• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.39-54
• /
• 2013

Recent explosive increase of electronic commerce provides many advantageous purchase opportunities to customers. In this situation, customers who do not have enough knowledge about their purchases, may accept product recommendations. Product recommender systems automatically reflect user's preference and provide recommendation list to the users. Thus, product recommender system in online shopping store has been known as one of the most popular tools for one-to-one marketing. However, recommender systems which do not properly reflect user's preference cause user's disappointment and waste of time. In this study, we propose a novel recommender system which uses data mining and multi-model ensemble techniques to enhance the recommendation performance through reflecting the precise user's preference. The research data is collected from the real-world online shopping store, which deals products from famous art galleries and museums in Korea. The data initially contain 5759 transaction data, but finally remain 3167 transaction data after deletion of null data. In this study, we transform the categorical variables into dummy variables and exclude outlier data. The proposed model consists of two steps. The first step predicts customers who have high likelihood to purchase products in the online shopping store. In this step, we first use logistic regression, decision trees, and artificial neural networks to predict customers who have high likelihood to purchase products in each product group. We perform above data mining techniques using SAS E-Miner software. In this study, we partition datasets into two sets as modeling and validation sets for the logistic regression and decision trees. We also partition datasets into three sets as training, test, and validation sets for the artificial neural network model. The validation dataset is equal for the all experiments. Then we composite the results of each predictor using the multi-model ensemble techniques such as bagging and bumping. Bagging is the abbreviation of "Bootstrap Aggregation" and it composite outputs from several machine learning techniques for raising the performance and stability of prediction or classification. This technique is special form of the averaging method. Bumping is the abbreviation of "Bootstrap Umbrella of Model Parameter," and it only considers the model which has the lowest error value. The results show that bumping outperforms bagging and the other predictors except for "Poster" product group. For the "Poster" product group, artificial neural network model performs better than the other models. In the second step, we use the market basket analysis to extract association rules for co-purchased products. We can extract thirty one association rules according to values of Lift, Support, and Confidence measure. We set the minimum transaction frequency to support associations as 5%, maximum number of items in an association as 4, and minimum confidence for rule generation as 10%. This study also excludes the extracted association rules below 1 of lift value. We finally get fifteen association rules by excluding duplicate rules. Among the fifteen association rules, eleven rules contain association between products in "Office Supplies" product group, one rules include the association between "Office Supplies" and "Fashion" product groups, and other three rules contain association between "Office Supplies" and "Home Decoration" product groups. Finally, the proposed product recommender systems provides list of recommendations to the proper customers. We test the usability of the proposed system by using prototype and real-world transaction and profile data. For this end, we construct the prototype system by using the ASP, Java Script and Microsoft Access. In addition, we survey about user satisfaction for the recommended product list from the proposed system and the randomly selected product lists. The participants for the survey are 173 persons who use MSN Messenger, Daum Caf$\acute{e}$, and P2P services. We evaluate the user satisfaction using five-scale Likert measure. This study also performs "Paired Sample T-test" for the results of the survey. The results show that the proposed model outperforms the random selection model with 1% statistical significance level. It means that the users satisfied the recommended product list significantly. The results also show that the proposed system may be useful in real-world online shopping store.

• Journal of Soil and Groundwater Environment
• /
• v.10 no.4
• /
• pp.1-12
• /
• 2005

Policies regarding to soil quality of the United States, the United Kingdom, the Netherlands, Germany, and Demark were analyzed to suggest Korean policy for improving soil quality concept and it's implementation. All countries met four criteria: I) Soil quality levels of contaminants are indebt to concept of contaminant risk to recipients (human and ecosystem); ii) Any soil quality value can't be a magic number to determine whether a site is contaminated or not. To determine risk of sites, risk assessment of the sites should be followed; iii) Concentrations of contaminants of sites are not always significantly certain to risk of human and ecosystem of the sites; and iv) Soil quality levels are adopted based on land uses and plans. Considering our rooms to improve policies and analysis of the other country reports on their legislations about soil quality levels, our policy implementation could be approached from these directions: i) Our concept for soil quality levels needs to develop in scientific and rational. ii) Soil quality levels and risk assessment should be implemented as determining tools of site contamination in parallel, and iii) Soil quality levels depending on land uses and plans should be developed in debt with rational and scientific concept of risk. Increasing efficacy of Korea policy regarding the soil quality levels would be in dept to applying concepts of SCL (Soil Contamination Level) and SRL (Soil Regulatory Level) developed, implementing soil quality levels and risk assessment of contaminated sites in conjunction, and classifying three distinctions of land uses based on sensitiveness of recipients (human and ecosystem) to contaminants in soil in this research.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.28 no.2
• /
• pp.1-8
• /
• 2010

