• Journal of KIISE:Software and Applications
• /
• v.29 no.12
• /
• pp.966-978
• /
• 2002

There are two important problems in improving text classification systems based on machine learning approach. The first one, called "selection problem", is how to select a minimum number of informative documents from a given document collection. The second one, called "composition problem", is how to reorganize selected training documents so that they can fit an adopted learning method. The former problem is addressed in "active learning" algorithms, and the latter is discussed in "boosting" algorithms. This paper proposes a new learning method, called AdaBUS, which proactively solves the above problems in the context of Naive Bayes classification systems. The proposed method constructs more accurate classification hypothesis by increasing the valiance in "weak" hypotheses that determine the final classification hypothesis. Consequently, the proposed algorithm yields perturbation effect makes the boosting algorithm work properly. Through the empirical experiment using the Routers-21578 document collection, we show that the AdaBUS algorithm more significantly improves the Naive Bayes-based classification system than other conventional learning methodson system than other conventional learning methods

• Journal of KIISE:Computer Systems and Theory
• /
• v.37 no.4
• /
• pp.226-238
• /
• 2010

OpenSSL is an open-source library implementing SSL that is a secure communication protocol. However, the library has a severe vulnerability that its security information can be easily exposed to malicious software when the library is used in a form of shared library on Linux and UNIX operating systems. We propose a scheme to attack the vulnerability of the OpenSSL library. The scheme injects codes into a running client program to execute the following attacks on the vulnerability in a SSL handshake. First, when a client sends a server a list of cryptographic algorithms that the client is willing to support, our scheme replaces all algorithms in the list with a specific algorithm. Such a replacement causes the server to select the specific algorithm. Second, the scheme steals a key for data encryption and decryption when the key is generated. Then the key is sent to an outside attacker. After that, the outside attacker decrypts encrypted data that has been transmitted between the client and the server, using the specified algorithm and the key. To show that our scheme is realizable, we perform an experiment of collecting encrypted login data that an ftp client using the OpenSSL shared library sends its server and then decrypting the login data.

• Journal of KIISE:Software and Applications
• /
• v.32 no.12
• /
• pp.1217-1228
• /
• 2005

Recently, since DNA computing has been widely studied for various applications, DNA sequence design which is the most basic and important step for DNA computing has been highlighted. In previous works, DNA sequence design has been formulated as a multi-objective optimization task, and solved by elitist non-dominated sorting genetic algorithm (NSGA-II). However, NSGA-II needed lots of computational time. Therefore, we use an $\varepsilon$- multiobjective evolutionarv algorithm ($\varepsilon$-MOEA) to overcome the drawbacks of NSGA-II in this paper. To compare the performance of two algorithms in detail, we apply both algorithms to the DTLZ2 benchmark function. $\varepsilon$-MOEA outperformed NSGA-II in both convergence and diversity, $70\%$ and $73\%$ respectively. Especially, $\varepsilon$-MOEA finds optimal solutions using small computational time. Based on these results, we redesign the DNA sequences generated by the previous DNA sequence design tools and the DNA sequences for the 7-travelling salesman problem (TSP). The experimental results show that $\varepsilon$-MOEA outperforms the most cases. Especially, for 7-TSP, $\varepsilon$-MOEA achieves the comparative results two tines faster while finding $22\%$ improved diversity and $92\%$ improved convergence in final solutions using the same time.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.10
• /
• pp.1264-1274
• /
• 1999

