• KDI Journal of Economic Policy
• /
• 제46권1호
• /
• pp.53-89
• /
• 2024

The effects of the betterment of enforced intellectual property rights (IPRs) provisions on services export diversification are investigated. The analysis used an unbalanced panel dataset of 76 developing countries over the period of 1970-2014. The empirical analysis is based on the feasible generalized least squares estimator. It suggests that the implementation of weaker IPR protection fosters services export diversification in less developed countries (i.e., those whose real per capita incomes are less than US$US$ 1458.60), including those with a low level of export product upgrading. Conversely, in relatively advanced developing countries (countries whose real per capita income exceeds US$ 3356.80), including those with high levels of export product upgrading, the implementation of stronger IPR laws induces greater services export diversification. Finally, the analysis revealed the existence of a non-linear relationship between IPR protection and services export diversification. The implementation of stronger intellectual property laws spurs services export diversification in countries with high degree of IPR protection, especially when IPR protection exceeds a certain level, recorded here as having a score of 1.197. In contrast, in countries with weaker IPR protection, in particular those with IPR protection levels that score less than 0.915, it is rather the implementation of weaker intellectual property laws that promotes services export diversification.

• 방송공학회논문지
• /
• 제20권3호
• /
• pp.430-439
• /
• 2015

최근 범죄 예방을 위해 폭력행위 검출에 대한 영상 분석 기술에 대한 요구가 증가되고 있다. 영상을 이용한 행동 인식 기술을 많은 연구되고 있지만, 폭력행위에 대한 검출 기술은 상대적으로 텔레비전 또는 영화의 폭력장면 검출에만 초점이 맞추어져 있다. 영화에서 촬영 된 폭력 장면에는 주로 피를 흘리는 모습들이 자주 등장하기 때문에 움직임 정보와 색상 정보를 함께 사용하는 방법을 많이 사용하였다. 하지만 실제 CCTV에서 촬영된 폭력행위의 경우 피가 묻은 장면은 자주 발생하지 않기 때문에 색상 정보를 이용한 폭력행위 검출에는 한계점이 존재한다. 본 논문에서는 영상에서의 움직임 벡터를 이용하여 감시영상에서의 폭력 행동을 검출하는 알고리즘을 제안하고자 한다. 제안하는 방법은 공개 데이터인 USI 데이터와 실제 폭력 행위가 발생한 YouTube 데이터를 사용하여 검출결과를 나타내었다.

• 한국컴퓨터정보학회논문지
• /
• 제27권12호
• /
• pp.161-169
• /
• 2022

최근 들어 모바일 디바이스와 GPS(Global Positioning System)의 발전으로 다양한 위치 기반 서비스(Location-Based Servises, LBS)를 활용할 수 있게 되었다. LBS 사용자는 서비스를 이용하기 위해 자신의 위치 정보를 서비스 제공자에게 노출한다. 이 과정에서 개인의 민감한 정보를 침해할 가능성이 있으므로 사용자의 위치 데이터를 변조하여 프라이버시를 보존할 수 있는 Geo-indistinguishability(Geo-Ind) 기법이 많이 활용되고 있다. 그러나 Geo-Ind 기법으로 인하여 사용자로부터 변조된 데이터를 수집하는 경우, LBS 제공자는 사용자 분포에 대한 정확한 정보를 얻을 수 없다. 그러므로 본 논문에서는 Geo-Ind 기법을 이용하여 사용자로부터 수집한 변조된 위치 데이터로부터 사용자 분포에 대한 정보를 정확하게 계산하기 위한 방법을 제안한다. 특히, Expectation-Maximization(EM) 기법을 이용하여 변조된 데이터로부터 사용자의 위치 분포를 정확하게 예측하기 위한 기법을 제안한다. 또한 실제 데이터를 이용해 제안 기법의 우수성을 입증한다.

