul. We first measure the reconstruction error and the data size for each truncated matrix. Then, to determine a range of principal components for removing random data, we measure the level of the regularity based on covariance coefficients of the demand data reconstructed with each range of principal components. Based on the distribution of the covariance coefficients, we found the range of principal components that covers the regular demand. The ranges are determined as 1~60 and 1~80 for Sejong and Seoul respectively.

최근 3년간 농업 관련 뉴스 및 동향에 따르면 양봉산업의 뿌리인 꿀 채밀량이 감소 추세에 있어 양봉산업 전체에 큰 손해가 있는 상황이다. 이러한 현상은 이상 기후에 대한 대응책 미비, 각종 해충과 질병의 발생 등이 원 인으로 제공되고 있는 것으로 보고되고 있다. 따라서, 본 논문에서는 벌꿀 생산량의 군집의 수, 군집 당 평균 수확 량, 총 생산량, 연도별 생산량의 변화에 따른 미래의 세계 양봉산업의 가능성을 예측하는 연구를 수행하였다. 이를 위해, 미국의 벌꿀 생산량에 대한 데이터의 종합적 통계를 구하고, 그룹별 생산량에 대한 군집 회귀 분석을 수행하 여 꿀벌 군집 소멸과의 상관성 분석을 수행하였다. 분석 결과, 기상 변화 데이터를 활용하여 평년 기온의 상승과 미국 내 벌꿀의 총생산량은 줄어들었음을 확인할 수 있었고 생존 군락이 점점 줄어드는 결론이 도출되었다.

According to agricultural news and trends for the past three years, the amount of honey harvested, which is the root of the beekeeping industry, is on the decline, causing great damage to the entire beekeeping industry. It is reported that this phenomenon is provided as the cause of the lack of countermeasures against abnormal climate and the occurrence of various pests and diseases. Therefore, in this paper, research was conducted to predict the future potential of the world's beekeeping industry according to the number of clusters of honey production, average yield per cluster, total production, and annual production. To this end, comprehensive statistics of honey production data in the United States were obtained, and cluster regression analysis was performed on production by group to analyze correlation with honey bee colony extinction. As a result of the analysis, it was confirmed that the average annual temperature increased and the total production of honey in the United States decreased by using the weather change data, and the conclusion was drawn that the surviving colony gradually decreased.

최근 인공지능, 딥러닝, 빅데이터 등 4차 산업의 핵심 분야에 대한 관심이 커지면서 기존의 의사결정 문제 를 전통적인 방법론의 한계점을 최소화하는 과학적 접근 방식이 대두되고 있다. 특히 이런 과학적인 기법들은 주 로 금융 상품의 방향성을 예측하는데 사용되는데 본 연구에서는 사회적으로 관심이 높은 아파트 가격의 요인을 자기조직화지도를 통해 분석하고자 한다. 이를 위해 아파트 가격의 실질 가격을 추출하고 아파트 가격에 영향을 주는 총 16개의 입력 변수를 선정한다. 실험 기간은 1986년 1월부터 2021년 6월까지이며 아파트 가격의 상승 및 횡보 구간을 나눠 각 구간 별 변수들의 특징을 살펴본 결과, 상승 구간과 횡보 구간의 입력 변수의 통계적 성향이 뚜렷하게 구분되는 것을 알 수 있었다. 더불어 U1~U3 구간이 N1~N3 구간에 비해서 변수들의 표준편차 가 상대적으로 크게 나왔다. 본 연구는 중장기적으로 상승과 하락이라는 큰 주기를 갖고 있는 부동산에 대해서 현재 시점의 현황을 정량적으로 분석한 것에 의미가 있으며 향후 이미지 학습을 통해 미래 방향성을 예측하는 연구에 도움이 되기를 기대한다.