본 논문에서는 분산된 클록들을 주기적으로 동기화 시키는 분산 실시간 시스템에서 시간적 제약을 만족시키기 위한 정적/동적 시간 제약(timing constraint) 변환 기법을 제안한다. 전형적인 이산클록동기화(discrete clock synchronization) 알고리즘은 클록의 값을 순간적으로 조정하여 클록의 시간이 불연속적으로 진행한다. 이러한 시간상의 불연속성은 시간적 이벤트를 잃어버리거나 다시 발생시키는 오류를 범하게 한다.클록 시간의 불연속성을 피하기 위해 일반적으로 연속클록동기화(continuous clock synchronization) 기법이 제안되고 있지만 소프트웨어적으로 구현되면 많은 오버헤드를 유발시키는 문제점이 있다. 본 논문에서는 시간적 제약을 동적으로 변환시키는 DCT (Dynamic Constraint Transformation) 기법을 제안하였으며, 이를 통해 기존의 이산클록동기화 알고리즘을 수정하지 않고서도 클록 시간의 불연속성에 의한 문제점들을 해결할 수 있도록 하였다. 아울러 DCT에 의해 이산클록동기화 하에서 생성된 태스크 스케쥴이 연속클록동기화에 의해 생성된 스케쥴과 동일함을 증명하여 DCT의 동작이 이론적으로 정확함을 증명하였다.또한 분산 실시간 시스템에서 지역 클록(local clock)이 기준 클록과 완벽하게 일치하지 않아서 발생하는 스케쥴링상의 문제점을 다루었다. 이를 위해 먼저 두 가지의 스케쥴링 가능성, 지역적 스케쥴링 가능성(local schedulability)과 전역적 스케쥴링 가능성(global schedulability)을 정의하고, 이를 위해 시간적 제약을 정적으로 변환시키는 SCT (Static Constraint Transformation) 기법을 제안하였다. SCT를 통해 지역적으로 스케쥴링 가능한 태스크는 전역적으로 스케쥴링이 가능하므로, 단지 지역적 스케쥴링 가능성만을 검사하면 스케쥴링 문제를 해결할 수 있도록 하였고 이를 수학적으로 증명하였다.Abstract In this paper, we present static and dynamic constraint transformation techniques for ensuring timing requirements in a distributed real-time system possessing periodically synchronized distributed local clocks. Traditional discrete clock synchronization algorithms that adjust local clocks instantaneously yield time discontinuities. Such time discontinuities lead to the loss or the gain of events, thus raising serious run-time faults.While continuous clock synchronization is generally suggested to avoid the time discontinuity problem, it incurs too much run-time overhead to be implemented in software. We propose a dynamic constraint transformation (DCT) technique which can solve the problem without modifying discrete clock synchronization algorithms. We formally prove the correctness of the DCT by showing that the DCT with discrete clock synchronization generates the same task schedule as the continuous clock synchronization.We also investigate schedulability problems that arise when imperfect local clocks are used in distributed real-time systems. We first define two notions of schedulability, global schedulability and local schedulability, and then present a static constraint transformation (SCT) technique. The SCT ensures that it is sufficient to check the schedulability of a task locally in a node with a local clock, since the global schedulability of the task is derived from its local schedulability through SCT. We formally prove the correctness of SCT.

• Journal of Distribution Science
• /
• v.13 no.1
• /
• pp.27-33
• /
• 2015

Purpose - The ICT market in the EU is lagging behind that of the US; however, algorithm and software development within the EU have grown steadily, and they involve focusing on the creative cultural convergence conceptualized as part of Horizon 2020 and connecting neighboring markets in the EE and the Mediterranean region. It is essential to study the requirements to market the EU's creative ICT development in emerging industrial countries after examining its applicability in these countries. Research design, data, and methodology - This study deals with data pertaining to the EU's creative industry and competitive edge. The global cultural expansion of the EU facilitates a new concept involving not only low-cost IT products to enhance local cultural artifacts through R&D and the construction of efficient infrastructure services, but also information exchange with a realistic commercialization of the technology that can be applied for creative cultural localization. In the European industry, research on algorithms has been applied for the benefit of consumers. We investigated how the process is conducted in the EU. Results - Europe needs to adjust its economic structure to the local culture as part of IT distribution convergence. The convergence has been converted into a production algorithm with IT in the form of low-cost production. This is because there is an attempt to improve the quality of transport infrastructure, workforce availability, and the distribution of the distance to the local industries and consumers, using IT algorithms. Integrated into the manufacturing industry, based on the ICT infrastructure and solutions, smart localized regional clusters are formed with the help of grafting. Europe has own strategy to increase the number of hub-and-spoke cities. Europe is now becoming integrated, with an EPC system for regional cooperation rather than national competition in ICT technology. Europe has also been recognized in this study as changing the step-by-step paradigm for global competitiveness through new creative culture industries. Conclusions - As a result, there are several ways of converging with others through EU R&D intensity; therefore, the EU can be seen as successfully increasing marginal value, which is useful in developing a special industrial cluster or local cultural cities that create converged development by connecting people and objects with IT. In fact, when compared to the US, Europe has a strong culture and the car industries have a tendency to overshadow the IT industries with integration of services in IT distribution. Considering the rapid environmental changes, the convergence of IT services is likely to take place in Europe, similar to the pharmaceutical industry and the automotive industry. This requires a focus on human resources and automated systems management. The trend is to move away from low-wage industries, switched to key personnel centers of the local university-industry. EU emphasizes the creation of IT market demand in Europe involving local cultural convergence for marketing as the second step to strengthen the economic hub-and-spoke areas.

