• Journal of KIISE
• /
• v.44 no.1
• /
• pp.51-56
• /
• 2017

The research area for solving question answering (QA) problems using artificial intelligence models is in a methodological transition period, and one such architecture, the dynamic memory network (DMN), is drawing attention for two key attributes: its attention mechanism defined by neural network operations and its modular architecture imitating cognition processes during QA of human. In this paper, we increased accuracy of the inferred answers, by adapting an automatic data augmentation method for lacking amount of training data, and by improving the ability of time perception. The experimental results showed that in the 1K-bAbI tasks, the modified DMN achieves 89.21% accuracy and passes twelve tasks which is 13.58% higher with passing four more tasks, as compared with one implementation of DMN. Additionally, DMN's word embedding vectors form strong clusters after training. Moreover, the number of episodic passes and that of supporting facts shows direct correlation, which affects the performance significantly.

• The Korean Journal of Food And Nutrition
• /
• v.30 no.4
• /
• pp.719-727
• /
• 2017

The objective of this study was to investigate the effect of alcohol content in Makgeolli made with Chinese matrimony vine (M-CMV) on the sensory profile and consumer acceptability. The M-CMVs were prepared with 6, 7, 8, and 9% alcohol content. Descriptive analysis of M-CMV was performed with six trained panelists. Thirteen attributes were generated and their intensities were alcohol content dependent. The consumer acceptance test was conducted with 57 consumers. M-CMV samples with 7% alcohol had the highest acceptance rate (5.8) followed by 6% M-CMV (5.6). Commercial rice Makgeolli (CRM) had the lowest consumer acceptance. Consumers were divided into two groups by clustering analysis. The majority of consumers (n=38) preferred M-CMV and did not like the commercial sample. Only 19 consumers indicated high acceptance ratings for CRM. However, these consumers also preferred 6 and 7% M-CMV. Partial least-squares regression analysis revealed moderate attribute intensities were related to greater consumer acceptability. The optimal alcohol content for the greatest consumer acceptance predicted by linear regression was 6.7%.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.39 no.2
• /
• pp.138-149
• /
• 2016

Quality function deployment (QFD) is a widely adopted customer-oriented product development methodology by translating customer requirements (CRs) into technical attributes (TAs), and subsequently into parts characteristics, process plans, and manufacturing operations. A main activity in QFD planning process is the determination of the target levels of TAs of a product so as to achieve a high level of customer satisfaction using the data or information included in the houses of quality (HoQ). Gathering the information or data for a HoQ may involve various inputs in the form of linguistic data which are inherently vague, or human perception, judgement and evaluation for the information and data. This research focuses on how to deal with this kind of impreciseness in QFD optimization. In this paper, it is assumed as more realistic situation that the values of TAs are taken as discrete, which means each TA has a few alternatives, as well as the customer satisfaction level acquired by each alternative of TAs and related cost are determined based on subjective or imprecise information and/or data. To handle these imprecise information and/or data, an approach using some basic definitions of fuzzy sets and the signed distance method for ranking fuzzy numbers is proposed. An example of a washing machine under two-segment market is provided for illustrating the proposed approach, and in this example, the difference between the optimal solution from the fuzzy model and that from the crisp model is compared as well as the advantage of using the fuzzy model is drawn.

• Journal of Digital Convergence
• /
• v.20 no.5
• /
• pp.919-929
• /
• 2022

As a new type of healthy public space, greenway users carry out leisure activities, exercise, sightseeing and necessary transportation in greenway. However, at present, there is little research on greenway users' evaluation and perception of greenway, and there is no comprehensive exploration of Greenway environment from a humanistic perspective. Combined with the research and actual situation of the existing representative greenways in Chinese cities, this paper refers to a large number of documents, applies the semantic difference method and multiple regression analysis method, analyzes the current situation of the Dongfengqu greenway, explores the influencing factors and impact evaluation of the greenway environment from the perspective of greenway users' perception, and puts forward suggestions on the optimization of the greenway environment in Zhengzhou from multiple levels. The main conclusion of this paper comes from the data conclusion obtained by semantic difference method, which is feasible in the resident evaluation of greenway use. The feedback results of post use evaluation can provide a reliable reference for the planning and design of similar greenways in the future.

• International Journal of High-Rise Buildings
• /
• v.4 no.1
• /
• pp.1-8
• /
• 2015

A team of researchers and practitioners were recently assembled to prepare a monograph on "Wind-Induced Motion of Tall Buildings: Designing for Habitability". This monograph presents a state-of-the-art report of occupant response to wind-induced building motion and acceptability criteria for wind-excited tall buildings. It provides background information on a range of pertinent subjects, including: ${\bullet}$ Physiological, psychological and behavioural traits of occupant response to wind-induced building motion; ${\bullet}$ A summary of investigations and findings of human response to real and simulated building motions based on field studies and motion simulator experiments; ${\bullet}$ A review of serviceability criteria to assess the acceptability of wind-induced building motion adopted by international and country-based standards organizations; ${\bullet}$ General acceptance guidelines of occupant response to wind-induced building motion based on peak acceleration thresholds; and ${\bullet}$ Mitigation strategies to reduce wind-induced building motion through structural optimization, aerodynamic treatment and vibration dissipation/absorption. This monograph is to be published by the American Society of Civil Engineers (ASCE) and equips building owners and tall building design professionals with a better understanding of the complex nature of occupant response to and acceptability of wind-induced building motion. This paper is a brief summary of the works reported in the monograph.