With the advent of the era of the fourth industry, more and more scientific techniques are being used to solve decision-making problems. In particular, big data analysis technology is developing as it becomes easier to collect numerical data. Therefore, in this study, in order to overcome the limitations of qualitatively analyzing investment trends, the association of various products was analyzed using associated analysis techniques. For the experiment, two experimental periods were divided based on the COVID-19 economic crisis, and sales information from individuals, institutions, and foreign investors was collected, and related analysis algorithms were implemented through r software. As a result of the experiment, institutions and foreigners recently invested in the KOSPI and KOSDAQ markets and bought futures and products such as ETF. Individuals purchased ETN and ETF products together, which is presumed to be the result of the recent great interest in sector investment. In addition, after COVID-19, all investors tended to be passive in investing in high-risk products of futures and options. This paper is thought to be a useful reference for product sales and product design in the financial field.

AMI(Advanced Metering Infrastructure, 지능형 계량 인프라) 구축이 활발하게 진행됨에 따 라서 전기사용량을 시간대별로 취득할 수 있고 우리생활에 밀접한 전기를 얼마나 어떻게 사용하는지 를 분석할 수 있게 됐다. 본 논문에서는 특정지역의 주택용과 일반용 전기의 여름철 한 달간 전기사 용량을 시간대별로 경부하와 최대부하로 구분하여 분석하고, 전기사용 용도에 따라 주택용과 일반용 전기의 일반적인 전기사용 패턴과 다른 이상 지역 분석을 통해 상업지역인지 주거지역인지 국세청자 료를 통해 검증을 수행하였다.

As the establishment of advanced metering infrastructure (AMI) is actively underway, electricity usage can be acquired on a time-to-hour basis and how much electricity is used that is close to our lives. In this paper, we analyzed the pattern of electricity usage by region, analyzed whether the area was a commercial area or a residential area, and performed verification by comparing it with the actual data of the National Tax Service data.

우리 주변에는 다양한 ‘무선 통신 기술’이 존재한다. 하지만 무선 통신 기술의 발달로 인해 주파수 자원을 필요로 하는 기술이 늘어나면서 주파수 부족 현상이 심각하게 대두되었다. 최근 이러한 문제를 해결할 수 있는 대안으로 떠오르는 통신기술로 ‘가시광 통신’이 주목받고 있다. ‘가시광 통신’은 직렬데이터 송수신을 기반으로 하는 통신 방법으로 발신부의 한계 및 포토다이오드를 이용하는 수신단의 문제 때문에 병렬데이터 송수신에 어려움이 있다. 본 논문에서는 이러한 가시광 통신의 병렬데이터 처리에 대하여 연구했다. 가시광 통신 시스템에 영상처리를 적용한 병렬데이터 분석을 통해 병렬데이터 분석 방법을 구현하였다. 제안 시스템에서 입출력 데이터 매칭 비교를 통해 병렬 통신 성능을 확인할 수 있었으며 병렬데이터 분석에 다양성을 제시할 것으로 기대된다.

There are various 'wireless communication technologies' around us. Wireless mobile communication has evolved through various stages, and its utilization is also diverse. However, due to the development of wireless communication technology, the demand for frequency resources is much higher than the supply, so frequency shortage is serious. Recently, 'visible light communication' has been attracting attention as an emerging communication technology that can solve the frequency shortage. 'Visible light communication' is a communication method based on serial data transmission / reception, and there is a difficulty in transmitting / receiving parallel data because the transmitter and the receiver are arbitrarily present. In this paper, we have studied parallel data processing of visible light communication. We could solve the problem by analyzing parallel data using image processing. Through this study, communication performance can be verified through I / O data comparison by implementing parallel data analysis method. It is expected that diversity in parallel data analysis will be presented through the results.

