• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.12
• /
• pp.485-496
• /
• 2018

So far, the deep learning, a field of artificial intelligence, has achieved remarkable results in solving problems from unstructured data. However, it is difficult to comprehensively judge situations like humans, and did not reach the level of intelligence that deduced their relations and predicted the next situation. Recently, deep neural networks show that artificial intelligence can possess powerful relational reasoning that is core intellectual ability of human being. In this paper, to analyze and observe the performance of Relation Networks (RN) among the neural networks for relational reasoning, two types of RN-based deep neural network models were constructed and compared with the baseline model. One is a visual question answering RN model using Sort-of-CLEVR and the other is a text-based question answering RN model using bAbI task. In order to maximize the performance of the RN-based model, various performance improvement experiments such as hyper parameters tuning have been proposed and performed. The effectiveness of the proposed performance improvement methods has been verified by applying to the visual QA RN model and the text-based QA RN model, and the new domain model using the dialogue-based LL dataset. As a result of the various experiments, it is found that the initial learning rate is a key factor in determining the performance of the model in both types of RN models. We have observed that the optimal initial learning rate setting found by the proposed random search method can improve the performance of the model up to 99.8%.

• Korean Journal of Artificial Intelligence
• /
• v.7 no.1
• /
• pp.13-16
• /
• 2019

While the Chinese banks have started the impact of foreign banks. At the same time, rising pressure on foreign exchange reserves and appreciation of the renminbi has prompted Chinese banks to go abroad and diversify their risks. The financial crisis of 2008 has caused the continued turbulence of the major financial markets around the world, and the valuation of foreign financial institutions has been drastically shrinking, providing opportunities for Chinese banks to carry out overseas M&A. Based on the overseas M&A status of Chinese commercial banks, this paper sums up the characteristics of the overseas M&A. Then taking a series of overseas M&A conducted by ICBC from 2006 to 2011 as an example, it analyzes the relationship between M&A and performance growth using grey incidence model. The test shows: there is a positive correlation between both overseas M&A and interest rate differential with performance growth of ICBC, and overseas M&A transactions role in promoting the performance growth is significantly higher than the interest rate differential.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.10
• /
• pp.1335-1343
• /
• 1999

This paper presents concrete pattern recognition method to identify the strength of concrete by evidence accumulation with multiple parameters based on artificial intelligence techniques. At first, zero-crossing(ZCR), mean frequency(MEANF), median frequency(MEDF) and autoregressive model coefficient(ARC) are extracted as feature parameters from ultrasonic signal of concrete. Pattern recognition is carried out through the evidence accumulation procedure using distance measured with reference parameters. A fuzzy mapping function is designed to transform the distances for the application of the evidence accumulation method. Results are presented to support the feasibility of the suggested approach for concrete pattern recognition.

• Korean Journal of Artificial Intelligence
• /
• v.2 no.1
• /
• pp.1-3
• /
• 2014

These days, interest and demand on smart learning has rapidly increased. Video English and mobile system based English speaking service have become popular. This study gave prototype of application to give and take voice message with world people and to give new concept of voice pen-pal beyond exchange of text messages. In modern society having rapidly increasing demand on smart learning, you can study foreign language by smart phone and communicate with foreigners by voice anytime and anywhere. The app allows global exchange to learn conversation. Recruitment of initial users and profit model have problems. We shall develop to improve problems and to solve difficulty.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.12
• /
• pp.671-678
• /
• 2003

This paper proposes an expert system to generate automatically the test table of test system which can highly enhance the quality and productivity of product by inspecting quickly and accurately the defect device on the electronic circuit board tested. The expert system identifies accurately the tested components and the circuit patterns by tracing automatically the connectivity of circuit from electronic circuit database. And it generates automatically the test table to detect accurately the missing components, the misplaced components, and the wrong components for analog components such as resistance, coil, condenser, diode, and transistor, based on the experience knowledge of veteran expert. It is implemented in C computer language for the purpose of the implementation of the inference engine using the dynamic memory allocation technique, the interface with the electronic circuit database and the hardware direct control. And, the validity of the builded expert system is proved by simulating for a typical electronic board model.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.311-318
• /
• 2022