• 한국전자통신학회논문지
• /
• 제19권2호
• /
• pp.389-396
• /
• 2024

생성형 AI는 최근 모든 분야에서 활용되고 있으며, 심층 데이터 분석 분야에서도 전문가를 대체할 수준으로 발전하고 있다. 그러나 과학기술 문헌에서의 지역명 식별은 학습 데이터의 부족과 이에 따른 인공지능 모델을 적용한 사례가 전무한 실정이다. 본 연구는 Web of Science에서 한국 기관 소속 저자들의 주소 데이터를 활용해 지역명을 분류하기 위한 데이터셋을 구축하고, 머신러닝 및 딥러닝 모델의 적용을 실험 및 평가했다. 실험 결과 BERT 모델이 가장 우수한 성능을 보였으며, 광역 분류에서는 정밀도 98.41%, 재현율 98.2%, F1 점수 98.31%를 기록하였다. 시군구 분류에서는 정밀도 91.79%, 재현율 88.32%, F1 점수 89.54%를 달성하였다. 이 결과는 향후 지역 R&D 현황, 지역 간 연구자 이동성, 지역 공동 연구 등 다양한 연구의 기반 데이터로 활용이 가능하다.

• 융합보안논문지
• /
• 제24권3호
• /
• pp.41-49
• /
• 2024

트랜스포머 모델의 핵심 요소인 토크나이저는 숫자 형태의 데이터를 제대로 이해하지 못한다. 따라서 패킷 페이로드를 문장처럼 학습하여 실제 네트워크에서 동작 가능한 트랜스포머 기반의 침입탐지 모델을 구축하기 위해서는 16진수 형태의 패킷 페이로드를 문자 형태로 변환하는 것이 필요하다. 이러한 문제 인식 하에 본 연구에서는 3종의 문자 인코딩 방식을 적용하여 패킷 페이로드를 숫자 및 문자 형태로 변환한 후 트랜스포머 모델에 학습시키면서 모델의 탐지성능이 어떻게 달라지는지를 분석하였다. 성능 분석 실험을 위한 데이터세트는 UNSW-NB15 데이터세트에 포함된 PCAP 파일에서 패킷 페이로드를 추출하여 구성하였으며, 학습 모델은 RoBERTa를 사용하였다. 실험 결과, ISO-8859-1 인코딩이 이진분류 및 다중분류에서 가장 우수한 성능을 달성하는 것으로 확인되었으며, 토큰의 수를 512개로 설정하고 최대 에포크를 15회로 증가한 경우에 다중분류 정확도가 88.77%까지 향상되었다.

• International journal of advanced smart convergence
• /
• 제11권1호
• /
• pp.19-27
• /
• 2022

Across the world, 'housing' comprises a significant portion of wealth and assets. For this reason, fluctuations in real estate prices are highly sensitive issues to individual households. In Korea, housing prices have steadily increased over the years, and thus many Koreans view the real estate market as an effective channel for their investments. However, if one purchases a real estate property for the purpose of investing, then there are several risks involved when prices begin to fluctuate. The purpose of this study is to design a real estate price 'return rate' prediction model to help mitigate the risks involved with real estate investments and promote reasonable real estate purchases. Various approaches are explored to develop a model capable of predicting real estate prices based on an understanding of the immovability of the real estate market. This study employs the LSTM method, which is based on artificial intelligence and deep learning, to predict real estate prices and validate the model. LSTM networks are based on recurrent neural networks (RNN) but add cell states (which act as a type of conveyer belt) to the hidden states. LSTM networks are able to obtain cell states and hidden states in a recursive manner. Data on the actual trading prices of apartments in autonomous districts between January 2006 and December 2019 are collected from the Actual Trading Price Disclosure System of the Ministry of Land, Infrastructure and Transport (MOLIT). Additionally, basic data on apartments and commercial buildings are collected from the Public Data Portal and Seoul Metropolitan Government's data portal. The collected actual trading price data are scaled to monthly average trading amounts, and each data entry is pre-processed according to address to produce 168 data entries. An LSTM model for return rate prediction is prepared based on a time series dataset where the training period is set as April 2015~August 2017 (29 months), the validation period is set as September 2017~September 2018 (13 months), and the test period is set as December 2018~December 2019 (13 months). The results of the return rate prediction study are as follows. First, the model achieved a prediction similarity level of almost 76%. After collecting time series data and preparing the final prediction model, it was confirmed that 76% of models could be achieved. All in all, the results demonstrate the reliability of the LSTM-based model for return rate prediction.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.237-250
• /
• 2022

