• 한국멀티미디어학회논문지
• /
• 제23권9호
• /
• pp.1181-1190
• /
• 2020

Detection of an anomaly pattern deviating normal data distribution in streaming data is an important technique in many application areas. In this paper, a method for detection of an newly emerging pattern in text streaming data which is an ordered sequence of texts is proposed based on text embedding and anomaly pattern detection. Using text embedding methods such as BOW(Bag Of Words), Word2Vec, and BERT, the detection performance of the proposed method is compared. Experimental results show that anomaly pattern detection using BERT embedding gave an average F1 value of 0.85 and the F1 value of 1 in three cases among five test cases.

• 한국지능시스템학회논문지
• /
• 제20권1호
• /
• pp.94-100
• /
• 2010

데이터 마이닝(Data Mining)은 환경으로부터 수집된 데이터에서 패턴을 추출하고 의미 있는 정보를 발견하기 위하여 주로 사용된다. 하지만, 기존의 방법은 데이터의 수집이 완료된 상태에서 분석하는 것을 기반으로 하고 있으며, 시간의 흐름에 따른 패턴의 변화를 반영하기 어렵다. 본 논문은 연속성(Continuity data), 대량성(Large scale) 그리고 패턴의 가변성(Changed pattern)과 같은 특성을 가지는 스트림 데이터(Stream Data)의 분석을 위한 스트리밍 의사결정 나무(Streaming Decision Tree : SDT) 방법을 소개한다. SDT는 연속적으로 발생하는 데이터를 블록으로 정의하고, 각 블록은 의사결정나무 학습 방법을 이용하여 규칙을 추출한다. 추출된 규칙은 발생 시간, 빈도 그리고 모순 등을 고려하여 결합하였다. 실험에서는 시계열 데이터를 이용하여 분석하였고, 적절한 결과를 확인하였다.

• 한국IT서비스학회지
• /
• 제18권4호
• /
• pp.115-134
• /
• 2019

The global music industry has been growing steadily every year. In particular, the proportion of music streaming services in the overall market are expanding. It shows same pattern in domestic market. Also, several companies are providing music streaming services competitively. It is clear that there is growing interest in effective operations of music streaming services. In this study, we examine the impact of customization and network effect on service quality and customer loyalty. We collect survey data and analyze relationships between latent variables using structural equations modeling. We find that customization and network effect of music streaming services positively affect service quality. However, the effect of customization on customer loyalty is not significant and there is a negative effect of network effect on customer loyalty. Finally, we find that service quality works as a mediator between customization and customer loyalty, and service quality also works as a mediator between network effect and customer loyalty.

• 정보과학회 논문지
• /
• 제45권1호
• /
• pp.86-93
• /
• 2018

다수의 클라이언트가 대역폭을 공유해 스트리밍 서비스를 받는 경우 HTTP 적응적 스트리밍(HTTP Adaptive Streaming)은 세그먼트 요청의 ON-OFF 패턴으로 인해 대역폭을 부정확하게 측정하는 문제가 있다. ON-OFF 패턴으로 인한 문제를 해결하기 위해 제안된 PANDA (Probe AND Adapt)는 목표 대역폭을 증가시키면서 요청할 세그먼트의 품질을 결정한다. 하지만 목표 대역폭을 고정된 양만큼 증가시키기 때문에 대역폭 활용도가 낮고 대역폭 변화에 느리게 반응하는 문제가 있다. 본 논문에서는 PANDA의 낮은 대역폭 활용도와 느린 반응성을 개선하기 위한 비디오 품질 조절 기법을 제안한다. 제안하는 기법은 세그먼트 다운로드 시간과 요청 간격을 비교해 대역폭 활용도를 판단한 후 대역폭 활용도에 따라 목표 대역폭을 증가시키는 양을 조절한다. 실험을 통하여 제안하는 기법이 대역폭을 충분히 활용하고 대역폭 변화에 빠르게 반응하는 것을 확인하였다.

• 정보과학회 논문지
• /
• 제43권6호
• /
• pp.718-723
• /
• 2016

스트리밍 데이터 분석에서 개념 변화가 일어나는 시점을 정확히 탐지하는 것은 분류 모델의 성능을 유지하는 데 있어서 매우 중요한 작업이다. 오류율은 스트리밍 데이터에서 개념 변화 탐지를 위해 많이 사용되는 척도이다. 그러나 0과 1로 이루어진 이진 값만으로 예측 결과를 묘사하는 것은 분류 모델의 행동 패턴을 나타내는 유용한 정보의 손실을 초래할 수 있다. 이 논문에서는 오류율을 이용하는 대신에 확률 예측치를 사용하여 분류기의 성능 패턴을 묘사하고 급격한 변화를 탐지하는 효과적인 개념 변화 탐지 방법을 제안한다. 합성데이터와 실제 스트리밍 데이터를 이용한 실험 결과는 제안한 방법이 개념 변화 시점을 탐지하는데 뛰어난 성능을 가짐을 보여준다.

• Journal of Communications and Networks
• /
• 제7권2호
• /
• pp.144-156
• /
• 2005

Video streaming over wireless networks is challenging due to node mobility and high channel error rate. In this paper, we propose a multi-source video streaming (MUVIS) system to support high quality video streaming service over IEEE 802.1l-based wireless networks. We begin by collocating a streaming proxy with the wireless access point to help leverage both the media server and peers in the WLAN. By tracking the peer mobility patterns and performing content discovery among peers, we construct a multi-source sender group and stream video using a rate-distortion optimized scheme. We formulate such a multi-source streaming scenario as a combinatorial packet scheduling problem and introduce the concept of asynchronous clocks to decouple the problem into three steps. First, we decide the membership of the multisource sender group based on the mobility pattern tracking, available video content in each peer and the bandwidth each peer allocates to the multi-source streaming service. Then, we select one sender from the sender group in each optimization instance using asynchronous clocks. Finally, we apply the point-to-point rate-distortion optimization framework between the selected sender-receiver pair. In addition, we implement two different caching strategies, simple caching simple fetching (SCSF) and distortion minimized smart caching (DMSC), in the proxy to investigate the effect of caching on the streaming performance. To design more realistic simulation models, we use the empirical results from corporate wireless networks to generate node mobility. Simulation results show that our proposed multi-source streaming scheme has better performance than the traditional server-only streaming scheme and that proxy-based caching can potentially improve video streaming performance.