Tobacco sales enterprises often need to summarize and verify the daily sales bills, which may consume substantial manpower, and manual verification is prone to occasional errors. The use of artificial intelligence technology to realize the automatic identification and verification of such bills offers important practical significance. This study presents a novel multi-branch residual network for tobacco sales bills to improve the efficiency and accuracy of tobacco sales. First, geometric correction and edge alignment were performed on the input sales bill image. Second, the multi-branch residual network recognition model is established and trained using the preprocessed data. The comparative experimental results demonstrated that the correct recognition rate of the proposed method reached 98.84% on the China Tobacco Bill Image dataset, which is superior to that of most existing recognition methods.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.588-591
• /
• 2021

4차 산업혁명 시대가 도래하면서 세상이 빠른 속도로 변하고 있다. 특히 데이터·인공지능(AI, Artificial Intelligence)의 활용이 적극적으로 다양한 분야에서 적용되기 시작하고 있다. 하지만 산림수종을 분석하는 업무를 수행하는 과정은 수작업으로 진행하다 보니 오류가 다수 발생하고 있다. 따라서 본 논문에서는 수도권 항공사진을 이용하여 소나무, 낙엽송, 침엽수, 활엽수를 대상으로 자동으로 분석하는 AI 학습용 데이터 약 60,000장을 구축하고, 수종을 구분할 수 있는 AI 모델을 개발하였다. 이를 통해 산림변화탐지 및 산림 분야 주제도 제작 시 수종 분할 이미지를 기초자료로 활용함으로써 업무효율 증대를 기대할 수 있다.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.694-696
• /
• 2021

본 논문에서는 CNN 모델에서 학습에 사용할 이미지 데이터를 늘리기 위해 사용되는 Mixup 알고리즘을 MLP 모델에 사용하는 데이터셋에 적용하여 data augmentation 효과를 얻을 수 있는 지에 대한 테스트를 수행했다. 테스트 결과 MLP 모델에 사용할 데이터셋에도 Mixup 알고리즘으로 data augmentation 효과를 기대할 수 있음을 보여준다.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2023.11a
• /
• pp.97-98
• /
• 2023

Recently, the use of digital technology in architectural technology is gradually increasing with the development of various industrial technologies. There are artificial intelligence and drones in the field of architecture, and among them, deep learning technology has been introduced to conduct research in areas such as precise inspection of buildings, and it is expressed in a highly reliable way. When a building is deteriorated, various defects such as cracks in the surface and subsidence of the structure may occur. Since these cracks can represent serious structural damage in the future, the detection of cracks was conducted using artificial intelligence that can detect and identify surface defects by detecting cracks and aging of buildings.

• International journal of advanced smart convergence
• /
• v.13 no.3
• /
• pp.135-142
• /
• 2024

Recently, along with digital transformation, technologies such as cloud computing, big data, and artificial intelligence have been actively introduced. In a situation where these technological changes are progressing rapidly, it is often difficult to manage processes efficiently using existing simple workflow management methods. Companies providing current cloud services are adopting virtualization technologies, including virtual machines (VMs) and containers, in their distributed system infrastructure for automated application deployment. Accordingly, this paper proposes a process-based orchestration system for integrated execution of corporate process-oriented workloads by integrating the potential of big data and machine learning technologies. This system consists of four layers as components for performing workload processes. Additionally, a common information model is applied to the data to efficiently integrate and manage the various formats and uses of data generated during the process creation stage. Moreover, a standard metadata protocol is introduced to ensure smooth exchange between data. This proposed system utilizes various types of data storage to store process data, metadata, and analysis models. This enables flexible management and efficient processing of data.