• Proceedings of the Korea Database Society Conference
• /
• 2010.06a
• /
• pp.109-123
• /
• 2010

Compared to research on the success of IT projects In general, research on the management of conflict in IT teams and its impact on project success is insufficient. This research performed a comparative study to analyze the nature and evolving phenomenon of conflicts in IT project teams. This study aimed at; First, Why a conflict is created within and around an IT project team. Second, What is the phenomenon of conflict evolution, diffusion, and/or resolution? Third, How the differences in conflict management style affects the IT project performance? Three propositions were drawn from the analysis of two intensive analysis of conflict cases and their comparison. Proposition 1 : Both 'the method of pointing out a problem' and 'the pointed-out contents' can be the trigger of conflicts around an IT project. Proposition 2 : While the focus of attention of one party is 'the method of pointing out a problem', if the other's and focus of attention of one party is 'the pointed-out contents', that is, if the focus of attention mismatch, the conflict can be amplified. Proposition 3 : According to the temporal nature of the relationship, e.g., short-term versus long-term, the way people approach and deal with the conflict differs.

• Journal of the Korea Computer Graphics Society
• /
• v.13 no.3
• /
• pp.1-6
• /
• 2007

Anticipation effect has been used as a traditional skill to enhance the dynamic motion of the traditional 2D animation. Basically, anticipation means the action of opposite direction which performs before the real action step. In this paper, we propose the perception-based video anticipation method to guide a user's visual attention to the important region. Using the image based attention map, we calculate the visual attention region and then combine this map with temporal saliency of video. We apply the anticipation effect in these saliency regions using the blur kernel. Using our method, we can generate the dynamic video motion which has attentive guidance.

• International Journal of Contents
• /
• v.15 no.4
• /
• pp.8-15
• /
• 2019

This paper presents a solution of the 'Quick, Draw! Doodle Recognition Challenge' hosted by Google. Doodles are drawings comprised of concrete representational meaning or abstract lines creatively expressed by individuals. In this challenge, a doodle is presented as a sequence of sketches. From the view of at the sketch level, to learn the pattern of strokes representing a doodle, we propose a sequential model stacked with multiple convolution layers and Long Short-Term Memory (LSTM) cells following the attention mechanism [15]. From the view at the image level, we use multiple models pre-trained on ImageNet to recognize the doodle. Finally, an ensemble and a post-processing method using the minimum cost flow algorithm are introduced to combine multiple models in achieving better results. In this challenge, our solutions garnered 11th place among 1,316 teams. Our performance was 0.95037 MAP@3, only 0.4% lower than the winner. It demonstrates that our method is very competitive. The source code for this competition is published at: https://github.com/ngxbac/Kaggle-QuickDraw.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.11
• /
• pp.947-953
• /
• 2020

The main purpose of this study is to find the factors affecting the consumer goods buyers' choice on e-commerce sites in Vietnam. By using the quantitative method, the paper examines the theoretical research model and tests four hypotheses. The sample was drawn from the population of e-commerce sites in Vietnam comprising about 1,000 respondents. This study used the questionnaire method to collect primary data to test the hypotheses. Data analysis of the questionnaire was done using SPSS. The results show that there is a correlation between personal preferences of consumers on colors and brands and their actual final choice on e-commerce sites in Vietnam. The most important factor affecting the consumer goods buyers' choice on e-commerce in Vietnam is the brand of E-commerce site, following by the color and position. Among all educational groups, respondents with a Master degree pay the biggest attention to the site's color attribute, and those with a Bachelor degree pay more attention to the brand attribute. Women pay much more attention to the location of the products on the screen than men, as do consumers with a PhD degree, over respondents with a Bachelor or Master degree, and foreigners over Vietnamese consumers.

• Archives of design research
• /
• v.13 no.3
• /
• pp.111-122
• /
• 2000

To achieve the advertising objectives, we must first draw out consumer's advertisingto advertising, and then realize the data processing course from comprehension to the retention of the advertising contents. In current media environment, however, the value of advertising as the method of attracting consumer's attention is declining due to the current situation of newspaper and magazine advertising as follows. As for newspapers, competitive messages are increasing on a continuous basis; the size of advertising section and the number of newspaper pages has been increased along with the popularization of color advertising. As for magazines, more pages are allocated for advertising rather than for the actual contents. In such communication environment, it is believed that only the advertising capable of uniquely appealing to consumers can achieve the purpose of advertising. Hence, this research attempts to discuss the effective ways of promoting attention in print media through the case studies covering the design principles relating to layout and the layout format, among the visualization processes related to the consumers'memory.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.5
• /
• pp.178-188
• /
• 2020

