• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.1
• /
• pp.47-55
• /
• 2024

We spend a lot of time in indoor space, and the space has a huge impact on our lives. Interior design plays a significant role to make an indoor space attractive and functional. However, it should consider a lot of complex elements such as color, pattern, and material etc. With the increasing demand for interior design, there is a growing need for technologies that analyze these design elements accurately and efficiently. To address this need, this study suggests a deep learning-based design analysis system. The proposed system consists of a semantic segmentation model that classifies spatial components and an image classification model that classifies attributes such as color, pattern, and material from the segmented components. Semantic segmentation model was trained using a dataset of 30000 personal indoor interior images collected for research, and during inference, the model separate the input image pixel into 34 categories. And experiments were conducted with various backbones in order to obtain the optimal performance of the deep learning model for the collected interior dataset. Finally, the model achieved good performance of 89.05% and 0.5768 in terms of accuracy and mean intersection over union (mIoU). In classification part convolutional neural network (CNN) model which has recorded high performance in other image recognition tasks was used. To improve the performance of the classification model we suggests an approach that how to handle data that has data imbalance and vulnerable to light intensity. Using our methods, we achieve satisfactory results in classifying interior design component attributes. In this paper, we propose indoor space design analysis system that automatically analyzes and classifies the attributes of indoor images using a deep learning-based model. This analysis system, used as a core module in the A.I interior recommendation service, can help users pursuing self-interior design to complete their designs more easily and efficiently.

• Journal of the Korean Society of Food Culture
• /
• v.23 no.5
• /
• pp.605-615
• /
• 2008

The principal objective of this study was to categorize service attributes on the basis of the asymmetric and non-linear relationship existing between service attributes and customer satisfaction. Researchers generally assume that service attribute performances and customer satisfaction are both symmetrical and linear. That is to say, improvements in attribute performance will inevitably result in increased customer satisfaction. However, this is not always the case. Certain attributes have been shown not to create satisfaction even when improved, and others do not create dissatisfaction even when their performance ratings become negative. Understanding this relationship is crucial not only to researchers, but also to service managers. Service managers can arrange their priorities with regard to which attributes must be improved or promoted first, in an environment of limited technical, financial, and human resources. Many studies into this asymmetric and non-linear relationship have recently been conducted, beginning with Herzberg's motivation-hygiene theory (1976) and the disconfirmation theory, which was eventually developed into Kano's model (1984). This study attempted to determine the impact level of service attributes on incidents of satisfaction or dissatisfaction. It used 30 service attributes generated by Park (2008) in the CIT research into family restaurants. The data were collected from 600 participants, 300 incidences of satisfaction and 300 incidents of dissatisfaction, via an online survey. The t-test was used to confirm the difference between the satisfaction group's and dissatisfaction group's attributes. 11 attributes were found to be significant at a level of p>0.05. This indicates that the 11 attributes exerted different impacts on satisfaction and dissatisfaction, which confirmed the asymmetric and non-linear relationship. 14 attributes were categorized into the core service, 1 attribute into the quality service, 7 attributes into the basic service, and 8 attributes into the neutral service. Strategic customer service management was recommended for the 'A' family restaurant as an example, on the basis of the asymmetric and non-linear relationship and the characteristics of the four service factors.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.2
• /
• pp.166-172
• /
• 2014

Image classification is an important problem in computer vision. However, it is a very challenging problem due to the variability, ambiguity and scale change that exists in images. In this paper, we propose a method of a hypergraph based modeling can consider the higher-order relationships of semantic attributes of a scene image and apply it to a scene image classification. In order to generate the hypergraph optimized for specific scene category, we propose a novel search method based on a probabilistic subspace method and also propose a method to aggregate the expression values of the member semantic attributes that belongs to the searched subsets based on a linear transformation method via likelihood based estimation. To verify the superiority of the proposed method, we showed that the discrimination power of the feature vector generated by the proposed method is better than existing methods through experiments. And also, in a scene classification experiment, the proposed method shows a competitive classification performance compared with the conventional methods.

• Journal of the Korean Chemical Society
• /
• v.60 no.6
• /
• pp.415-427
• /
• 2016

This study analyzed the attributes of the conceptions of $3^{rd}$ to $6^{th}$ grade elementary school students on three states of matter and investigated the characteristics of the classified results of various examples of matter by grades. Through discussion activities, we confirmed the stabilization of conception attributions. For this study, 113 participants from two $3^{rd}$ to $6^{th}$ grade elementary school classes were selected. The concentration analysis (C-factor) and normalized gain (G-factor) of the conceptions for the quantitative analysis of the conception changes were used. The elementary school students retained different percentages of the attributes for states of matter. The characteristic of the grades were different between the 3rd grade and other grades. Based on these results, we pointed out the problems with the present teaching methods in science textbooks and stated the advantages of the effects of the representation of mixtures.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.5
• /
• pp.358-365
• /
• 2016

