• Proceedings of the Optical Society of Korea Conference
• /
• 2003.07a
• /
• pp.60-61
• /
• 2003

A new approach to reduce the computation time of genetic algorithm (GA) for making binary phase holograms is described. Synthesized holograms having diffraction efficiency of 75.8% and uniformity of 5.8% are proven in computer simulation and experimentally demonstrated. Recently, computer-generated holograms (CGHs) having high diffraction efficiency and flexibility of design have been widely developed in many applications such as optical information processing, optical computing, optical interconnection, etc. Among proposed optimization methods, GA has become popular due to its capability of reaching nearly global. However, there exits a drawback to consider when we use the genetic algorithm. It is the large amount of computation time to construct desired holograms. One of the major reasons that the GA' s operation may be time intensive results from the expense of computing the cost function that must Fourier transform the parameters encoded on the hologram into the fitness value. In trying to remedy this drawback, Artificial Neural Network (ANN) has been put forward, allowing CGHs to be created easily and quickly (1), but the quality of reconstructed images is not high enough to use in applications of high preciseness. For that, we are in attempt to find a new approach of combiningthe good properties and performance of both the GA and ANN to make CGHs of high diffraction efficiency in a short time. The optimization of CGH using the genetic algorithm is merely a process of iteration, including selection, crossover, and mutation operators [2]. It is worth noting that the evaluation of the cost function with the aim of selecting better holograms plays an important role in the implementation of the GA. However, this evaluation process wastes much time for Fourier transforming the encoded parameters on the hologram into the value to be solved. Depending on the speed of computer, this process can even last up to ten minutes. It will be more effective if instead of merely generating random holograms in the initial process, a set of approximately desired holograms is employed. By doing so, the initial population will contain less trial holograms equivalent to the reduction of the computation time of GA's. Accordingly, a hybrid algorithm that utilizes a trained neural network to initiate the GA's procedure is proposed. Consequently, the initial population contains less random holograms and is compensated by approximately desired holograms. Figure 1 is the flowchart of the hybrid algorithm in comparison with the classical GA. The procedure of synthesizing a hologram on computer is divided into two steps. First the simulation of holograms based on ANN method [1] to acquire approximately desired holograms is carried. With a teaching data set of 9 characters obtained from the classical GA, the number of layer is 3, the number of hidden node is 100, learning rate is 0.3, and momentum is 0.5, the artificial neural network trained enables us to attain the approximately desired holograms, which are fairly good agreement with what we suggested in the theory. The second step, effect of several parameters on the operation of the hybrid algorithm is investigated. In principle, the operation of the hybrid algorithm and GA are the same except the modification of the initial step. Hence, the verified results in Ref [2] of the parameters such as the probability of crossover and mutation, the tournament size, and the crossover block size are remained unchanged, beside of the reduced population size. The reconstructed image of 76.4% diffraction efficiency and 5.4% uniformity is achieved when the population size is 30, the iteration number is 2000, the probability of crossover is 0.75, and the probability of mutation is 0.001. A comparison between the hybrid algorithm and GA in term of diffraction efficiency and computation time is also evaluated as shown in Fig. 2. With a 66.7% reduction in computation time and a 2% increase in diffraction efficiency compared to the GA method, the hybrid algorithm demonstrates its efficient performance. In the optical experiment, the phase holograms were displayed on a programmable phase modulator (model XGA). Figures 3 are pictures of diffracted patterns of the letter "0" from the holograms generated using the hybrid algorithm. Diffraction efficiency of 75.8% and uniformity of 5.8% are measured. We see that the simulation and experiment results are fairly good agreement with each other. In this paper, Genetic Algorithm and Neural Network have been successfully combined in designing CGHs. This method gives a significant reduction in computation time compared to the GA method while still allowing holograms of high diffraction efficiency and uniformity to be achieved. This work was supported by No.mOl-2001-000-00324-0 (2002)) from the Korea Science & Engineering Foundation.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.32 no.1
• /
• pp.75-88
• /
• 2005

