• Journal of KIISE
• /
• v.44 no.5
• /
• pp.486-495
• /
• 2017

Collaborative Filtering is limited because of the cost that is required to perform the recommendation (such as the time complexity and space complexity). The RFWBP (Rating Frequency Weight-based Baseline Predictor) that approximates the precision of the existing methods is one of the efficiency methods to reduce the cost. But, the following issues need to be considered regarding the RFWBP: 1) It does not reduce the error because the RFWBP does not learn for the recommendation, and 2) it recommends all of the items because there is no condition for an appropriate recommendation list when only the RFWBP is used for the achievement of efficiency. In this paper, the BBP (Bias-Based Predictor) is proposed to solve these problems. The BBP reduces the error range, and it determines some of the cases to make an appropriate recommendation list, thereby forging a recommendation list for each case.

• The KIPS Transactions:PartB
• /
• v.19B no.1
• /
• pp.31-36
• /
• 2012

In recent years, there has been increasing interest in recommender systems which provide users with personalized suggestions for products or services. In particular, researches of collaborative filtering analyzing relations between users and items has become more active because of the Netflix Prize competition. This paper presents the reinforcement learning approach for collaborative filtering. By applying reinforcement learning techniques to the movie rating, we discovered the connection between a time sequence of past ratings and current ratings. For this, we first formulated the collaborative filtering problem as a Markov Decision Process. And then we trained the learning model which reflects the connection between the time sequence of past ratings and current ratings using Q-learning. The experimental results indicate that there is a significant effect on current ratings by the time sequence of past ratings.

• Journal of Korea Technology Innovation Society
• /
• v.8 no.2
• /
• pp.583-604
• /
• 2005

The future will see all industries become technology-driven in the competitive global market place. Firms with deep technological roots and innovation strategies have some advantages. In this situation widely used scoring approach is not enough to evaluate technology's relative competitiveness and to assign relative ranking category. Therefore, a more useful and comprehensive approach, which is called technology composite index, is needed to complement and enhance the existing scoring approach. In this research, factor analysis is applied to determine the common factors and to estimate associated weights. And technology composite index is used to measure the technology's relative strength and also to assign its ranking category instead of technology scoring.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.329-345
• /
• 2021

Recommender systems use data from customers to suggest personalized products. The recommender systems can be categorized into three cases; collaborative filtering, contents-based filtering, and hybrid recommender system that combines the first two filtering methods. In this work, we introduce and compare deep learning-based recommender system using autoencoder. Autoencoder is an unsupervised deep learning that can effective solve the problem of sparsity in the data matrix. Five versions of autoencoder-based deep learning models are compared via three real data sets. The first three methods are collaborative filtering and the others are hybrid methods. The data sets are composed of customers' ratings having integer values from one to five. The three data sets are sparse data matrix with many zeroes due to non-responses.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.5
• /
• pp.3025-3036
• /
• 2014

This study was conducted to figure out the influential factors on premenstrual syndrome(PMS) of college female students which are to be utilized as the basic data to develop and apply programs for preventing and controlling such symptom. The subjects were 330 college female students. The data were collected from April 2, 2012 to April 6, 2012. From the results, There has been significant correlation between stress and PMS(r=.36, p<.001) and the attitude to menstruation has appeared to have significant positive correlation with PMS as well(r=.34, p<.001). Multiple regression analysis has been employed to identify the influential factors on PMS and the result has shown that menstrual attitude, grade point average for stress, smoking and dysmenorrhea have been the most significant influential factors with 27% of explanatory power. The level of significance has been high in menstrual attitude(${\beta}$=.28, p<.001), grade point average for stress(${\beta}$=.27, p<.001), smoking(${\beta}$=.20, p<.001) and dysmenorrhea(${\beta}$=.15, p<.001) respectively. In conclusion, it needs to find nursing interventions for PMS related to psychosocial factors and suggest a narrative study for improving quality of life of women with PMS.

• Korean Journal of Occupational Health Nursing
• /
• v.1
• /
• pp.30-51
• /
• 1991