"Dokseoyeoga Do"(An Enjoyable Pleasure with Reading) is a work included in a collection book of both poems and works called as "Gyunggyo Myungseung Cheop"(Collections of Beautiful Scenary in the Suburb of Seoul) with a promise that a great painter Gyumjae Jung-Sun(謙齋 鄭敾) and his lifelong friend, Sacheon Lee Byung-Yeon(李秉淵) should hopefully exchange their poems and painted works. So far, general perspective of art history toward this work is either self-portrait of Gyumjae himself or genre painting of the aristocrat. The purpose of this study is to interpret in-depth meanings implied in this work with various considerations of plant elements appeared in this work. As a result of this study, We could draw newer and additional meanings beyond the existing perspectives of art history research categories, and the details can be summarized as following below. First of all, we could approach a new interpretation implying that Gyumjae and Sacheon hopefully wished their well-being, friendship, and reunion. Second, as a means of expressing this meaning, plant elements like old juniper(Juniperus chinensis L.), orchid(Orchidaceae), and peony(Paeonia lactiflora var. hortensis) were used. Third, each plant element can be literally seen as an icon implying message that Gyumjae desired to give to Sacheon. To be more detailed, experience and greenness of old juniper imply well-being of two people, and orchid implies fragrant friendship, and peony implies the feeling that Gyumjae hopefully desired to make a reunion with a parted friend. The significance of this study is that throughout old paintings, we could observe substantial examples of recognizing the meanings of plant elements in a category of traditional landscaping and utilizing them. Moreover, adding various point-of-views of many professional categories like Korean Landscaping history in the traditional painting research, we could also recognize the potentials for making rich interpretation toward implicated meanings of old paintings.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.66 no.3
• /
• pp.190-200
• /
• 2021

The heading response of rice varieties, originated from South Korea, North Korea, and northern China, was examined under the temperature and day-length conditions of 10 low-temperature regions in North Korea. When 30-day-old broadcast tray-raised seedlings (BS) were used, only one or a few varieties reached the heading stage within the safe marginal heading date (SMHD) under the Hamhung, Pyonggang, Huichon, and Kanggye conditions. In the Yangdok and Sinpo environments, no varieties reached heading until SMHD. The South Korean varieties that reached the heading stage between SMHD and late marginal heading date (LMHD) were Jinbuol, Baegilmi, Joun, Jopum, Jinok, Jopyeong, Sanhomi, and Odae for the Hamhung, Pyonggang, Huichon, and Kanggye; Joun for Sinpo; and none for Yangdok, Kimchaek, Sonbong, and Chunggang environments. The number of days to heading (DTH) was positively correlated among all 10 regions. Reduced DTH in the 30-day-old pot tray-raised seedlings (PS), compared to BS, was not consistent with varieties and regions. Reduced DTH in PS, compared to BS, was the largest in Jinok and Sonbong9 among the South and North Korean varieties, respectively, whereas it was small in the northern Chinese varieties. The South Korean varieties that headed until SMHD in PS were Jinok for the Hamhung, Pyonggang, Yangdok, Sinpo, and Kanggye environments; and Jinbuol, Jopum, and Jinok for the Huichon condition. In the Kimchaek and Chunggang conditions, Jinok reached the heading stage within LMHD. With the use of PS for some responsive varieties such as Jinok, varietal adaptability could be improved to the conditions of low-temperature regions in North Korea. None of the PS or BS among the tested varieties reached the heading stage until LMHD for the Chongjin and Sonbong conditions.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.4
• /
• pp.428-434
• /
• 2019

