• International Journal of Computer Science & Network Security
• /
• 제23권9호
• /
• pp.47-54
• /
• 2023

For Bengali music emotion classification, deep learning models, particularly CNN and RNN are frequently used. But previous researches had the flaws of low accuracy and overfitting problem. In this research, attention-based Conv1D and BiGRU model is designed for music emotion classification and comparative experimentation shows that the proposed model is classifying emotions more accurate. We have proposed a Conv1D and Bi-GRU with the attention-based model for emotion classification of our Bengali music dataset. The model integrates attention-based. Wav preprocessing makes use of MFCCs. To reduce the dimensionality of the feature space, contextual features were extracted from two Conv1D layers. In order to solve the overfitting problems, dropouts are utilized. Two bidirectional GRUs networks are used to update previous and future emotion representation of the output from the Conv1D layers. Two BiGRU layers are conntected to an attention mechanism to give various MFCC feature vectors more attention. Moreover, the attention mechanism has increased the accuracy of the proposed classification model. The vector is finally classified into four emotion classes: Angry, Happy, Relax, Sad; using a dense, fully connected layer with softmax activation. The proposed Conv1D+BiGRU+Attention model is efficient at classifying emotions in the Bengali music dataset than baseline methods. For our Bengali music dataset, the performance of our proposed model is 95%.

• 한국IT서비스학회지
• /
• 제17권2호
• /
• pp.143-163
• /
• 2018

The business models has a great impact on the successful management of enterprises. Business environment has been shifting from industrial economy to knowledge-based economy. Enterprises go through numerous trials for successful management in the changing environment. Along with trial tests, research areas have been growing simultaneously. Although many researches have been conducted with regard to business models, it is very insufficient to systematically analyze the knowledge flow of research. Accordingly, successive researchers who want to study the business model may find it difficult to establish the orientation of future application research based on understanding the process of changing the knowledge structure that have accumulated so far. This study is intended to determine the current state of the business model research and to understand the process of knowledge structure changes in keywords that appear in 2,667 business model articles in the SCOPUS database. Identifying the knowledge structure has been completed through social network analysis, a methodology based on the 'relationship', and the changes in the knowledge structure were identified by classifying them into four different periods. The analysis showed that, first, the number of business model co-author increases over time with the need for academic diversity. Second, the 'innovation' keyword has the biggest center in the network, and over time, the lower-rank keyword which was in the former period has emerged as the top-rank keyword. Third, the cohesiveness group decreased from 12 before 2000 to 5 in 2015 and also the modularity decreased as well. Finally, examining characteristics of study area through a cognitive map showed that the relationships between domains increased gradually over time. The study has provided a systematic basis for understanding the current state of the business model research and the process of changing knowledge structure. In addition, considering that no research has ever systematically analyzed the knowledge structure accumulated by individual researches, it is considered as a significant study.

• 한국국방경영분석학회지
• /
• 제28권1호
• /
• pp.115-135
• /
• 2002

The limits of current DN(Defense networks), private and closed network, become to reality; for Example, high expense of construction and maintenance of networks, restriction of new subscribers on DN. Therefore, a network using web environment that reflect fast development of If and IS(Information Security) technology is demanded for MND. Meeting the requirement of reliable IS system and extension and improvement of DN using common network, we can reduce the expense to extend, maintain, repair DN, form the environment that makes military business cooperate better with civil company and government agency, advance implementing Defense computing and networking service for field small size units that was a exception of Defense digitalization. But it is essential to construct DN based on common network that there are security requisites; confidentiality, integrity, availability, efficiency, log, backup, restoration, that have to be realized at demanding level for IS. This thesis suggested four measurements; replacement DN with common network to resolve the requirements of building new network and improvement of performance for private DN, linkage with common network for new requirement, distribution of traffic using common network, configuration of DN using Internet and Proposed a refinement of IS management organization to treat security threat of common network flexibly, and LAN IS standard model of DN based on the web environment.

• Journal of Information Processing Systems
• /
• 제7권1호
• /
• pp.75-84
• /
• 2011

The enhanced pyramid graph was recently proposed as an interconnection network model in parallel processing for maximizing regularity in pyramid networks. We prove that there are two edge-disjoint Hamiltonian cycles in the enhanced pyramid networks. This investigation demonstrates its superior property in edge fault tolerance. This result is optimal in the sense that the minimum degree of the graph is only four.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제33권8호
• /
• pp.692-702
• /
• 2006

본 연구에서는 패턴 분류문제를 위하여 가중치 개념을 갖는 퍼지 최대 최소 신경망 모델을 제안한다. 제안된 모델은 기존의 FMM 신경망 모델에 대하여 학습데이타에 포함되는 특징값의 빈도요소를 효과적으로 반영할 수 있도록 수정한 구조를 갖는다. 본 논문에서는 제안된 모델에 대하여 하이퍼박스 소속함수로 정의되는 새로운 활성화 특성과 학습알고리즘을 정의한다. 학습알고리즘은 하이퍼박스 생성 및 확장, 중첩 테스트, 하이퍼박스 축소의 3 단계 과정으로 이루어지며, 각 과정에서 특정값의 빈도요소를 고려하여 가중치값을 갱신하는 규칙이 새롭게 정의된다. 본 연구에서는 또한 제안된 모텔의 응용으로서 특정분석 기법을 제안한다. 이를 위하여 특정값, 특정유형, 하이퍼박스, 패턴클래스 상호간 연관도 요소를 4 가지 유형의 척도로 정의하여, 주어진 패턴분류 문제에서 각 특정의 상대적 중요도를 평가할 수 있도록 한다. 아이리스 데이타와 클리블랜드 의료데이타에 대한 분류문제에 적용한 실험결과를 통하여 제안된 방법의 타당성을 고찰하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권3호
• /
• pp.881-895
• /
• 2023