Ameloblastoma is the most commonly occurring odontogenic tumor in oral cavity. Although most are benign epithelial neoplasm, they are generally considered to be locally aggressive and destructive, exhibiting a high rate of recurrence. The biological behavior of this neoplasm is a slowly growing, locally invasive tumor without metastasis, therefore malignant neoplasm, changed its histological appearance to carcinoma or showed distant metastasis, is only defined clinically. In this study, we identified the differentially expressed genes(DEGs) in stages under benign or malignant ameloblastoma compared with normal patient using ordered differential display(ODD) reverse transcription(RT)-PCR and $GeneFishing^{TM}$ technology. ODD RT-PCR is rather effective when the investigation of samples containing very small amounts of total RNA must be accomplished. ODD RT-PCR used the means of amplification with anchored T-primer and adaptor specific primer. bearing definite two bases at their 3' ends and so this method could display differential 3'-expressed sequence taqs(ESTs) patterns without using full-length cDNAs. Compared with standard differential display, ODD RT-PCR is more simple and have enough sensitivity to search for molecular markers by comparing gene expression profiles, However, this method required much effort and skill to perform. $GeneFishing^{TM}$ modified from DD-PCR is an improved method for detecting differentially expressed genes in two or more related samples. This two step RT-PCR method uses a constant reverse primer(anchor ACP-T) to prime the RT reaction and arbitrary primer pairs(annealing control primers, ACPs) during PCR. Because of high annealing specificity of ACPs than ODD RT-PCR, the application of $GeneFishing^{TM}$ to DEG discovery generates reproducible, authentic, and long(100bp to 2kb) PCR products that are detectable on agarose gels. Consequently, various DEGs observed differential expression levels on agarose gels were isolated from normal, benign, and malignant tissues using these methods. The expression patterns of the some isolated DEGs through ODD RT-PCR and $GeneFishing^{TM}$ were confirmed by Northern blot analysis and RT-PCR. The results showed that these identified DEGs were implicated in ameloblastoma neoplasm processes. Therefore, the identified DEGs will be further studied in order to be applied in candidate selection for marker as an early diagnosis during ameloblastoma neoplasm processes.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.205-225
• /
• 2018

Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.

• Archives of design research
• /
• v.18 no.4 s.62
• /
• pp.73-84
• /
• 2005

The purpose of this study is to analyze the interior color characteristics of elderly housing facilities in Korea. For this purpose, a filed survey was conducted between October 16 and October 25, 2003, visiting 10 elderly housing facilities. A three step analysis of the facilities was involved. First, floor, walls, ceiling, base boards and doors in common spaces were examined for the color scheme using the 'Pantone for Fashion and Home Color Guide' for each room. Measured Pantone numbers were converted to RGB color using 'Chooser 3.0 of Pantone, Inc.' Finally, the conversions were recorded as Munsell numbers. The results are as follows. First, the brightness used in elderly housing facilities ranges from high value to low value while the chroma ranges from the middle chroma to low chroma. The warm colors of Y and YR are mainly used, followed by GY and R. Second, the colors used for elderly housing facilities are found inter-related. For example, the main color in the lobby is also used for corridor and used again in dining room, library, and treatment room. Third, in terms of the color scheme, it is best help the elderly, who generally have lower recognition capability, by highlighting the distinction and attention by using the dearly distinctive accent color for the transition space, edge of the hallways or in front of the door to the units and having different color for each floor. But in most facilities except one or two, this consideration is ignored. Fourth, there is difference in color selection between elderly housing with low and high rental value. Only one facility, the most high priced one, has aggressive color arrangement, and the others show very little consideration of the elderly's needs.

• Korean Journal of Cognitive Science
• /
• v.27 no.2
• /
• pp.159-219
• /
• 2016

Mathematical ability is important for academic achievement and technological renovations in the STEM disciplines. This study concentrated on the relationship between neural basis of mathematical cognition and its mechanisms. These cognitive functions include domain specific abilities such as numerical skills and visuospatial abilities, as well as domain general abilities which include language, long term memory, and working memory capacity. Individuals can perform higher cognitive functions such as abstract thinking and reasoning based on these basic cognitive functions. The next topic covered in this study is about individual differences in mathematical abilities. Neural efficiency theory was incorporated in this study to view mathematical talent. According to the theory, a person with mathematical talent uses his or her brain more efficiently than the effortful endeavour of the average human being. Mathematically gifted students show different brain activities when compared to average students. Interhemispheric and intrahemispheric connectivities are enhanced in those students, particularly in the right brain along fronto-parietal longitudinal fasciculus. The third topic deals with growth and development in mathematical capacity. As individuals mature, practice mathematical skills, and gain knowledge, such changes are reflected in cortical activation, which include changes in the activation level, redistribution, and reorganization in the supporting cortex. Among these, reorganization can be related to neural plasticity. Neural plasticity was observed in professional mathematicians and children with mathematical learning disabilities. Last topic is about mathematical creativity viewed from Neural Darwinism. When the brain is faced with a novel problem, it needs to collect all of the necessary concepts(knowledge) from long term memory, make multitudes of connections, and test which ones have the highest probability in helping solve the unusual problem. Having followed the above brain modifying steps, once the brain finally finds the correct response to the novel problem, the final response comes as a form of inspiration. For a novice, the first step of acquisition of knowledge structure is the most important. However, as expertise increases, the latter two stages of making connections and selection become more important.

