• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.4
• /
• pp.270-275
• /
• 2016

We present an approach for acceleration actor-critic algorithm for reinforcement learning with continuous action space. Actor-critic algorithm has already proved its robustness to the infinitely large action spaces in various high dimensional environments. Despite that success, the main problem of the actor-critic algorithm remains the same-speed of convergence to the optimal policy. In high dimensional state and action space, a searching for the correct action in each state takes enormously long time. Therefore, in this paper we suggest a search accelerating function that allows to leverage speed of algorithm convergence and reach optimal policy faster. In our method, we assume that actions may have their own distribution of preference, that independent on the state. Since in the beginning of learning agent act randomly in the environment, it would be more efficient if actions were taken according to the some heuristic function. We demonstrate that heuristically-accelerated actor-critic algorithm learns optimal policy faster, using Educational Process Mining dataset with records of students' course learning process and their grades.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.18 no.4
• /
• pp.283-294
• /
• 2014

In this paper we study consumption-labor supply decision of an agent who prepares for retirement at a known time in the future. The agent is assumed to have a preference which is represented by the von Neumann-Morgenstern utility function in which the felicity function has constant relative risk aversion over the composite of consumption and leisure. The composite is obtained by the Cobb-Douglas function. A general problem has been studied by Bodie et al. (2004). We contribute to the literature by deriving the Slutsky equations and conducting comparative statics. In particular, we show that wealth effect can exhibit an interesting property depending upon the time until retirement, as the interest rate increases.

• Korean Journal of Organic Agriculture
• /
• v.25 no.2
• /
• pp.257-277
• /
• 2017

This study was conducted to suggest policy implications of providing public function of organic farming. Define the difficulties of rice farmers to provide public utility of organic farming. We estimated the amount of the farmers would like to receive for the practice of organic agriculture using choice experiment methods. By evaluating individual economic values through WTA estimates, its meaningful to grasp farmers preference for organic farming practice and conversion. These results could be used as useful data for future organic agriculture policy. For the expansion of organic farming, it is required to evaluate correct and accurate economic value and reflect this in policy.

• Journal of the Korean Society of Systems Engineering
• /
• v.1 no.2
• /
• pp.56-62
• /
• 2005

In this study, we first define user requirements to fulfill the reconnaissance and the search missions, by analyzing the system characteristics and operation environment. By investigating the design technology level, the development and procurement costs, the strong system design concepts and possible alternatives will be proposed. To analyze the system requirements, the Quality Function Deployment of the systems engineering approach will be implemented. The promising design alternatives that satisfy the user requirements are extracted by constructing the Morphological Matrix, then the best design concept will be obtained using the Pugh Concept Selection Matrix and the TOPSIS(Technique of Order Preference by Similarity to Ideal Solution).

• KIEAE Journal
• /
• v.12 no.5
• /
• pp.53-61
• /
• 2012

As aging society progresses, it is necessary to establish a housing standard that provides a healthy, safe, and convenient environment for the elderly. The objective of the study is to propose a housing space standard that is geared towards elderly living in the urban area. Three steps were taken to attain the objective. First, characteristics of housing preference, spatial usage and furniture preference of the elderly were attained through a survey. The survey was conducted targeting the respondents of the ages 60 to 79 living in three different administrative districts in Seoul. Second, the required floor space to perform individual specific behavior using anthropometric dimension of the elderly was investigated by literature review. Lastly, floor space of each space is suggested combining each area for individual specific behavior which reflects characteristics of housing preference and spatial usage. The results of the research as follows. A bedroom was planned, which includes the function of receiving guests by giving the area for the placement of a sofa. The space for the use of dinning table for two person was planned in the kitchen and The space for the behavior of showering on a chair was planned in a bathroom. Lastly, the space for the behavior of taking off shoes on a chair was considered in the entrance area. Consequently, $35.4m^2$ is suggested as the necessary floor space for the elderly housing considering the characteristics of the elderly in urban area. However, because of space composition and space for accessibility, the additional space should be considered to plan the elderly housing.

• Archives of design research
• /
• v.12 no.3
• /
• pp.105-114
• /
• 1999

As the microelectronics technology is developed, new types of smart intelligent products are being emerged. OSD user interface is one of the critical factor in this kind of product, especially brown goods and information devices, as it is responsible for imput and output function. OSD is being treated as accompaniment to hardware in spite of its importance, and therefore is developed from only simple and separate usability testing based on performance measurement. This study propose a usability evaluation method of OSD based on subjective preference to support existing usability testing. The purpose of this analysis is to make clear what is important factor and how its preference level is from the user's viewpoint. The various attributes of OSD are clarified from user's questionaire and interview, and orthogonal array is generated with specified factor levels. The prototypes are generated from rapid prototyping tool and tested in natural simulation environment. The preference data which collected in this usability testing is analyzed with conjoint analysis module. This usability evaluation is not the final stage in user interface design process but the early planned and circulated stage.

