• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.31-40
• /
• 2021

The increasing number of credit card fraud cases has become a considerable problem since the past decades. This phenomenon is due to the expansion of new technologies, including the increased popularity and volume of online banking transactions and e-commerce. In order to address the problem of credit card fraud detection, a rule-based approach has been widely utilized to detect and guard against fraudulent activities. However, it requires huge computational power and high complexity in defining and building the rule base for pattern matching, in order to precisely identifying the fraud patterns. In addition, it does not come with intelligence and ability in predicting or analysing transaction data in looking for new fraud patterns and strategies. As such, Data Mining and Machine Learning algorithms are proposed to overcome the shortcomings in this paper. The aim of this paper is to highlight the important techniques and methodologies that are employed in fraud detection, while at the same time focusing on the existing literature. Methods such as Artificial Neural Networks (ANNs), Support Vector Machines (SVMs), naïve Bayesian, k-Nearest Neighbour (k-NN), Decision Tree and Frequent Pattern Mining algorithms are reviewed and evaluated for their performance in detecting fraudulent transaction.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.19 no.11
• /
• pp.94-101
• /
• 2020

In the age of industry 4.0, artificial intelligence is being widely used to realize machinery condition monitoring. Due to their excellent performance and the ability to handle large volumes of data, machine learning techniques have been applied to realize the fault diagnosis of different equipment. In this study, we performed the failure mode effect analysis (FMEA) of an aluminum electrolytic capacitor by using deep learning and big data. Several tests were performed to identify the main failure mode of the aluminum electrolytic capacitor, and it was noted that the capacitance reduced significantly over time due to overheating. To reflect the capacitance degradation behavior over time, we employed the Vanilla long short-term memory (LSTM) neural network architecture. The LSTM neural network has been demonstrated to achieve excellent long-term predictions. The prediction results and metrics of the LSTM and Vanilla LSTM models were examined and compared. The Vanilla LSTM outperformed the conventional LSTM in terms of the computational resources and time required to predict the capacitance degradation.

• Food Science and Industry
• /
• v.49 no.4
• /
• pp.19-30
• /
• 2016

Food industry in Korea is one of the most important manufacturing field since the history of this country. Recent days all industries in the world move to $4^{th}$ industrial revolution beginning from 1st revolution. This means that connections between human to human, human to things and things to things should be settled down. food industry in this country should escape from the conventional manufacturing fields until now and accept new or cutting edge technology NT including artificial intelligence robot system and platform system using Internet of Thing. To overcome the saturation condition of domestic food market, it should be extended our market to overseas. To do this Korean food industry should be reformed the processing system to convergence and fusion inner or multi-disciplinary research in not only research field but also manufacturing field. The food industry must introduce new technology and concept of controlling all manufacturing systems. This paper present the fields should be convergence and the field study together and the new techniques, methods and new products be developed in the future.

• Food Science and Industry
• /
• v.50 no.2
• /
• pp.60-73
• /
• 2017

Recently, the whole world is facing an unprecedented moment of opportunity, so-called The Fourth Industrial Revolution. As emphasized in the World Economic Forum held in January of 2016 at Davos, the Fourth Industrial Revolution is not merely a changes of technological devices. The fundamental of the revolution is new, innovative, and visionary business models which change the whole systems dramatically. One of the greatest challenges is to feed an expected population of 9 billion by 2050 in a impactful way. The system should be sustainable as well as beneficial in improving the lives of people in the food chain along with the ecological health of environment. The technological advances of the Fourth Industrial Revolution are expected to improve our food system. The smart farm technology such as precision planting and irrigation techniques will improve the yields of food materials. The smart food transportation and logistics systems will substantially improve the safety and human nutrition. The adaptation the Fourth Industrial Revolution technology will induce the smart supply chains, smart production, and smart products in food industry due to its flexibility and standardization. This will lead the manufactures to adapt to customers' changing product specifications and traceable services in a timely manner.

• Electronics and Telecommunications Trends
• /
• v.36 no.3
• /
• pp.12-22
• /
• 2021

Since the technical realization of self-aligned planar complementary metal-oxide-semiconductor field-effect transistors in 1960s, semiconductor manufacturing has aggressively pursued scaling that fruitfully resulted in tremendous advancement in device performances and realization of features sizes smaller than 10 nm. Due to many intrinsic material and technical obstacles, continuing the scaling progress of semiconductor devices has become increasingly arduous. As an effort to circumvent the areal limit, stacking devices in a three-dimensional fashion has been suggested. This approach is commonly called monolithic three-dimensional (M3D) integration. In this work, we examined technical issues that need to be addressed and overcome to fully realize energy efficiency, short latency and cost competency. Full-fledged M3D technologies are expected to contribute to various new fields of artificial intelligence, autonomous gadgets and unknowns, which are to be discovered.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.107-114
• /
• 2021

Recently artificial intelligence technology has been introduced in various fields and various machine learning models have been operated in various frameworks as academic interest has increased. However, these frameworks have different data formats, which lack interoperability, and to overcome this, the open neural network exchange format, ONNX, has been proposed. In this paper we describe how to transform multiple machine learning models to ONNX, and propose algorithms and inference systems that can determine machine learning techniques in an integrated ONNX format. Furthermore we compare the inference results of the models before and after the ONNX transformation, showing that there is no loss or performance degradation of the learning results between the ONNX transformation.