Many countries have interested animal welfare. Similarly, domestic people have concern for the welfare from companion animals to livestock. Environmental enrichments (EE) are tool to enforce the welfare, however, research with sows is limited. Therefore, this study was investigate to effects of environmental enrichments on performance and behavior properties of gestating sows. A total of 30 pregnant sows (Landrace) were assigned into three treatments that control, T1 (plastic device) and T2 (Rice straw). Period of trial was from Mar. 03. 18. to Mary 19. 18. The EE were allotted to center of experimental pen ($11.6{\times}6.0m$). Body weight (BW), backfat thickness (BF) and cortisol were identified at experimental initial or end date. Behavior was recorded during 24 hours on days 91 of gestation, and then analyzed the patterns. BF was reduced (15.73 vs. 16.56 mm; p>0.05) in T1 than control, but Ending BW, total litter size and alive piglets did not differ. Born dead piglets showed lower tendency (1.00 and 0.63 vs. 1.50 heads; p>0.05) in T1 and T2 than control. Similarly, the enrichments declined farrowing mortality (C, 8.68%; T1, 6.86%; T2, 3.40%; p>0.05). Cortisol was not differed among treatments. In the behavior characteristics, eating showed lower (1.81 vs. 9.68 and 6.99%; p<0.05) in T2 than control and T1. Furthermore, playing or digging were only observed (0.33 and 2.10%; p<0.05) in T1 and T2, respectively, whereas rubbing (0.91%, p<0.05) only showed in the control. These results suggest that the provision of EE would be not negatively affected the performance of the gestating sows and could be led to improvement of the livestock welfare.

• Journal of Intelligence and Information Systems
• /
• v.27 no.2
• /
• pp.17-32
• /
• 2021

Speaker recognition is generally divided into speaker identification and speaker verification. Speaker recognition plays an important function in the automatic voice system, and the importance of speaker recognition technology is becoming more prominent as the recent development of portable devices, voice technology, and audio content fields continue to expand. Previous speaker recognition studies have been conducted with the goal of automatically determining who the speaker is based on voice files and improving accuracy. Speech is an important sociolinguistic subject, and it contains very useful information that reveals the speaker's attitude, conversation intention, and personality, and this can be an important clue to speaker recognition. The final ending used in the speaker's speech determines the type of sentence or has functions and information such as the speaker's intention, psychological attitude, or relationship to the listener. The use of the terminating ending has various probabilities depending on the characteristics of the speaker, so the type and distribution of the terminating ending of a specific unidentified speaker will be helpful in recognizing the speaker. However, there have been few studies that considered speech in the existing text-based speaker recognition, and if speech information is added to the speech signal-based speaker recognition technique, the accuracy of speaker recognition can be further improved. Hence, the purpose of this paper is to propose a novel method using speech style expressed as a sentence-final ending to improve the accuracy of Korean speaker recognition. To this end, a method called sentence sequencing that generates vector values by using the type and frequency of the sentence-final ending appearing in the utterance of a specific person is proposed. To evaluate the performance of the proposed method, learning and performance evaluation were conducted with a actual drama script. The method proposed in this study can be used as a means to improve the performance of Korean speech recognition service.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.329-352
• /
• 2022

It is reported that particulate matter(PM) penetrates the lungs and blood vessels and causes various heart diseases and respiratory diseases such as lung cancer. The subway is a means of transportation used by an average of 10 million people a day, and although it is important to create a clean and comfortable environment, the level of particulate matter pollution is shown to be high. It is because the subways run through an underground tunnel and the particulate matter trapped in the tunnel moves to the underground station due to the train wind. The Ministry of Environment and the Seoul Metropolitan Government are making various efforts to reduce PM concentration by establishing measures to improve air quality at underground stations. The smart air quality management system is a system that manages air quality in advance by collecting air quality data, analyzing and predicting the PM concentration. The prediction model of the PM concentration is an important component of this system. Various studies on time series data prediction are being conducted, but in relation to the PM prediction in subway stations, it is limited to statistical or recurrent neural network-based deep learning model researches. Therefore, in this study, we propose four transformer-based models including spatiotemporal transformers. As a result of performing PM concentration prediction experiments in the waiting rooms of subway stations in Seoul, it was confirmed that the performance of the transformer-based models was superior to that of the existing ARIMA, LSTM, and Seq2Seq models. Among the transformer-based models, the performance of the spatiotemporal transformers was the best. The smart air quality management system operated through data-based prediction becomes more effective and energy efficient as the accuracy of PM prediction improves. The results of this study are expected to contribute to the efficient operation of the smart air quality management system.