• Journal of Internet Computing and Services
• /
• v.5 no.4
• /
• pp.53-62
• /
• 2004

This paper presents a study on the optimization for brokering between cargos and ships for future e-logistics. The primary contribution of this research is that we establish an optimization model by formalizing the criteria for the brokering such as time constraints, weight constraints, and preference values between cargos and ships. Another important contribution is that we not only investigate the complexity and the tractability of the optimal brokering problem but present how to evaluate the performance of the optimization program through an experiment. We first derive the preference values between cargos and ships using the time and the weight constraints. These preference values between each pair of cargos and ships are assigned to corresponding binary decision variables as coefficients in the objective function. The optimization model selects pairs of cargos and ships in a way that the sum of the preference values is maximized while satisfying given constraints. Experiment shows that the Davis-Putnam based optimization program finds optimal solutions in reasonable time for the problems with less than 90 decision variables.

• Proceedings of the Korea Contents Association Conference
• /
• 2007.11a
• /
• pp.146-150
• /
• 2007

As the internet has become a mainstream information tool, searching answers has become crucial as well. The collaborative filtering estimates and recommends items based upon the similar preference. However, because it refers to partial users information who have the similar preference, it tends to ignore the rest. In this paper, we propose the apparel coordination based on human sensibility ergonomics using the female students preference. This proposed method calculates evaluation values using fitness function based genetic algorithm, and gathers users through a-cut. Finally, the collaborative filtering recommends apparel coordination. To estimate the performance, the suggested method is compared with FAIMS-I, FAIMS-II in the questionnaire dataset.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.42 no.2
• /
• pp.1-17
• /
• 2017

Based on increased popularity and use of social network services as a marketing tool, social commerce became an emerging trend in e-commerce platforms. Social commerce involves sellers offering potential consumers the products and services at a lower price in a limited time period. Through comparison of the performances of domestic social commerce websites, we found that the buying environment such as price, number of available products, and the remaining time period for sale has a significant difference influencing on the purchase decisions of consumers. This study aims to analyze the interaction effects and preference levels of four characteristics (price, discount rate, number of purchases and purchase time) by conducting choice-based conjoint analysis. Survey experiment was performed using a sample of 146 undergraduate and graduate students. The results showed that consumers importantly consider purchase time, discount rate, price, number of purchases in the order of their preference. Also, discount effect is more significant on purchase decisions than price effect and consumers distinguish less the differences among the buying environment characteristics in the closing days of purchase period. Customer segmentation using the preference levels of characteristics indicates that the preference levels have different effects in the purchase utility of each segment. The proposed customer segmentation and differences in feature utilities are expected to be valuable in forming future sales promotion strategies in social commerce.

• The Korean Journal of Community Living Science
• /
• v.19 no.4
• /
• pp.481-495
• /
• 2008

The aim of this study was to investigate the effects of socio-demographic and health-related perceptions and twenty dietary behaviors on the superiority and preference, convenient use of and purchase motives of vegetables among 370 women(mean age 43.9) in Seoul and metropolitan and Gangwon area. Data was collected using a questionnaire developed in advanced studies. Almost all subjects rated vegetables more predominantly than grains or meats on the basis of health, diversity, and convenience to purchase. Recognition of healthly funcions of diet, purpose of diet and some dietary behaviors had influence on preference for vegetables. Those who recognized that diet had the function of disease control and prevention, knew the healthy options vegetables provide us, such as, preventing us from geriatric diseases, good sources of vitamins and minerals and considered other motives aside from health issues. From Pearsons's correlation analysis, it was revealed that sufficient time for meal preparation and pleasant eating and desirable life style and dietary attitudes, and a better understanding of nutrition were positively correlated with consumption of more vegetables and convenient use of vegetables. The subjects who recognized the ability of food to help prevent diseases, had a preference for vegetables and attitude to use nutritional knowledge and also considered the nutrition values of vegetables and environment friendliness. Everyday use of vegetables such as consuming a variety of different vegetables, a families favorite vegetables and variation of cooking methods had correlated positively with purchase motives. Therefore to encourage consumption of vegetables, it is important to notice three points continuously; first, practical information of nutritive values of vegetables, the contribution of vegetables in current animal protein foods centered diet and healthful functionalities. Second, the importance of regular diet and steadfast attitude and lifestyle. Finally, the quality of the market where the products are bought such as, quality control of freshness, sanitation, expression of producer and a pleasant environment to shop in.