We propose a novel application of random forest, a machine learning-based general classification algorithm, to analyze the influence of design attributes on the silicon-to-SPICE (S2S) gap. To improve modeling accuracy, we introduce magnification of learning data as well as randomization for the counting of design attributes to be used for each tree in the forest. From the automatically generated decision trees, we can extract the so-called importance and impact indices, which identify the most significant design attributes determining the S2S gap. We apply the proposed method to actual silicon data, and observe that the identified design attributes show a clear trend in the S2S gap. We finally unveil 10nm key fin-shaped field effect transistor (FinFET) structures that result in a large S2S gap using the measurement data from 10nm test vehicles specialized for model-hardware correlation.

• Journal of Information Technology Applications and Management
• /
• v.10 no.4
• /
• pp.29-50
• /
• 2003

This research provide the basic information to enhance the user-orientedness of usability design guidelines for software products and an effective empirical guidance to classify design attributes of internet shopping mall sites. The results of analysis show that design attributes can be classified into the procedural attribute group, the shopping tool attribute group, the visual attribute group, linguistic attribute group, and others. The results show that shopping tool attribute group can be divided further into the search tool attribute group and purchase tool attribute group and that the visual attribute group can be divided further into the screen condition attribute group and the character legibility attribute group. The research reveals that when designers design software interfaces and features they should take the compound effect of a group of design attributes into consideration to enhance the usability of the system.

• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.1
• /
• pp.37-45
• /
• 2022

It is well known that the Kano model measures customer satisfaction and classifies quality attributes into must-be, attractive as well as one-dimensional. The main purpose of this study is to investigate the dynamics of e-learning quality attributes by applying the proposed method using Kano's satisfaction index in the rapidly changing online learning environment. For this, the current study examined 27 e-learning quality attributes and conducted a comparative study using Kano's results obtained in 2013 and 2020. The result shows that the dynamics of quality attributes suggested by Kano(2001) is confirmed in the case of e-learning. The proposed approach shows better results in terms of Kano's direct classification method, and has potential application areas such as IPA(Importance-Performance Analysis) in the area of risk assemement. Some suggestions for better understanding of the proposed SI-DI diagram are also included in this study.

• Journal of Korea Multimedia Society
• /
• v.20 no.4
• /
• pp.704-712
• /
• 2017

Various researches on content sensibility with the development of digital contents are under way. Emotional research on fonts is also underway in various fields. There is a requirement to use the content expressions in the same way as the content, and to use the font emotion and the textual sensibility of the text in harmony. But it is impossible to select a proper font emotion in Korea because each of more than 6,000 fonts has a certain emotion. In this paper, we analysed emotional classification attributes and constructed the Hangul font recommendation system. Also we verified the credibility and validity of the attributes themselves in order to apply to Korea Hangul fonts. After then, we tested whether general users can find a proper font in a commercial font set through this emotional recommendation system. As a result, when users want to express their emotions in sentences more visually, they can get a recommendation of a Hangul font having a desired emotion by utilizing font-based emotion attribute values collected through the crowdsourced method.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.5
• /
• pp.83-87
• /
• 2011

Naive Bayesian learning has been widely used in machine learning. However, in traditional naive Bayesian learning, we make two assumptions: (1) each attribute is independent of each other (2) each attribute has same importance in terms of learning. However, in reality, not all attributes are the same with respect to their importance. In this paper, we propose a new paradigm of calculating the importance of attributes for naive Bayesian learning. The performance of the proposed methods has been compared with those of other methods including SBC and general naive Bayesian. The proposed method shows better performance in most cases.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1031-1033
• /
• 2005

This paper presents a system of handwritten numerals recognition, which is based on Ant-miner algorithm (data mining based on Ant colony optimization). At the beginning, three distinct fractures (also called attributes) of each numeral are extracted. The attributes are Loop zones, End points, and Feature codes. After these data are extracted, the attributes are in the form of attribute = value (eg. End point10 = true). The extraction is started by dividing the numeral into 12 zones. The numbers 1-12 are referenced for each zone. The possible values of Loop zone attribute in each zone are "true" and "false". The meaning of "true" is that the zone contains the loop of the numeral. The Endpoint attribute being "true" means that this zone contains the end point of the numeral. There are 24 attributes now. The Feature code attribute tells us how many lines of a numeral are passed by the referenced line. There are 7 referenced lines used in this experiment. The total attributes are 31. All attributes are used for construction of the classification rules by the Ant-miner algorithm in order to classify 10 numerals. The Ant-miner algorithm is adapted with a little change in this experiment for a better recognition rate. The results showed the system can recognize all of the training set (a thousand items of data from 50 people). When the unseen data is tested from 10 people, the recognition rate is 98 %.