• Journal of the Korean Orthopaedic Association
• /
• v.54 no.4
• /
• pp.309-316
• /
• 2019

Purpose: This study examined the residential environment and accessibility of rehabilitation for cerebral palsy (CP) to identify the problems with residential laws pertaining to the disabled and provide basic data on the health legislation for the rights of the disabled. Materials and Methods: The literature was searched using three keywords: residence, rehabilitation, and accessibility. Two items were selected: residential environment and rehabilitation accessibility. The questionnaire included 51 items; 24 were scored using a Likert scale and 27 were in the form of multiple-choice questions. Results: This study included 100 subjects, of which 93 lived at home and seven lived in a facility. Of these 93 subjects, 65% were living in apartments, usually two or more floors above ground, and 40% of them were living without elevators. According to the Gross Motor Function Classification System, subjects with I to III belonged to the ambulatory group and IV, V were in the non-ambulatory group. Subjects from both groups who lived at home found it most difficult to visit the rehabilitation center by themselves. In contrast, among those who lived at the facility, the ambulatory group found it most difficult to leave the facility alone, while the non-ambulatory group found it most difficult to use the toilet alone. Moreover, 83% of respondents thought that rehabilitation was necessary for CP. On the other hand, 33% are receiving rehabilitation services. Rehabilitation was performed for an average of 3.6 sessions per week, 39 minutes per session. Conclusion: There is no law that ensures secure and convenient access of CP to higher levels. Laws on access routes to enter rooms are insufficient. The disabled people's law and the disabled person's health law will be implemented in December 2017. It is necessary to enact laws that actually reflect the difficulties of people with disabilities. Based on the results of this study, an investigation of the housing and rehabilitation of patients with CP through a large-scale questionnaire will necessary.

• (The) Research of the performance art and culture
• /
• no.39
• /
• pp.549-580
• /
• 2019

There are many festivals in the region that mainly use traditional performing arts. Traditional performing arts has a story that incorporates the history of the area, and it is easy to build a brand that can be distinguished from other areas through traditional performing arts, so it has conditions that are easy to grow in the local festival it can. In this paper, I researched the relationship between the significance and regionality of regional performing arts, referring to the case where traditional performing arts, which is a joint cultural heritage of East Asia, are transmitted as a festival. The performance art with strong locality has grown into a local festival, and the Kanto Festival (竿燈) held in Akita prefecture in northeastern Japan was selected as a representative example of that area. Along with the Aomori Nebuta Festival and the Sendai Tanabata Festival, the Kanto Festival is called the 3 major festivals in the northeastern part of Japan. It was designated as an important intangible folk cultural property in 1980 and became more famous. It visited Seoul several times after the 2005 Japan-Korea Exchange Festival event. It is widely known as a regional festival that represents Japan. The Kanto Festival, which was a participatory event on a village basis, has faced problems such as the migration and aging of young people since the 1970s. In order to solve this, they led the participation of schools, educational institutions, and various groups beyond the village unit and persuaded the participation to the local companies. They have been steadily promoting free lectures on technical skills and school visit events that induce children's interest. As a result, the number of moths mobilized in the current festival has reached 250, and the Kanto tournament also shows great popularity every year.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.4
• /
• pp.37-50
• /
• 2023

Venture capital invests the necessary capital and supports management and technology in promising small and medium-sized venture companies in the early stages of start-up with promising technology and excellent manpower. It plays a role as a key player in the venture ecosystem that realizes profits by collecting the investment through various means after growth. Venture capital's job is to recruit various investors(LPs) to invest in small and medium-sized venture companies with growth potential through the formation of venture investment funds, and to collect investment as companies grow, distribute and reinvest. The main tasks of venture capitalists, which play the most important role in venture investment, are finding promising companies, corporate analysis and evaluation, investment screening, follow-up management, and investment recovery. Venture capital's success indicators are fund formation and return on investment, and venture capitalists are rewarded with annual salary, performance-based incentive, and promotion with work performance such as investment, exit, and fund formation. Compared to the recent rapidly growing venture investment market, investment manpower is insufficient, and venture capital is making great efforts to foster manpower and establish infrastructure and systems for long-term service, but research has been conducted mainly from a quantitative perspective. Accordingly, this study aims to empirically analyzed the impact of investment experience, delegation of authority, job fit, and peer relationships on fund formation and return on investment according to the characteristics of the venture capital industry. The results of these empirical studies suggested that future venture capital needs a job environment and manpower operation strategy so that venture capitalists with high job fit and investment experience can work for a long time.