This paper proposes a lane recognition CNN network using split-attention network as a backbone to extract feature. Split-attention is a method of assigning weight to each channel of a feature map in the CNN feature extraction process; it can reliably extract the features of an image during the rapidly changing driving environment of a vehicle. The proposed deep neural networks in this paper were trained and evaluated using the Tusimple data set. The change in performance according to the number of layers of the backbone network was compared and analyzed. A result comparable to the latest research was obtained with an accuracy of up to 96.26, and FN showed the best result. Therefore, even in the driving environment of an actual vehicle, stable lane recognition is possible without misrecognition using the model proposed in this study.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.22 no.4
• /
• pp.246-252
• /
• 2011

Background : The objective of this study was to evaluate the diagnostic efficiency of the Comprehensive Attention Test (CAT) by comparing the results of the continuous performance test (CPT) and the CAT in children and adolescents with attention-deficit hyperactivity disorder (ADHD). Method : A total of 110 children and adolescents with ADHD (mean age : 11.2${\pm}$2.9 years, 76 boys) and 36 children and adolescents without ADHD (mean age : 12.0${\pm}$2.7 years, 25 boys) completed the CAT. We compared the specificity and sensitivity of the CPT and CAT at two different cut-off points via the McNemar test. The areas under the receiver operating characteristic curves (AUC) between the two groups were compared using Medcalc software. Results : The sensitivity and specificity of the CPT were .419 and .806, respectively. The sensitivity of the CAT was .827, which was significantly higher than that of the CPT (p<.001), and the specificity of CAT is .444, which was significantly lower than that of CPT (p<.001). The AUC of the CPT and CAT was .664 and .692, respectively, and there were no significant difference between the two groups on the paired comparison (p=.513). Conclusions : The results of this study suggest that the CAT has moderate specificity and high sensitivity. The CAT can be used as a useful tool to evaluate the neuropsychological function of children and adolescents with ADHD.

• The Journal of the Korea Contents Association
• /
• v.19 no.4
• /
• pp.102-110
• /
• 2019

Among the visual sensibility studies, many studies of color or movement have been done, but not much have been done about whether it can evoke sensibility in static form itself. Therefore, in this study, visual attention by AOIs(Area of Interests) combined with color based on three basic forms was analyzed using eye tracking, and the results were visually expressed through Heat Map and Gaze Plot. In addition, A Paired t-test was performed on the mean difference between the two groups to verify the statistical significance of each color and form. As a result of the experiment, the chromatic color form was more visual attention than the achromatic color form and warm color form was higher mean than cool color, so the visual attention was greater. In this study, it is meaningful to suggest a quantitative method which is easy to interpret objectively the design element that is easily interpreted subjectively. Based on the results of this study, if more further studies and quantitative analysis methods are presented that can identify differences in visual attention from various colors and forms, it can be used to provide guidelines for basic design.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.2
• /
• pp.94-101
• /
• 2023

A deep neural network composed of encoders and decoders, such as U-Net, used for speech enhancement, concatenates the encoder to the decoder through skip-connection. Skip-connection helps reconstruct the enhanced spectrum and complement the lost information. The features of the encoder and the decoder connected by the skip-connection are incompatible with each other. In this paper, for complex-valued spectrum based speech enhancement, Self-Attention (SA) method is applied to skip-connection to transform the feature of encoder to be compatible with the features of decoder. SA is a technique in which when generating an output sequence in a sequence-to-sequence tasks the weighted average of input is used to put attention on subsets of input, showing that noise can be effectively eliminated by being applied in speech enhancement. The three models using encoder and decoder features to apply SA to skip-connection are studied. As experimental results using TIMIT database, the proposed methods show improvements in all evaluation metrics compared to the Deep Complex U-Net (DCUNET) with skip-connection only.

• Proceedings of The KACE
• /
• 2017.08a
• /
• pp.253-256
• /
• 2017

딥러닝으로 만들어진 모델의 내부는 black box와 같은 특성을 가져 동작 규칙을 알기 어렵다. 최근 기계학습의 발전으로 인공지능이 전보다 더 복잡한 문제를 해결할 수 있으나 위와 같은 이유로, 모델이 내린 판단의 근거를 알기 어렵다. 그러므로 딥러닝의 동작 규칙을 사람이 이해할 수 있는 형식으로 나타내려는 노력이 필요하다. 본 연구에서는 Attention과 LIME 기법을 활용하여 IMDB 데이터를 감성 분석한 순환신경망의 의사결정 요인을 분석하였다. 각 기법을 활용했을 때의 장단점과 실제 구현에 있어 등장하는 문제에 대해 알아보고자 한다.