• International journal of advanced smart convergence
• /
• v.5 no.4
• /
• pp.21-25
• /
• 2016

Recent development in multi-channel surround is emerging in various formats to provide better stereoscopic and sound effects to consumers in recent broadcasting. The ability sound localize the sound sources in space is most considerable design factor on multi-channel surround system for human earing perception model. However, this paper propose the change of the sound localization according to the spacing of the speakers, which is not covered in the existing research focus on sound system design. Presently the sound system uses the position and number of the speakers to localize the sound. In the multi-channel surround environment, the proposed design uses the sound localization is caused by the directional characteristics of the speaker, the distance between the speakers and the distance between the listener and the speaker according to the directivity is required. The proposed design is simulated using virtual measurement with MATLAB simulation environment and performances are measured.

• Human Ecology Research
• /
• v.54 no.3
• /
• pp.239-250
• /
• 2016

This study identifies the aspects of a managers' perceived work experience in a Multicultural Family Support Center and analyzes how work experience backgrounds are formulated. In-depth interviews were conducted with 10 managers from March to September 2015 to understand managers' work experience. Descriptions from the interviews were analyzed using phenomenological research methods. The result show that their work experience can be categorized into 73 meanings, 10 subcategories, and five themes. The five themes are 'program development,' 'employees assignment optimization,' 'use and control network resources,' 'deal with changes in needs of multicultural families,' and 'supports vs. controls.' The analysis of the themes and subcategories from each theme allows us to first interpret that managers consider their significant and diversified work tasks overwhelming. Second, they find their jobs fit them and feel personally interested with a sense of duty from their work to overcome stress from heavy workloads. Third, managers put a high value on their work as a hands-on experience that is an officially authorized position from the government. Fourth, they are proud that they contribute to offering welfare services to multicultural families as members of Multicultural Family Support Centers.

• Proceedings of the KIEE Conference
• /
• 2006.10a
• /
• pp.145-146
• /
• 2006

본 논문은 부분방전(PD: Partial Discharge)의 패턴인식 확률 극대화를 목적으로 신경망(NN: Neural Network) 파라미터 중에서 은닉층 뉴런의 수, 모멘텀(momentum)의 Step size와 Decay rate 를 최적화하기 위하여 유전 알고리즘(GA: Genetic Algonthm)을 적응하였다. 실험적 연구의 대상으로서, GIS(Gas Insulated Switchgear)사고의 주요 원인으로 보고되어있는 결함들을 인위적으로 모의한 16개 Test cell을 이용하여 부분방전을 발생시켰다. 부분방전 신호는 본 연구팀이 개발한 센서를 이용하여 검출되어 데이터베이스가 구축되어 그로부터 추출된 학습 데이터들의 학습에 다음과 같은 5가지 신경망 모델이 적응되었다: Multilayer Perception (MLP), Jordan-Elman Network (JEN), Recurrent Network (RN), Self-Organizing Feature Map (SOFM), Time-Lag Recurrent Network (TLRN). 유전 알고리즘 적용 효율성을 분석하기 위하여 동일한 데이터를 이용하여 다음과 같은 두 가지 방법을 적용한 결과를 상호 비교하였다. 우선 상기 선택된 모델만 적용하였고 다근 하나는 상기 모델과 Genetic Algorithm이 동시에 적용되었다. 모든 모델에 대하여 학습오차와 패턴 분류 확률을 비교한 결과, 유전 알고리즘 적응 시 부분방전 패턴인식 확률이 향상되었음이 확인되어 향후 신뢰성 있는 GIS 부분방전 진단기술에 활용될 수 있을 것으로 사료된다.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.23 no.6
• /
• pp.565-574
• /
• 2015

Recently, a medium-level representation named "Stixel" has been extensively researched in stereo vision-based environmental perception. Obstacle detection using Stixel representation consists of three steps: static Stixel generation, dynamic Stixel generation, and Stixel segmentation. This paper focuses on the Stixel segmentation step and has two contributions. One is that it shows that Stixel segmentation performance can be enhanced by utilizing both image domain and real world domain features. The other is that it suggests that parameters used for Stixel segmentation can be effectively tuned based on genetic algorithm. The proposed method was quantitatively evaluated and the result showed that the proposed method increased Stixel segmentation accuracy compared with the previous method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.2
• /
• pp.383-403
• /
• 2021

With the development of mobile edge computing (MEC), some late-model application technologies, such as self-driving, augmented reality (AR) and traffic perception, emerge as the times require. Nevertheless, the high-latency and low-reliability of the traditional cloud computing solutions are difficult to meet the requirement of growing smart cars (SCs) with computing-intensive applications. Hence, this paper studies an efficient offloading decision and resource allocation scheme in collaborative vehicular edge computing networks with multiple SCs and multiple MEC servers to reduce latency. To solve this problem with effect, we propose a context-aware offloading strategy based on differential evolution algorithm (DE) by considering vehicle mobility, roadside units (RSUs) coverage, vehicle priority. On this basis, an autoregressive integrated moving average (ARIMA) model is employed to predict idle computing resources according to the base station traffic in different periods. Simulation results demonstrate that the practical performance of the context-aware vehicular task offloading (CAVTO) optimization scheme could reduce the system delay significantly.