• Journal of KIISE:Software and Applications
• /
• v.26 no.11
• /
• pp.1261-1269
• /
• 1999

다층 퍼셉트론은 다양한 응용 분야에 성공적으로 적용되고 있는 대표적인 신경회로망 모델이다. 그러나 다층 퍼셉트론의 학습에서 나타나는 플라토에 기인한 느린 학습 속도와 지역 극소는 실제 응용문제에 적용함에 있어서 가장 큰 문제로 지적되어왔다. 이 문제를 해결하기 위해 여러 가지 다양한 학습알고리즘들이 개발되어 왔으나, 계산의 비효율성으로 인해 실제 문제에는 적용하기 힘든 예가 많은 등, 현재까지 만족할 만한 해결책은 제시되지 못하고 있다. 본 논문에서는 다층퍼셉트론의 베이시스 함수로 사용되는 시그모이드 함수를 보다 일반화된 형태로 정의하여 사용함으로써 학습에 있어서의 플라토를 완화하고, 지역극소에 빠지는 것을 줄이는 접근방법을 소개한다. 본 방법은 기존의 변형된 가중치 수정식을 사용한 학습 속도 향상의 방법들과는 다른 접근 방법을 택함으로써 기존의 방법들과 함께 사용하는 것이 가능하다는 특징을 갖고 있다. 제안하는 방법의 성능을 확인하기 위하여 간단한 패턴 인식 문제들에의 적용 실험 및 기존의 학습 속도 향상 방법을 함께 사용하여 시계열 예측 문제에 적용한 실험을 수행하였고, 그 결과로부터 제안안 방법의 효율성을 확인할 수 있었다. Abstract A multilayer perceptron is the most well-known neural network model which has been successfully applied to various fields of application. Its slow learning caused by plateau and local minima of gradient descent learning, however, have been pointed as the biggest problems in its practical use. To solve such a problem, a number of researches on learning algorithms have been conducted, but it can be said that none of satisfying solutions have been presented so far because the problems such as computational inefficiency have still been existed in these algorithms. In this paper, we propose a new learning approach to minimize the effect of plateau and reduce the possibility of getting trapped in local minima by generalizing the sigmoidal function which is used as the basis function of a multilayer perceptron. Adapting a new approach that differs from the conventional methods with revised updating equation, the proposed method can be used together with the existing methods to improve the learning performance. We conducted some experiments to test the proposed method on simple problems of pattern recognition and a problem of time series prediction, compared our results with the results of the existing methods, and confirmed that the proposed method is efficient enough to apply to the real problems.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.6
• /
• pp.86-94
• /
• 2019

This paper describes the development of unmanned vehicle remote control system which is configured with steering and accelerating/braking hardware to improve the sense of reality and safety of control. Generally, in these case of the remote control system, a joystick-type device is used for steering and accelerating/braking control of unmanned vehicle in most cases. Other systems have been developing using simple steering wheel, but there is no function of that feedback the feeling of driving situation to users and it mostly doesn't include the accelerating/braking control hardware. The technology of feedback means that a reproducing the feeling of current driving situation through steering and accelerating/braking hardware when driving a vehicle in person. In addition to studying feedback technologies that reduce unfamiliarity in remote control of unmanned vehicles, it is necessary to develop the remote control system with hardware that can improve sense of reality. Therefore, in this study, the reliable remote control system is developed and required system specification is defined for applying force-feedback haptic control technology developed through previous research. The system consists of a steering-wheel module similar to a normal vehicle and an accelerating/braking pedal module with actuators to operate by feedback commands. In addition, the software environment configured by CAN communication to send feedback commands to each modules. To verify the reliability of the remote control system, the force-feedback haptic control algorithms developed through previous research were applied, to assess the behavior of the algorithms in each situation.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.2
• /
• pp.59-76
• /
• 2023