Structural health monitoring (SHM) plays an important role in ensuring the safety and functionality of critical civil infrastructure. In recent years, numerous researchers have conducted studies to develop computer vision and machine learning techniques for SHM purposes, offering the potential to reduce the laborious nature and improve the effectiveness of field inspections. However, high-quality vision data from various types of damaged structures is relatively difficult to obtain, because of the rare occurrence of damaged structures. The lack of data is particularly acute for fatigue crack in steel bridge girder. As a result, the lack of data for training purposes is one of the main issues that hinders wider application of these powerful techniques for SHM. To address this problem, the use of synthetic data is proposed in this article to augment real-world datasets used for training neural networks that can identify fatigue cracks in steel structures. First, random textures representing the surface of steel structures with fatigue cracks are created and mapped onto a 3D graphics model. Subsequently, this model is used to generate synthetic images for various lighting conditions and camera angles. A fully convolutional network is then trained for two cases: (1) using only real-word data, and (2) using both synthetic and real-word data. By employing synthetic data augmentation in the training process, the crack identification performance of the neural network for the test dataset is seen to improve from 35% to 40% and 49% to 62% for intersection over union (IoU) and precision, respectively, demonstrating the efficacy of the proposed approach.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권2호
• /
• pp.775-792
• /
• 2015

Person re-identification is an important and challenging task in computer vision with numerous real world applications. Despite significant progress has been made in the past few years, person re-identification remains an unsolved problem. This paper presents a novel appearance-based approach to person re-identification. The approach exploits region covariance matrix and color histograms to capture the statistical properties and chromatic information of each object. Robustness against low resolution, viewpoint changes and pose variations is achieved by a novel signature, that is, the combination of Log Covariance Matrix feature and HSV histogram (LCMH). In order to further improve re-identification performance, third-party image sets are utilized as a common reference to sufficiently represent any image set with the same type. Distinctive and reliable features for a given image set are extracted through decision boundary between the specific set and a third-party image set supervised by max-margin criteria. This method enables the usage of an existing dataset to represent new image data without time-consuming data collection and annotation. Comparisons with state-of-the-art methods carried out on benchmark datasets demonstrate promising performance of our method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권7호
• /
• pp.1252-1273
• /
• 2011

In Web Service Composition (WSC) area, services selection aims at selecting an appropriate candidate from a set of functionally-equivalent services to execute the function of each task in an abstract WSC according to their different QoS values. In despite of many related works, few of previous studies consider transactional constraints in QoS-aware WSC, which guarantee reliable execution of Composite Web Service (CWS) that is composed by a number of unpredictable web services. In this paper, we propose a novel global selection-optimal approach in WSC by considering both transactional constraints and end-to-end QoS constraints. With this approach, we firstly identify building rules and the reduction method to build layer-based Directed Acyclic Graph (DAG) model which can model transactional relationships among candidate services. As such, the problem of solving global optimal QoS utility with transactional constraints in WSC can be regarded as a problem of solving single-source shortest path in DAG. After that, we present Graph-building algorithms and an optimal selection algorithm to explain the specific execution procedures. Finally, comprehensive experiments are conducted based on a real-world web service QoS dataset. The experimental results show that our approach has better performance over other competing selection approaches on success ratio and efficiency.

• Journal of Electrical Engineering and Technology
• /
• 제10권2호
• /
• pp.699-707
• /
• 2015

Although digit character recognition has got a significant improvement in recent years, it is still challenging to achieve satisfied result if the data contains an amount of distracting factors. This paper proposes a novel digit character recognition approach using a multi-layer hierarchical model, Hybrid Restricted Boltzmann Machines (HRBMs), which allows the learning architecture to be robust to background distracting factors. The insight behind the proposed model is that useful high-level features appear more frequently than distracting factors during learning, thus the high-level features can be decompose into hybrid hierarchical structures by using only small label information. In order to extract robust and compact features, a stochastic 0-1 layer is employed, which enables the model's hidden nodes to independently capture the useful character features during training. Experiments on the variations of Mixed National Institute of Standards and Technology (MNIST) dataset show that improvements of the multi-layer hierarchical model can be achieved by the proposed method. Finally, the paper shows the proposed technique which is used in a real-world application, where it is able to identify digit characters under various complex background images.