본 연구는 산업간호사가 수행하는 산업간호업무 수행실태를 파악하고 업무수행에 영향을 미치는 요인이 무엇인지를 직무만족도와 지식의 측면에서 조사 분석하기 위한 목적으로 이 연구를 시도하였다. 연구방법은 1991년 4월 26일 부터 5일 3일까지의 총 8일간 서울시내에 소재하고 있는 사업체의 산업간호사 77명을 대상으로 연구자와 조사자 1명이 직접방문을 하여 자료수집 하였다. 자료는 SPSS를 이용하여 전산처리 하였다. 실수와 백분율, 평균 표준편차를 구하였으며, Pearson Correlation Coefficient, ANOVA 분석방법올 활용하였다. 본 연구의 결과는 다음과 같다. 1. 산업 간호업무 수행 정도는 근로자건강관리업무 중의 통상질환관리 업무를 대상자의 97.6%가 실시하여 수행정도가 가장 높았고, 가장 수행정도가 낮은 업무는 근로자복리중 진업무로 대상자의 40.2%만이 실시하고 있었다. 산업환경위생관리업무(실시율 45.5%), 직업병 관리업무(실시율 43.1%), 산업보건교육업무(실시율41.5%)로 역시 낮은 실시율을 보였다. 2. 산업간호사 직무만족도는 최대가능점수 240점에 대하여 점수 143.8점이며 최대평점 5점에 평점 3.0으로 나타났고 상호작용, 전문적위치, 자율성면에서 3.4점으로 높았으며 보수에 관한 만족도는 평점 2.3으로 가장 낮아 산업간호사들이 보수에 대해 불만족하고 있는 것으로 나타났다. 3. 산업간호사의 산업간호업무 관계지식정도는 최대가능점수 21점에 대하여 점수 17.9점을 나타내었다. 영역별로 가장 많이 알고 있는 지식은 건강관리실운영업무 관계지식으로 정확하게 알고있는 사람이 63명으로 전체의 81.4%이었고, 가장 낮게 알고 있는 지식은 산업환경위생관리 업무 관계지식으로 정답자는 42명(54.2%)이었다. 4. 산업간호업무 수행정도와 직무만족도 및 지식정도와의 상관관계를 분석하여 본 결과 직무만족도가 높을수록 업무수행정도가 높은 것으로 나타났고 (r=.3010, P<.01), 업무수행정도와 지식정도 역시 순상관관계 (r=.2591, P<.05)가 있음이 통계적으로 유의하여 지식정도가 높을수록 업무수행정도가 높음을 알 수 있었다. 그러나 직무만족도와 지식정도는 상관관계가 없었다. 결론적으로 연구에서 나타난 산업 간호사의 업무수행은 통상질환관리 위주의 업무로서 직업병 예방, 산업보건교육, 산업환경위생관리 업무등에는 수행이 미비한 것으로 나타났으며, 업무수행과 직무만족도 및 지식정도는 순상관관계가 있어 직무만족도와 지식정도가 높을수록 업무수행이 높은 것으로 나타났다.

• Tunnel and Underground Space
• /
• v.15 no.1 s.54
• /
• pp.39-46
• /
• 2005

The RMR ratings, one by horizontal pre-boring in a tunnel and another by tunnel face mapping, are compared at the fault zone in a tunnel. Generally. the horizontal pre-borings were so effective as to forecast reasonably the supporting patterns after tunnel excavation. But the maximum difference in RMR ratings estimated by two methods was about 50 at a certain section of a tunnel. The differences were analyzed on each parameter of the RMR system: the rating differences were 24 in the condition of discontinuities, 15 in the RQD and 13 in the uniaxial compressive strength of rock. To minimize the gap between RMR by pre-borings and by face mappings, it is necessary to select the horizontal pre-boring location where tunnel stability could be critical and to evaluate in detail the sub-parameters of the condition of discontinuities.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.4
• /
• pp.727-736
• /
• 2014

The fundamental concerns of this paper are to analyze the effects of some characteristics on the employment of new college graduated students in viewpoint of data visualization. We use individual and department characteristic data of K-university graduated students in 2010. We apply multiple correspondence analysis, decision tree analysis, association rules and social network analysis for data visualization. The results of the analysis are summarized as follows. First, an analysis of the determinants of employment shows that GPA, department category, age and number of majors, recruiting time affect the employment rate. Second, higher GPA and natural category of department positively affect the employment rate. Finally, low age, single major and early recruiting time also positively affect the employment rate.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.5
• /
• pp.899-908
• /
• 2021

Since the modern weapon system is composed of a complex system, it is quite difficult to derive the priority of performance improvement factors. In addition, research on the method for prioritizing quantitative performance improvement factors considering the opinions of stakeholders is insufficient. In this study, in order to quantitatively derive the priorities of performance improvement factors for K1 tanks and K2 tanks, a survey was conducted with experts with experience in tank development, and the preface data was analyzed using AHP, scoring method, and TOPSIS methods. As a result, priorities were quantitatively derived. The results of this study are expected to be used as decision-making indicators for stakeholders in the future weapon system development and performance improvement stage.

• Information Systems Review
• /
• v.23 no.1
• /
• pp.155-172
• /
• 2021

With the recent increase in online product purchases, a recommender system that recommends products considering users' preferences has still been studied. The recommender system provides personalized product recommendation services to users. Collaborative Filtering (CF) using user ratings on products is one of the most widely used recommendation algorithms. During CF, the item-based method identifies the user's product by using ratings left on the product purchased by the user and obtains the similarity between the purchased product and the unpurchased product. CF takes a lot of time to calculate the similarity between products. In particular, it takes more time when using text-based big data such as review data of Amazon store. This paper suggests a hybrid recommendation system using a 2-phase methodology and text data mining to calculate the similarity between products easily and quickly. To this end, we collected about 980,000 online consumer ratings and review data from the online commerce store, Amazon Kinder Store. As a result of several experiments, it was confirmed that the suggested hybrid recommendation system reflecting the user's rating and review data has resulted in similar recommendation time, but higher accuracy compared to the CF-based benchmark recommender systems. Therefore, the suggested system is expected to increase the user's satisfaction and increase its sales.