Since the COVID-19 era, the rise in apartment prices has been unconventional. In this uncertain real estate market, price prediction research is very important. In this paper, a model is created to predict the actual transaction price of future apartments after building a vast data set of 870,000 from 2015 to 2020 through data collection and crawling on various real estate sites and collecting as many variables as possible. This study first solved the multicollinearity problem by removing and combining variables. After that, a total of five variable selection algorithms were used to extract meaningful independent variables, such as Forward Selection, Backward Elimination, Stepwise Selection, L1 Regulation, and Principal Component Analysis(PCA). In addition, a total of four machine learning and deep learning algorithms were used for deep neural network(DNN), XGBoost, CatBoost, and Linear Regression to learn the model after hyperparameter optimization and compare predictive power between models. In the additional experiment, the experiment was conducted while changing the number of nodes and layers of the DNN to find the most appropriate number of nodes and layers. In conclusion, as a model with the best performance, the actual transaction price of apartments in 2021 was predicted and compared with the actual data in 2021. Through this, I am confident that machine learning and deep learning will help investors make the right decisions when purchasing homes in various economic situations.

• Journal of the Korean Society of Radiology
• /
• v.85 no.4
• /
• pp.714-726
• /
• 2024

Researchers have developed various algorithms utilizing artificial intelligence (AI) to automatically and objectively diagnose patterns and extent of pulmonary emphysema or interstitial lung diseases on chest CT scans. Studies show that AI-based quantification of emphysema on chest CT scans reveals a connection between an increase in the relative percentage of emphysema and a decline in lung function. Notably, quantifying centrilobular emphysema has proven helpful in predicting clinical symptoms or mortality rates of chronic obstructive pulmonary disease. In the context of interstitial lung diseases, AI can classify the usual interstitial pneumonia pattern on CT scans into categories like normal, ground-glass opacity, reticular opacity, honeycombing, emphysema, and consolidation. This classification accuracy is comparable to chest radiologists (70%-80%). However, the results generated by AI are influenced by factors such as scan parameters, reconstruction algorithms, radiation doses, and the training data used to develop the AI. These limitations currently restrict the widespread adoption of AI for quantifying pulmonary emphysema and interstitial lung diseases in daily clinical practice. This paper will showcase the authors' experience using AI for diagnosing and quantifying emphysema and interstitial lung diseases through case studies. We will primarily focus on the advantages and limitations of AI for these two diseases.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1062-1062
• /
• 2001

The concept of “precision agriculture” or “site-specific farming” is usually confined to the fields of soil science, crop science and agronomy. However, because plants grow in soil, animals eat plants, and humans eat animal products, it could be argued (perhaps with some poetic licence) that the fields of feed quality, animal nutrition and animal production should also be considered in this context. NIR spectroscopy has proved over the last 20 years that it can provide a firm foundation for quality measurement across all of these fields, and with the continuing developments in instrumentation, computer capacity and software, is now a major cog in the wheel of precision agriculture. There have been a few giant leaps and a lot of small steps in the impact of NIR on the animal world. These have not been confined to the amazing advances in hardware and software, although would not have occurred without them. Rapid testing of forages, grains and mixed feeds by NIR for nutritional value to livestock is now commonplace in commercial laboratories world-wide. This would never have been possible without the pioneering work done by the USDA NIR Forage Research Network in the 1980's, following the landmark paper of Norris et al. in 1976. The advent of calibration transfer between instruments, algorithms which utilize huge databases for calibration and prediction, and the ability to directly scan whole grains and fresh forages can also be considered as major steps, if not leaps. More adventurous NIR applications have emerged in animal nutrition, with emphasis on estimating the functional properties of feeds, such as in vivo digestibility, voluntary intake, protein degradability and in vitro assays to simulate starch digestion. The potential to monitor the diets of grazing animals by using faecal NIR spectra is also now being realized. NIR measurements on animal carcasses and even live animals have also been attempted, with varying degrees of success, The use of discriminant analysis in these fields is proving a useful tool. The latest giant leap is likely to be the advent of relatively low-cost, portable and ultra-fast diode array NIR instruments, which can be used “on-site” and also be fitted to forage or grain harvesters. The fodder and livestock industries are no longer satisfied with what we once thought was revolutionary: a 2-3 day laboratory turnaround for fred quality testing. This means that the instrument needs to be taken to the samples rather than vice versa. Considerable research is underway in this area, but the challenge of calibration transfer and maintenance of instrument networks of this type remains. The animal world is currently facing its biggest challenges ever; animal welfare, alleged effects of animal products on human health, environmental and economic issues are difficult enough, but the current calamities of BSE and foot and mouth disease are “the last straw” NIR will not of course solve all these problems, but is already proving useful in some of these areas and will continue to do so.