• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2018.06a
• /
• pp.102-102
• /
• 2018

Electronic industry had required the finer size and the higher performance of the device. Therefore, 3-D die stacking technology such as TSV (through silicon via) and micro-bump had been used. Moreover, by the development of the 3-D die stacking technology, 3-D structure such as chip to chip (c2c) and chip to wafer (c2w) had become practicable. These technologies led to the appearance of HBM (high bandwidth memory). HBM was type of the memory, which is composed of several stacked layers of the memory chips. Each memory chips were connected by TSV and micro-bump. Thus, HBM had lower RC delay and higher performance of data processing than the conventional memory. Moreover, due to the development of the IT industry such as, AI (artificial intelligence), IOT (internet of things), and VR (virtual reality), the lower pitch size and the higher density were required to micro-electronics. Particularly, to obtain the fine pitch, some of the method such as copper pillar, nickel diffusion barrier, and tin-silver or tin-silver-copper based bump had been utillized. TCB (thermal compression bonding) and reflow process (thermal aging) were conventional method to bond between tin-silver or tin-silver-copper caps in the temperature range of 200 to 300 degrees. However, because of tin overflow which caused by higher operating temperature than melting point of Tin ($232^{\circ}C$), there would be the danger of bump bridge failure in fine-pitch bonding. Furthermore, regulating the phase of IMC (intermetallic compound) which was located between nickel diffusion barrier and bump, had a lot of problems. For example, an excess of kirkendall void which provides site of brittle fracture occurs at IMC layer after reflow process. The essential solution to reduce the difficulty of bump bonding process is copper to copper direct bonding below $300^{\circ}C$. In this study, in order to improve the problem of bump bonding process, copper to copper direct bonding was performed below $300^{\circ}C$. The driving force of bonding was the self-annealing properties of electrodeposited Cu with high defect density. The self-annealing property originated in high defect density and non-equilibrium grain boundaries at the triple junction. The electrodeposited Cu at high current density and low bath temperature was fabricated by electroplating on copper deposited silicon wafer. The copper-copper bonding experiments was conducted using thermal pressing machine. The condition of investigation such as thermal parameter and pressure parameter were varied to acquire proper bonded specimens. The bonded interface was characterized by SEM (scanning electron microscope) and OM (optical microscope). The density of grain boundary and defects were examined by TEM (transmission electron microscopy).

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.169-176
• /
• 2020

Recently, deep neural networks (DNNs) have brought revolutions to many mobile and embedded devices by providing human-level machine intelligence for various applications. However, high inference accuracy of such DNNs comes at high computational costs, and, hence, there have been significant efforts to reduce computational overheads of DNNs either by compressing off-the-shelf models or by designing a new small footprint DNN architecture tailored to resource constrained devices. One notable recent paradigm in designing small footprint DNN models is sharing parameters in several layers. However, in previous approaches, the parameter-sharing techniques have been applied to large deep networks, such as ResNet, that are known to have high redundancy. In this paper, we propose a parameter-sharing method for already parameter-efficient small networks such as ShuffleNetV2. In our approach, small templates are combined with small layer-specific parameters to generate weights. Our experiment results on ImageNet and CIFAR100 datasets show that our approach can reduce the size of parameters by 15%-35% of ShuffleNetV2 while achieving smaller drops in accuracies compared to previous parameter-sharing and pruning approaches. We further show that the proposed approach is efficient in terms of latency and energy consumption on modern embedded devices.

• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.173-193
• /
• 2010

The World Wide Web is transitioning from being a mere collection of documents that contain useful information toward providing a collection of services that perform useful tasks. The emerging Web service technology has been envisioned as the next technological wave and is expected to play an important role in this recent transformation of the Web. By providing interoperable interface standards for application-to-application communication, Web services can be combined with component-based software development to promote application interaction and integration within and across enterprises. To make Web services for service-oriented computing operational, it is important that Web services repositories not only be well-structured but also provide efficient tools for an environment supporting reusable software components for both service providers and consumers. As the potential of Web services for service-oriented computing is becoming widely recognized, the demand for an integrated framework that facilitates service discovery and publishing is concomitantly growing. In our research, we propose a framework that facilitates Web service discovery and publishing by combining clustering techniques and leveraging the semantics of the XML-based service specification in WSDL files. We believe that this is one of the first attempts at applying unsupervised artificial neural network-based machine-learning techniques in the Web service domain. We have developed a Web service discovery tool based on the proposed approach using an unsupervised artificial neural network and empirically evaluated the proposed approach and tool using real Web service descriptions drawn from operational Web services repositories. We believe that both service providers and consumers in a service-oriented computing environment can benefit from our Web service discovery approach.

• Journal of Intelligence and Information Systems
• /
• v.11 no.2
• /
• pp.59-79
• /
• 2005

Export containers in a container terminal are usually classified into a few weight groups and those belonging to the same group are placed together on a same stack. The reason for this stacking by weight groups is that it becomes easy to have the heavier containers be loaded onto a ship before the lighter ones, which is important for the balancing of the ship. However, since the weight information available at the time of container arrival is only an estimate, those belonging to different weight groups are often stored together on a same stack. This becomes the cause of extra moves, or rehandlings, of containers at the time of loading to fetch out the heavier containers placed under the lighter ones. In this paper, we use machine learning techniques to derive a classifier that can classify the containers into the weight groups with improved accuracy. We also show that a more useful classifier can be derived by applying a cost-sensitive learning technique, for which we introduce a scheme of searching for a good cost matrix. Simulation experiments have shown that our proposed method can reduce about 5$\sim$7% of rehandlings when compared to the traditional weight grouping method.

• Journal of Science Education
• /
• v.45 no.2
• /
• pp.156-171
• /
• 2021

This study aims at investigating the characteristics of trends of future education over time though the literature review and examining the accuracy of the framework for forecasting future education proposed by the previous studies by comparing the outcomes between the literature review and media articles. Thus, this study collects the articles dealing with future education searched from the Web of Science and categorized them into four periods during the new millennium. The new articles from media were selected to find out the present of education so that we can figure out the appropriateness of the proposed framework to predict the future of education. Research findings reveal that gradual tendencies of topics could not be found except teacher education and they are diverse from characteristics of agents (students and teachers) to the curriculum and pedagogical strategies. On the other hand, the results of analysis on the media articles focuses more on the projects launched by the government and the immediate responses to the COVID-19, as well as educational technologies related to big data and artificial intelligence. It is surprising that only a few key words are occupied in the latest articles from the literature review and many of them have not been discussed before. This indicates that the predictive framework is not effective to establish the long-term plan for education due to the uncertainty of educational environment, and thus this study will give some implications for developing the model to forecast the future of education.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.4
• /
• pp.44-57
• /
• 2019

Various studies have been conducted to solve traffic congestions in many metropolitan cities through accurate traffic flow prediction. Most studies are based on the assumption that past traffic patterns repeat in the future. Models based on such an assumption fall short in case irregular traffic patterns abruptly occur. Instead, the approaches such as predicting traffic pattern through big data analytics and artificial intelligence have emerged. Specifically, deep learning algorithms such as RNN have been prevalent for tackling the problems of predicting temporal traffic flow as a time series. However, these algorithms do not perform well in terms of long-term prediction. In this paper, we take into account various external factors that may affect the traffic flows. We model the correlation between the multi-dimensional context information with temporal traffic speed pattern using deep neural networks. Our model trained with the traffic data from TOPIS system by Seoul, Korea can predict traffic speed on a specific date with the accuracy reaching nearly 90%. We expect that the accuracy can be improved further by taking into account additional factors such as accidents and constructions for the prediction.

