• International Journal of Computer Science & Network Security
• /
• 제23권3호
• /
• pp.81-93
• /
• 2023

The popularity of short-form video platforms like TikTok has increased recently. Short-form videos are significantly shorter than traditional videos, and viewers regularly switch between different types of content to watch. Therefore, a successful prefetching strategy is essential for this novel type of video. This study provides a resource-effective prefetching technique for streaming short-form videos. The suggested solution dynamically adjusts the quantity of prefetched video data based on user viewing habits and network traffic conditions. The results of the experiments demonstrate that, in comparison to baseline approaches, our method may reduce data waste by 21% to 83%, start-up latency by 50% to 99%, and the total time of Re-buffering by 90% to 99%.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2011년도 추계학술발표대회
• /
• pp.429-430
• /
• 2011

Roughness set theory is a popular topic to use in color-image segmentation. A new popular color image segmentation algorithm is proposed by scientists with the point using traditional histogram and Histon construct roughness set histogram. But, there is still a problem about that is the correlativity of color vector in roughness set histogram, which take an inactive effect in the process of color-image segmentation. Therefore, this paper represents further research based on this and proposed an improved method proved through lot of experiments. The experimental result reduces the correlativity of color vector in roughness set histogram and calculation time remarkably.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2021년도 하계학술대회
• /
• pp.23-25
• /
• 2021

This paper describes an approach of composing music with multiple AI composers. This approach enriches more the creativity space of artificial intelligence music composition than using only one composer. This paper presents a simple example with 2 different deep learning composers working together for composing one music. For the experiment, the two composers adopt the same deep learning architecture of an LSTM model trained with different data. The output of a composer is a sequence of notes. Each composer alternatively appends its output to the resulting music which is input to both the composers. Experiments compare different music generated by the proposed multiple composer approach with the traditional one composer approach.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권7호
• /
• pp.1894-1915
• /
• 2023

Software-Defined Networking (SDN) offers several advantages in dynamic routing, flexible programmable control and custom application-driven network management. However, the programmability of the data plane in traditional SDN is limited. A network operator cannot change the ability of the data plane and perform complex packet processing on the data plane, which limits the flexibility and extendibility of SDN. In the paper, AP-SDN (Action Program enabled Software-Defined Networking) architecture is proposed, which extends the action set of SDN data plane. In the proposed architecture, a modified Open vSwitch is utilized in the data plane allowing the execution of action programs at runtime, thus enabling complex packet processing. An example action program is also implemented which transparently encrypts traffic for terminals. At last, a prototype system of AP-SDN is developed and experiments show its effectiveness and performance.

• International Journal of Computer Science & Network Security
• /
• 제23권7호
• /
• pp.119-130
• /
• 2023

Worldwide organizations use the benefits offered by Cloud Computing (CC) to store data, software and programs. While running hugely complicated and sophisticated software on cloud requires more energy that causes global warming and affects environment. Most of the time energy consumption is wasted and it is required to explore opportunities to reduce emission of carbon in CC environment to save energy. Many improvements can be done in regard to energy efficiency from the software perspective by considering and paying attention on the energy consumption aspects of software's that run on cloud infrastructure. The aim of the current research is to propose a framework with an additional phase called parameterized development phase to be incorporated along with the traditional Software Development Life cycle (SDLC) where the developers need to consider the suggested techniques during software implementation to utilize low energy for running software on the cloud and contribute in green computing. Experiments have been carried out and the results prove that the suggested techniques and methods has enabled in achieving energy consumption.

• Current Optics and Photonics
• /
• 제7권5호
• /
• pp.562-568
• /
• 2023

We designed a combined nanostructure of inverted pyramidal pits and nanoparticles, which can obtain much stronger field enhancement than traditional periodic pits or nanoparticles. The field enhancement |E|/|E₀| is greater than 10 in a large area at 750-820 nm in incident wavelength. |E_max|/|E₀| is greater than 60. Moreover, the hot spot is obtained outside the pits instead of localized inside them, which is beneficial for experiments such as surface-enhanced Raman scattering. The relations between resonant wavelength and structural parameters are investigated. The resonant wavelength shows a linear dependence on the structure's period, which provides a direct way to tune the resonant wavelength. The excitation of a propagating surface plasmon on the periodic structure's surface, a localized surface plasmon of nanoparticles, and a standing-wave effect contribute to the enhancement.

• Journal of information and communication convergence engineering
• /
• 제21권4호
• /
• pp.294-299
• /
• 2023

Micro ribonucleic acids (miRNAs) can regulate the protein expression levels of genes in the human body and have recently been reported to be closely related to the cause of disease. Determining the genes related to miRNAs will aid in understanding the mechanisms underlying complex miRNAs. However, the identification of miRNA-related genes through wet experiments (in vivo, traditional methods are time- and cost-consuming). To overcome these problems, recent studies have investigated the prediction of miRNA relevance using deep learning models. This study presents a method for predicting the relationships between miRNAs and genes. First, we reconstruct a negative dataset using the proposed method. We then extracted the feature using an autoencoder, after which the feature vector was concatenated with the original data. Thereafter, the concatenated data were used to train a long short-term memory model. Our model exhibited an area under the curve of 0.9609, outperforming previously reported models trained using the same dataset.

• Current Optics and Photonics
• /
• 제8권2호
• /
• pp.183-191
• /
• 2024

High-resolution retinal imaging based on adaptive optics (AO) is important for early diagnosis related to retinal diseases. However, in practical applications, closed-loop AO correction takes a relatively long time, and traditional open-loop correction methods have low accuracy in correction, leading to unsatisfactory imaging results. In this paper, a SH-U-net-based open-loop AO wavefront correction method is presented for a retinal AO imaging system. The SH-U-net builds a mathematical model of the entire AO system through data training, and the Root mean square (RMS) of the distorted wavefront is 0.08λ after correction in the simulation. Furthermore, it has been validated in experiments. The method improves the accuracy of wavefront correction and shortens the correction time.