• 한국소음진동공학회논문집
• /
• 제20권11호
• /
• pp.1078-1088
• /
• 2010

In order to analyze the quantitative characteristics of acoustic streaming, experimental setup of 3-D stereoscopic PIV(particle imaging velocimetry) was designed and quantitative ultrasonic flow fields in the gap between the ultrasonic vibrator and heat source were measured. Utilizing acoustic streaming induced by ultrasonic vibration, surface temperature drop of cooling object was also measured. The study on smart cooling method by acoustic streaming induced by ultrasonic vibration was performed due to the empirical relations of flow pattern, average flow velocity, different gaps, and enhancement on cooling rates in the gap. Average velocity fields and maximum acoustic streaming velocity in the open gap between the stationary cylindrical heat source and ultrasonic vibrator were experimentally measured at no vibration, resonance, and non-resonance. It was clearly observed that the enhancement of cooling rates existed owing to the acoustic air flow in the gap at resonance and non-resonance induced by ultrasonic vibration. The ultrasonic wave propagating into air in the gap creates steady-state secondary eddy called acoustic streaming which enhances heat transfer from the heat source to encompassing air. The intensity of the acoustic streaming induced by ultrasonic vibration experimentally depended upon the gap between the heat source and ultrasonic vibrator. The ultrasonic vibration at resonance caused the increase of the acoustic streaming velocity and convective heat transfer augmentation when the flow fields by 3D stereoscopic PIV and temperature drop of the heat source were measured experimentally. The acoustic streaming velocity of air enhancement on cooling rates in the gap is maximal when the gap agrees with the multiples of half wavelength of the ultrasonic wave, which is specifically 12 mm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권1호
• /
• pp.308-324
• /
• 2018

HTTP adaptive streaming (HAS) is a promising technology for delivering video content over the Internet. HAS-based video streaming solutions rely on bandwidth estimation to select the appropriate video bitrate. Video streaming solutions that consider network conditions provide users with seamless video playback. However, when multiple clients compete for a common bottleneck link, conventional bandwidth estimation schemes that consider only one client overestimate the network bandwidth due to the ON-OFF traffic pattern. The bandwidth overestimation can cause Quality of Experience (QoE) degradation, such as unnecessary changes in video quality, and unfairness of video quality. In this paper, we propose a client-side bandwidth estimation scheme to obtain a better QoE of HAS in the multiple-client environment. The proposed scheme differentiates the client buffer status according to the buffer occupancy, and then estimates the available network bandwidth based on the buffer status and segment throughput. We evaluate the performance of HAS implemented in the ns-3 network simulator. Simulation results show that compared with the conventional schemes, the proposed scheme can enhance the QoE.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2008년도 추계학술대회B
• /
• pp.2511-2514
• /
• 2008

As the minimum feature size decreases, techniques to avoid contamination and processes to maintain clean wafer surfaces have become very important. The deposition and detachment of nanoparticles from surfaces are major problem to integrated circuit fabrication. Therefore, cleaning technology which reduces nanoparticles is essential to increase yield. Previous megasonic cleaning technology has reached the limits to reduce nanoparticles. Megasonic cleaning is one of the efficiency method to reduce contamination nanoparticle. Two major mechanisms are active in a megasonic cleaning, namely, acoustic streaming and cavitation. Acoustic streaming does not lead to sufficiently strong force to cause damage to the substrates or patterns. Sonoluminescence is a phenomenon of light emission associated with the cavitation of a bubble under ultrasound. We studied a correlation between sonoluminescence and sound pressure distribution for the minimum of pattern damage in megasonic cleaning.

• 방송공학회논문지
• /
• 제23권6호
• /
• pp.813-823
• /
• 2018

끊김 없는 UHD 콘텐츠 스트리밍 서비스를 제공하기 위해서는 많은 가용 대역폭이 필요하다. 기존의 HTTP 적응적 스트리밍 방식은 정상 상태에서 가용 대역폭보다 낮은 품질의 비디오 세그먼트를 주기적으로 요청하여 다운로드 받는다. 정상 상태에서는 가용 대역폭과 콘텐츠의 인코딩 율에 차이에 따라 On-Off 구간의 패턴이 반복되어 발생하고, 반복적인 Off 구간에 의해서 대역폭을 효율적으로 사용하지 못하는 문제점이 있다. 지연과 손실률이 높고 대역폭 변화가 심한 무선 환경에서는 네트워크의 가용대역폭을 부정확하게 측정하기 때문에 문제점이 심화된다. 본 논문에서는 무선 환경에서 UHD 콘텐츠의 효율적인 대역폭 활용을 위한 HTTP 적응적 스트리밍의 세그먼트 스케줄링 기법을 제안한다. 제안하는 기법은 OFF 구간의 빈도수를 줄이기 위해 한 번의 요청 메시지에 여러 세그먼트를 동시에 요청하는 집단 세그먼트 방식이다. 실험을 통해서 제안하는 기법이 대역폭 이용율을 향상시키는 것을 확인 하였다.