• Journal of Engineering Education Research
• /
• v.24 no.3
• /
• pp.50-57
• /
• 2021

Although the mathematics education process in AI education is a very important issue, little cases are reported in developing effective methods on AI and mathematics education at the university level. The universities cover all fields of mathematics in their curriculums, but they lack in connecting and applying the math knowledge to AI in an efficient manner. Students are hardly interested in taking many math courses and it gets worse for the students in humanities, social sciences and arts. But university education is very slow in adapting to rapidly changing new technologies in the real world. AI is a technology that is changing the paradigm of the century, so every one should be familiar with this technology but it requires fundamental math knowledge. It is not fair for the students to study all math subjects and ride on the AI train. We recognize that three key elements, SW knowledge, mathematical knowledge, and domain knowledge, are required in applying AI technology to the real world problems. This study proposes a modular approach of studying mathematics knowledge while connecting the math to different domain problems using AI techniques. We also show a modular curriculum that is developed for using math for AI-driven autonomous driving.

• International journal of advanced smart convergence
• /
• v.10 no.1
• /
• pp.75-81
• /
• 2021

As the number of connected smart devices and applications increases explosively, the existing orthogonal multiple access (OMA) techniques have become insufficient to accommodate mobile traffic, such as artificial intelligence (AI) and the internet of things (IoT). Fortunately, non-orthogonal multiple access (NOMA) in the fifth generation (5G) mobile networks has been regarded as a promising solution, owing to increased spectral efficiency and massive connectivity. In this paper, we investigate the achievable data rate for non-orthogonal multiple access (NOMA) with negatively-correlated information sources (CIS). For this, based on the linear transformation of independent random variables (RV), we derive the closed-form expressions for the achievable data rates of NOMA with negatively-CIS. Then it is shown that the achievable data rate of the negatively-CIS NOMA increases for the stronger channel user, whereas the achievable data rate of the negatively-CIS NOMA decreases for the weaker channel user, compared to that of the positively-CIS NOMA for the stronger or weaker channel users, respectively. We also show that the sum rate of the negatively-CIS NOMA is larger than that of the positively-CIS NOMA. As a result, the negatively-CIS could be more efficient than the positively-CIS, when we transmit CIS over 5G NOMA networks.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.26 no.1
• /
• pp.1-8
• /
• 2021

For the safe and reliable operation of lithium-ion batteries in electric vehicles or energy storage systems, having accurate information of the battery, such as the state of charge (SOC), is essential. Many different techniques of battery SOC estimation have been developed, such as the Kalman filter. However, when this filter is applied to multiple batteries, it has difficulty maintaining the accuracy of the estimation over all cells owing to the difference in parameter values of each cell. The difference in the parameter of each cell may increase as the operation time accumulates due to aging. In this paper, a novel deep neural network (DNN)-based SOC estimation method for multi-cell application is proposed. In the proposed method, DNN is implemented to determine the nonlinear relationships of the voltage and current at different SOCs and temperatures. In the training, the voltage and current data obtained at different temperatures during charge/discharge cycles are used. After the comprehensive training with the data obtained from the cycle test with a cell, the resulting algorithm is applied to estimate the SOC of other cells. Experimental results show that the mean absolute error of the estimation is 1.213% at 25℃ with the proposed DNN-based SOC estimation method.

• International Journal of High-Rise Buildings
• /
• v.9 no.4
• /
• pp.351-360
• /
• 2020

Monitoring civil structures periodically is necessary for ensuring the fitness of the structures. Cracks on inner and outer surfaces of the building plays a vital role in indicating the health of the building. Conventionally, human visual inspection techniques were carried up to human reachable altitudes. Monitoring of high rise infrastructures cannot be done using this primitive method. Also, there is a necessity for more accurate prediction of cracks on building surfaces for ensuring the health and safety of the building. The proposed research focused on developing an efficient crack classification model using Transfer Learning enabled EfficientNet (TL-EN) architecture. Though many other pre-trained models were available for crack classification, they rely on more number of training parameters for better accuracy. The TL-EN model attained an accuracy of 0.99 with less number of parameters on large dataset. A bench marked METU dataset with 40000 images were used to test and validate the proposed model. The surfaces of high rise buildings were investigated using vision enabled Unmanned Arial Vehicles (UAV). These UAV is fabricated with TL-EN model schema for capturing and analyzing the real time streaming video of building surfaces.