Modern image inpainting techniques based on deep learning have achieved remarkable performance, and more and more people are working on repairing more complex and larger missing areas, although this is still challenging, especially for facial image inpainting. For a face image with a huge missing area, there are very few valid pixels available; however, people have an ability to imagine the complete picture in their mind according to their subjective will. It is important to simulate this capability while maintaining the identity features of the face as much as possible. To achieve this goal, we propose a three-stage network model, which we refer to as the identity and structure feature refinement network (ISFRNet). ISFRNet is based on 1) a pre-trained pSp-styleGAN model that generates an extremely realistic face image with rich structural features; 2) a shallow structured network with a small receptive field; and 3) a modified U-net with two encoders and a decoder, which has a large receptive field. We choose structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), L1 Loss and learned perceptual image patch similarity (LPIPS) to evaluate our model. When the missing region is 20%-40%, the above four metric scores of our model are 28.12, 0.942, 0.015 and 0.090, respectively. When the lost area is between 40% and 60%, the metric scores are 23.31, 0.840, 0.053 and 0.177, respectively. Our inpainting network not only guarantees excellent face identity feature recovery but also exhibits state-of-the-art performance compared to other multi-stage refinement models.

• 패션비즈니스
• /
• 제15권5호
• /
• pp.87-102
• /
• 2011

The research intends to exploit a strategy method of personal branding improving a personal value for construction of a personal image. As an assessment, the model of construction strategy of personal branding is developed in four steps of a model, construction of personal branding, by using elements of personal image and researching about personal branding strategy of scholars. In order to substantiate a validity of presented model, the case analyses of Martha Stewart. The strategy of four steps for construction of effective personal image is explained below. First step is an analysis of personal brand equity, deciding a direction of the concept of a personal branding through analyzing into a core value and core competence of one. Second step is a personal brand identity, constructing personal specification and identity with elements of personal image by using effective strategy, being able to be perceived to population. Third step is a personal brand positioning, constructing competitive brand image by using analysis of SWOT and strategy STP. Fourth step is a promotion of personal brand, advertising and extending a brand image of one by using a public activity and communication methods such as publication, mass media, and social network. By using the four kinds of processes, constructed strategy of a personal brand will be significant for construction of an effective personal image by having increment of a value and power of the brand.

• Journal of Construction Engineering and Project Management
• /
• 제6권3호
• /
• pp.1-7
• /
• 2016

The purpose of this paper is to develop a model for selecting the best contractor in the Gaza Strip using the Artificial Neural Network (ANN). The contractor's selection methods and criteria were identified using a field survey. Fifty four engineers were asked to fill a questionnaire that covers factors related to the selection criteria of contractors practiced in Gaza Strip. The results shows that the dominant part of respondents (91%) confirmed that the current awarding method "the lowest bid price" is considered one of the major problems of the construction sector, "award the bid to the highest weight after combination of the technical and financial scores" represented 50% of the respondents. The criteria weights were determined based on Relative Importance Index (RII. Ninety-one tenders(13 projects) were used to train and test the ANN model after re-evaluating the contractors depend on the weights of factors to select the best contractor who achieves the highest score. Neurosolution software was used to train the models. The results of the trained models indicated that neural network reasonably succeeded in selection the best contractor with 95.96% accuracy. The performed sensitivity analysis showed that the profitability and capital of company are the most influential parameters in selection contractors. This model gives chance to the owner to be more accurate in selecting the most appropriate contractor.

• Journal of Information Technology Applications and Management
• /
• 제20권1호
• /
• pp.87-106
• /
• 2013

This study is to explore the airline network patterns of Incheon and Narita International Airports using passenger flight departure and arrival data of the two airports. The so-called Origin-Destination data is collected from the airports' websites and some of the important data items are flight number, city of origin, destination city, departure/arrival time, number of flights, and delay time. A snowflake schema dimensional model is proposed and implemented. Tableau Public, a well-known visual analytic tool, is used to connect the dimensional model and played an important role in navigating the data space to find interesting and visual patterns among corresponding airports and airlines. For the efficiency of analyzing this spacious data mart, data visualization method was used. Four types of visualization method proposed by Yau was used; visualizing patterns over time, visualizing proportions, visualizing relationships, and visualizing spatial relationships. The strength of connectivity of each flight segments is calculated to evaluate the degree of globalization of Seoul and Tokyo. We anticipate that various patterns and new findings produced by the data mart would provide airline managers, airport authorities, and policy makers in the field of travel and transportation with insightful information.

• 한국비블리아학회지
• /
• 제14권2호
• /
• pp.119-140
• /
• 2003

국가 통합정보망 구축을 통하여 지식과 정보의 생산, 유통 및 활용을 극대화할 수 있다. 지식기반사회를 형성하는 기본적인 정보망은 교육, 노동시장, 산학협동 그리고 평생교육 정보망으로구성된다. 이들 네 영역의 정보망에 대한 현황과 문제점 분석을 기반으로 하여 국가인적자원개발을 위한 체계적인 통합정보망 구축방안을 제시하고자 한다. 국가 통합정보망 구축 방향을 정보기반 인프라와 정보지원 인프라로 구분하여 제시하고 있다. 정보기반 인프라를 위하여 지식과 정보에 대한 디지털화의 촉진, 접근성 향상, 컨텐츠 개발 등이 필요하며, 정보지원 인프라를 위해서는 법적$\cdot$제도적 기반 정비, 행$\cdot$재정 지원체제 정비 그리고 다양한 인적자원개발 정책이 요구된다.