• The Journal of the Korea Contents Association
• /
• v.19 no.7
• /
• pp.384-392
• /
• 2019

Big data-related research that deals with the amount of explosive information in the era of the Fourth Industrial Revolution is actively underway. Big data is an essential element that promotes the development of artificial intelligence with a wide range of data that become learning data for machine learning, or deep learning. The use of deep learning and big data in various fields has produced meaningful results. In this paper, we have investigated the use of Big Data in the cultural arts industry, focusing on video contents. Noteworthy is that big data is used not only in the distribution of cultural and artistic contents but also in the production stage. In particular, we first looked at what kind of achievements and changes the Netflix in the US brought to the OTT business, and analyzed the current state of the OTT business in Korea. After that, Netflix analyzed the success stories of 'House of Cards', which was produced / circulated through 'Deep Learning' cinematique, which is a prediction algorithm, through accumulated customer data. After that, FGI (Focus Group Interview) was held for cultural and artistic contents experts. In this way, the future prospects of Big Data in the domestic culture and arts industry are divided into technical aspect, creative aspect, and ethical aspect.

• Journal of Internet Computing and Services
• /
• v.23 no.1
• /
• pp.77-86
• /
• 2022

The Republic of Korea is a country in which coastal surveillance is an imperative national task as it is surrounded by seas on three sides under the confrontation between South and North Korea. However, due to Defense Reform 2.0, the number of R/D (Radar) operating personnel has decreased, and the period of service has also been shortened. Moreover, there is always a possibility that a human error will occur. This paper presents specific guidelines for developing an AI learning model for the intelligent coastal surveillance system. We present a three-step strategy to realize the guidelines. The first stage is a typical stage of building an AI learning model, including data collection, storage, filtering, purification, and data transformation. In the second stage, R/D signal analysis is first performed. Subsequently, AI learning model development for classifying real and false images, coastal area analysis, and vulnerable area/time analysis are performed. In the final stage, validation, visualization, and demonstration of the AI learning model are performed. Through this research, the first achievement of making the existing weapon system intelligent by applying the application of AI technology was achieved.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.7
• /
• pp.1-7
• /
• 2022

Recently, as the use of applications such as big data programs and machine learning programs that are driven while generating large amounts of data in the program itself becomes common, the existing main memory alone lacks memory, making it difficult to execute the program quickly. In particular, the need to derive results more quickly has emerged in a situation where it is necessary to analyze whether the entire sequence is genetically altered due to the outbreak of the coronavirus. As a result of measuring performance by applying large-capacity data to a computing system equipped with a self-developed memory pool MOCA host adapter instead of processing large-capacity data from an existing SSD, performance improved by 16% compared to the existing SSD system. In addition, in various other benchmark tests, IO performance was 92.8%, 80.6%, and 32.8% faster than SSD in computing systems equipped with memory pool MOCA host adapters such as SortSampleBam, ApplyBQSR, and GatherBamFiles by task of workflow. When analyzing large amounts of data, such as electrical dielectric pipeline analysis, it is judged that the measurement delay occurring at runtime can be reduced in the computing system equipped with the memory pool MOCA host adapter developed in this research.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.2
• /
• pp.71-82
• /
• 2022

In order to replace the sampling tensile test of products produced in the tab welding process, which is one of the automotive battery manufacturing processes, vision inspectors are currently being developed and used. However, the vision inspection has the problem of inspection position error and the cost of improving it. In order to solve these problems, there are recent cases of applying deep learning technology. As one such case, this paper tries to examine the usefulness of applying Faster R-CNN, one of the deep learning technologies, to existing product inspection. The images acquired through the existing vision inspection machine are used as training data and trained using the Faster R-CNN ResNet101 V1 1024x1024 model. The results of the conventional vision test and Faster R-CNN test are compared and analyzed based on the test standards of 0% non-detection and 10% over-detection. The non-detection rate is 34.5% in the conventional vision test and 0% in the Faster R-CNN test. The over-detection rate is 100% in the conventional vision test and 6.9% in Faster R-CNN. From these results, it is confirmed that deep learning technology is very useful for detecting welding error of lead tabs in automobile batteries.