The global population is increasing rapidly because of increasing urbanization and such increasing urbanization directs the up-growing need of urban safety and preventions. This urbanization is also responsible for two things that is increased job opportunities and increased the crime rates. In this era technology has gone far more forward in a positive way. By making use of these technologies such as machine learning, artificial intelligence and big data we presented an approach through which crime pattern analysis is done. We have used apache spark (scala-programming) and machine learning algorithm for predictive crime pattern analysis. The data that we have used is a real-world data set based on Chicago city of United State of America. Our main goal of work is to define a predictive crime analysis which shows top crime patterns related to the top community areas of Chicago city.

The coat and muzzled colors in mammals are determined by the relative distribution of pheomelanin and eumelanin. In melanocytes, the expression of these two pigments is controlled by the melanocortin 1 receptor (MC1R) as well as by agouti locus alleles. For example, high expression of the AC (Adenylate cyclase 1) family genes, which interact with MC1R may affect the expression of CREB. leading to the over-expression of melanogenesis-associated genes such as PRKACB and CAMK2α. Further, increased expression of genes involved in tyrosine metabolism, such as TYR, is likely to affect the synthesis of eumelanin in the dark-muzzle cows. Our previous result, that the found intriguing distinction between the dark and light muzzle with respect to the expression of genes involved in the MAPK vs. Wnt signaling pathways. The aim of the present study was to identify melanogenesis-associated genes that are differentially expressed in the yellow and brindle coat of Korean native cattles. The first results, The expressed evaluation of MC1R protein were mostly highly between KBC to KDC, and there level at ST of KDC was higher than we expected, and decreased expressed in large sweat gland of epidermis zone. The RT-PCR analysis found that MC1R expressed at the KBC showed the same results like it was localization analyzed. Especially, levels of MC1R mRNA were elevated in HC compared to another group, while that the expression mostly higher in KDC. Next secondly results, was showed that the associated- genes of Want and MARK pathway were differential expression between KBC and KHC. Especially, in the KBC of malanogenesis major genes expression was highly, but the Want and minor genes were highly expression from KHC. In addition, analysis result of DNA methylation in Bos taurus genomes to survey the coat color–related epigenetic markers (differential methylated CpGs, DM CpGs) by RRBS was the appears that KBC has the biological functions of the DM CpG-associated genes. Regarding the increasing interest in the genetic diversity of cattle stocks, genes we identified for differential expression of melanogenesis and DM CpG-associated genes in the KBC vs. KHC may serve as novel markers for genetic diversity among cattle based on the coat color phenotype.

The application of the information and communication technologies (ICTs) towards sound solid waste management (SWM) is a challenging research area. This study is aimed at identifying evolving technological trends, competitor’s distribution and technological convergence pattern between ICTs and SWM technology. A total of 1041 patents’ applications that were submitted to Korean Intellectual Property Office (KIPO) from 1996 to 2016 were investigated as a dataset. Convergence patterns were obtained by applying the association rules of the International Patent Classification (IPC) codes. The results show some related technological fields where convergence technologies are mostly adopted and can contribute to establishing the development strategies of the ICTs - SWM technology in Korea.

Spatial point-patterns analyses were conducted to provide insight into the ecological process behind competition and mortality in two lodgepole pine (Pinus contorta Dougl. ex Loud. var. latifolia Engelm.) stands, one in the Lower Foothills, and the other in the Upper Foothills natural subregions in the boreal forest of Alberta, Canada. Spatial statistical tests were applied to live and dead trees and included Clark-Evans nearest neighbor statistic (R), nearest neighbor distribution function (G(r)), and a variant of Ripley’s K function (L(r)). In both lodgepole pine plots, the results indicated that there was significant regularity in the spatial point-pattern of the surviving trees which indicates that competition has been a key driver of mortality and forest dynamics in these plots. Dead trees generally showed a clumping pattern in higher density patches. There were also significant bivariate relationships between live and dead trees, but the relationships differed by natural subregion. In the Lower Foothills plot there was significant attraction between live and dead tees which suggests mainly one-sided competition for light. In contrast, in the Upper Foothills plot, there was significant repulsion between live and dead trees which suggests two-sided competition for soil nutrients and soil moisture.