• 디자인학연구
• /
• 제17권3호
• /
• pp.335-342
• /
• 2004

최근 한국의 전통조형에 관련되는 연구가 활발하게 이뤄지고 있다. 그러나 그간 이뤄졌던 선행연구의 대부분이 극히 개인적이고 부분적이며 산발적으로 이뤄졌기에, 그 내용이 단편적이고, 체계화되지 못한 경향이 있다. 따라서 이러한 선행연구들을 체계적 틀로 종합하여 객관적으로 정리할 필요가 있다. 대체적으로 인간의 모든 행위(조형행위 또한)는 목표와 절차와 수단을 갖고 있다. 즉 인간은 어떤 사상을 구현하기 위해서는 세 가지 요소가 필요한데, 이 세 가지 요소는 내용적, 형식적, 그리고 실질적 요소로 구분하는 것이 보편화되어 있다. 내용적 요소는 사상의 가치, 관념, 의미, 그리고 목적으로서 이루고자 하는 목표이며, 형식적 요소는 목표를 이루기 위해 그 사상의 단위들을 구성하는 방법, 원리, 규범, 절차, 형태 그리고 양식이며, 실질적 요소는 형식을 통하여 내용을 구체화시킬 수 있는 구체적인 수단, 도구, 매체, 재료 그리고 기술 등을 가리킨다. 이 세 가지는 상호연관 되어 있어 이중 한 가지라도 결여되면 완벽한 구현을 이룰 수 없다. 인간이 표현코자 하는 사상과 의미는 거의 언어로 이뤄지고 있다. 문장의 주성분에는 주어(생략가능), 목적어(목적), 서술어(방법), 보어(수단)가 있으며, 그것의 구성요소를 층위별로 규칙을 갖고 체계화시켜놓은 것이 문장이므로 위의 내용들을 언어와 비교 연구하여 디자인(조형)과의 상관관계를 살펴 전통조형의 체계에 관련된 분석의 틀을 만들었다. 또한 위의 방법으로 분석ㆍ정리된 결과를 갖고 전통 조형정신을 시대에 적합하게 실현하기 위한 방안을, 단계별로 정리하였다. 터미널에 상호 호환적 형태로 소비될 수 있는 터미널 구조 및 구현, 그리고 실험 결과를 처음으로 제시 하였다는데 의미가 있다고 할 수 있겠다.있을 것이다. for business transactions, which is composed by ′Classify Phase′ that classify transactions. We called this model "3-Phase Commit Method Applied by Classify Phase", we design this model to manage an resource of enterprise efficiently. The proposed method is compared by the method based on 2-Phase commit that could be a problem of management the resource of enterprise, and the advantage of this method is certified to propose the solution of that problem.should be more cautious in interpreting data obtained from so-called "in vivo counter-transport" experiments.d in many countries now. Especially, development of high power/quality diode laser will be accelerate the introduction of this magnificent

• 지능정보연구
• /
• 제18권1호
• /
• pp.23-38
• /
• 2012

온라인 쇼핑몰은 인터넷을 통해 손쉽게 접근이 가능하기 때문에, 최초 구매의사가 발생한 시점으로부터 이에 대한 실제 구매가 실현되기까지의 기간이 오프라인 쇼핑몰에 비해 비교적 짧게 나타난다. 즉 오프라인 쇼핑몰의 경우 구매희망 물품을 바로 구매하기 보다는 몇 개의 물품들을 모아서 구매하는 행태가 일반적이다. 하지만, 인터넷 쇼핑몰의 경우 단 하나의 물품만을 포함하고 있는 주문이 전체 주문의 절반 이상을 차지한다. 따라서 온라인 쇼핑몰 데이터의 장바구니 분석에 전통적 데이터마이닝 기법을 그대로 적용할 경우, Null Transaction의 수가 지나치게 많음으로 인해 합리적 수준의 지지도(Support)를 만족시키는 규칙을 찾는 것이 매우 어렵게 된다. 이러한 이유로 온라인 데이터를 사용한 많은 연구는 동시성 기준을 여러 방법으로 확장하여 사용하였는데, 이들 동시성 기준은 명확한 근거나 합의 없이 연구자의 상황에 따라 임의로 선택된 측면이 있다. 따라서 본 연구에서는 온라인 마켓 분석에 적용되는 구매의 동시성 기준을 정확도 측면에서 평가함으로써, 구매의 동시성 기준 선정을 위한 근거를 제시하고자 한다. 또한 동시성 기준의 정확도가 고객의 평균 구매간격에 따라 상이하게 나타나는 것을 파악하여, 향후 고객의 특성에 따른 차별화된 추천 시스템 구축을 위한 기본 방향을 제시하고자 한다. 이를 위해 국내 대형 인터넷 쇼핑몰의 최근 2년간 실제 거래 내역을 대상으로 실험을 수행하였으며, 실험 결과 단골 고객의 구매 추천을 위한 분석의 경우 추천 범위와 분석 데이터의 동시성 기준을 맞추어 연관규칙을 도출하는 것이 바람직하며, 비단골 고객의 경우 대부분의 추천 범위에 대해서 분석 데이터의 동시성 기준을 비교적 길게 설정하여 연관규칙을 도출하는 것이 바람직한 것으로 나타났다.