• Journal of the Korea Computer Graphics Society
• /
• v.6 no.4
• /
• pp.1-7
• /
• 2000

Traditionally, three dimension model s have been used for building virtual worlds, and a data structure called the "scene graph" is often employed to organize these 3D objects in the virtual space. On the other hand, image-based rendering has recently been suggested as a probable alternative VR platform for its photo-realism, however, due to limited interactivity. it has only been used for simple navigation systems. To combine the merits of these two approaches to object/scene representations, this paper proposes for a scene graph structure in which both 3D models and various image-based scenes/objects can be defined. traversed, and rendered together. In fact, as suggested by Shade et al. [1]. these different representations can be used as different LOD's for a given object. For in stance, an object might be rendered using a 3D model at close range, a billboard at an intermediate range. and as part of an environment map at far range. The ultimate objective of this mixed platform is to breath more interactivity into the image based rendered VE's by employing 3D models as well. There are several technical challenges in devising such a platform : designing scene graph nodes for various types of image based techniques, establishing criteria for LOD/representation selection. handling their transition s. implementing appropriate interaction schemes. and correctly rendering the overall scene. Currently, we have extended the scene graph structure of the Sense8's WorldToolKit. to accommodate new node types for environment maps. billboards, moving textures and sprites, "Tour-into-the-Picture" structure, and view interpolated objects. As for choosing the right LOD level, the usual viewing distance and image space criteria are used, however, the switching between the image and 3D model occurs at a distance from the user where the user starts to perceive the object's internal depth. Also. during interaction, regardless of the viewing distance. a 3D representation would be used, if it exists. Finally. we carried out experiments to verify the theoretical derivation of the switching rule and obtained positive results.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• Journal of Global Scholars of Marketing Science
• /
• v.18 no.4
• /
• pp.61-118
• /
• 2008