The purpose of this study was to comparatively analyze normal gait on the plains by gender for old people reference data for the normal gait pattern for the old people. Participants were selected according to the Korean standard body type provided by the Ministry of Health and Wel-fare and used a three-dimensional motion analysis system. Cortex, Or-thotrak, and Excel were used as the software for analyzing the extract-ed data, and IBM SPSS Statistics ver. 24.0 was used for statistical anal-ysis. When data standardization was performed using the dimension-less numbers conversion, the step length and stride length of the lower extremities, which had differences between genders before dimension-less numbers conversion, showed no difference after dimensionless numbers conversion. Cadence, step time, and single support time of the left lower extremity, which had no difference between genders before dimensionless numbers conversion, were found to have significant dif-ferences after dimensionless numbers conversion. In addition, as a re-sult of analyzing the coefficient of variation value to find out the degree of change in data due to dimensionless numbers conversion, there were increase and decrease in the coefficient of variation value rang-ing from -8.11% to 6.67% before and after dimensionless numbers con-version, which means dimensionless numbers conversion can affect the statistical test.

The game of Go represents a major challenge for artificial as well as human intelligences due to its profound complexity. Although computer programs capable of playing Go have existed for decades, the period from 2015 to 2025 has marked a turning point, with these systems achieving—and even exceeding—professional human performance. A notable shift in gameplay strategy emerged around 2016, driven by the adoption of unconventional yet effective moves demonstrated by AI programs, challenging previous conventions of preferred moves. Nowadays, the process of learning Go has evolved from traditional knowledge transfer, based on centuries of established conventions regarding strategy, positional judgement and move value to modern trends developed by AI playing style, lacking the explanatory heuristics inherent in both human style of playing and oral tradition. Facing this new challenge, we rely on our curiosity to understand our opponents’ moves beyond technological and language barriers as we continue to explore the mysterious depths of this timeless game.

The proliferation of social network services affects individual’ lifestyle. Users get a lot of information through social network services. Accordingly, the social network services are useful tools, as means of promotion and marketing. In this research, we analyze the information exchange network using social network analysis in a specific-purpose trade show. That is, social features of interested parties (organizer, exhibitors and visitors) can be analyzed in information exchange network of the trade show. The aim of this research is that how the social network services influence the behavior of interested parties. We expected that the analysis can utilize the social features of interested parties to design better trade shows for promotion and be a role as an exploratory analysis for discovering a useful communication channel.

Mobile applications (apps) have become central to the digital economy, yet the proliferation of Dark Patterns—deceptive interface designs hindering user autonomy—increasingly undermines user experience (UX) and trust, posing a significant challenge. While this issue is internationally recognized (e.g., by the OECD), systematic analysis leveraging large-scale app store review data, which captures authentic user voices, to investigate the prevalence of dark patterns and compare their characteristics across service types remains limited. This study addresses this gap by applying text mining techniques to user reviews from major South Korean mobile apps, focusing on leading e-commerce and OTT streaming platforms. The primary objective is to exploratorily identify core complaint themes, keywords, and specific user experience narratives related to Dark Patterns within this dataset. Furthermore, the research aims to deepen the multifaceted understanding of the issue by comparatively analyzing whether distinct patterns of Dark Pattern-related complaints emerge across these different service types. Methodologically, the study involves collecting review data, preprocessing it using Natural Language Processing (NLP), applying LDA topic modeling and keyword analysis to uncover complaint patterns, and qualitatively examining review texts for contextual insights. The findings are expected to provide empirical evidence on the real-world landscape of Dark Pattern issues in the Korean app ecosystem, demonstrating the utility of text mining for this purpose. This research will contribute to the academic discourse on Information Systems (IS) design and UX, while offering practical insights for businesses towards ethical interface improvements and informing future consumer protection policies and regulatory considerations.