The purpose of this research is to construct research framework for international franchising based on existing literature and to identify research components in the framework. Franchise can be defined as management styles that allow franchisee use various management assets of franchisor in order to make or sell product or service. It can be divided into product distribution franchise that is designed to sell products and business format franchise that is designed for running it as business whatever its form is. International franchising can be defined as a way of internationalization of franchisor to foreign country by providing its business format or package to franchisee of host country. International franchising is growing fast for last four decades but academic research on this is quite limited. Especially in Korea, research about international franchising is carried out on by case study format with single case or empirical study format with survey based on domestic franchise theory. Therefore, this paper tries to review existing literature on international franchising research, providing research framework, and then stimulating new research on this field. International franchising research components include motives and environmental factors for decision of expanding to international franchising, entrance modes and development plan for international franchising, contracts and management strategy of international franchising, and various performance measures from different perspectives. First, motives of international franchising are fee collection from franchisee. Also it provides easier way to expanding to foreign country. The other motives including increase total sales volume, occupying better strategic position, getting quality resources, and improving efficiency. Environmental factors that facilitating international franchising encompasses economic condition, trend, and legal or political factors in host and/or home countries. In addition, control power and risk management capability of franchisor plays critical role in successful franchising contract. Final decision to enter foreign country via franchising is determined by numerous factors like history, size, growth, competitiveness, management system, bonding capability, industry characteristics of franchisor. After deciding to enter into foreign country, franchisor needs to set entrance modes of international franchising. Within contractual mode, there are master franchising and area developing franchising, licensing, direct franchising, and joint venture. Theories about entrance mode selection contain concepts of efficiency, knowledge-based approach, competence-based approach, agent theory, and governance cost. The next step after entrance decision is operation strategy. Operation strategy starts with selecting a target city and a target country for franchising. In order to finding, screening targets, franchisor needs to collect information about candidates. Critical information includes brand patent, commercial laws, regulations, market conditions, country risk, and industry analysis. After selecting a target city in target country, franchisor needs to select franchisee, in other word, partner. The first important criteria for selecting partners are financial credibility and capability, possession of real estate. And cultural similarity and knowledge about franchisor and/or home country are also recognized as critical criteria. The most important element in operating strategy is legal document between franchisor and franchisee with home and host countries. Terms and conditions in legal documents give objective information about characteristics of franchising agreement for academic research. Legal documents have definitions of terminology, territory and exclusivity, agreement of term, initial fee, continuing fees, clearing currency, and rights about sub-franchising. Also, legal documents could have terms about softer elements like training program and operation manual. And harder elements like law competent court and terms of expiration. Next element in operating strategy is about product and service. Especially for business format franchising, product/service deliverable, benefit communicators, system identifiers (architectural features), and format facilitators are listed for product/service strategic elements. Another important decision on product/service is standardization vs. customization. The rationale behind standardization is cost reduction, efficiency, consistency, image congruence, brand awareness, and competitiveness on price. Also standardization enables large scale R&D and innovative change in management style. Another element in operating strategy is control management. The simple way to control franchise contract is relying on legal terms, contractual control system. There are other control systems, administrative control system and ethical control system. Contractual control system is a coercive source of power, but franchisor usually doesn't want to use legal power since it doesn't help to build up positive relationship. Instead, self-regulation is widely used. Administrative control system uses control mechanism from ordinary work relationship. Its main component is supporting activities to franchisee and communication method. For example, franchisor provides advertising, training, manual, and delivery, then franchisee follows franchisor's direction. Another component is building franchisor's brand power. The last research element is performance factor of international franchising. Performance elements can be divided into franchisor's performance and franchisee's performance. The conceptual performance measures of franchisor are simple but not easy to obtain objectively. They are profit, sale, cost, experience, and brand power. The performance measures of franchisee are mostly about benefits of host country. They contain small business development, promotion of employment, introduction of new business model, and level up technology status. There are indirect benefits, like increase of tax, refinement of corporate citizenship, regional economic clustering, and improvement of international balance. In addition to those, host country gets socio-cultural change other than economic effects. It includes demographic change, social trend, customer value change, social communication, and social globalization. Sometimes it is called as westernization or McDonaldization of society. In addition, the paper reviews on theories that have been frequently applied to international franchising research, such as agent theory, resource-based view, transaction cost theory, organizational learning theory, and international expansion theories. Resource based theory is used in strategic decision based on resources, like decision about entrance and cooperation depending on resources of franchisee and franchisor. Transaction cost theory can be applied in determination of mutual trust or satisfaction of franchising players. Agent theory tries to explain strategic decision for reducing problem caused by utilizing agent, for example research on control system in franchising agreements. Organizational Learning theory is relatively new in franchising research. It assumes organization tries to maximize performance and learning of organization. In addition, Internalization theory advocates strategic decision of direct investment for removing inefficiency of market transaction and is applied in research on terms of contract. And oligopolistic competition theory is used to explain various entry modes for international expansion. Competency theory support strategic decision of utilizing key competitive advantage. Furthermore, research methodologies including qualitative and quantitative methodologies are suggested for more rigorous international franchising research. Quantitative research needs more real data other than survey data which is usually respondent's judgment. In order to verify theory more rigorously, research based on real data is essential. However, real quantitative data is quite hard to get. The qualitative research other than single case study is also highly recommended. Since international franchising has limited number of applications, scientific research based on grounded theory and ethnography study can be used. Scientific case study is differentiated with single case study on its data collection method and analysis method. The key concept is triangulation in measurement, logical coding and comparison. Finally, it provides overall research direction for international franchising after summarizing research trend in Korea. International franchising research in Korea has two different types, one is for studying Korean franchisor going overseas and the other is for Korean franchisee of foreign franchisor. Among research on Korean franchisor, two common patterns are observed. First of all, they usually deal with success story of one franchisor. The other common pattern is that they focus on same industry and country. Therefore, international franchise research needs to extend their focus to broader subjects with scientific research methodology as well as development of new theory.