• Journal of the Korea Convergence Society
• /
• v.10 no.4
• /
• pp.1-6
• /
• 2019

Neural networks have been used to predict the direction of stock index movement from past data. The conventional research that predicts the upward or downward movement of the stock index predicts a rise or fall even with small changes in the index. It is highly likely that losses will occur when trading ETFs by use of the prediction. In this paper, a neural network model that predicts the movement direction of the daily KOrea composite Stock Price Index (KOSPI) to reduce ETF trading losses and earn more than a certain amount per trading is presented. The proposed model has outputs that represent rising (change rate in index ${\geq}{\alpha}$), falling (change rate ${\leq}-{\alpha}$) and neutral ($-{\alpha}$ change rate < ${\alpha}$). If the forecast is rising, buy the Leveraged Exchange Traded Fund (ETF); if it is falling, buy the inverse ETF. The hit ratio (HR) of PNN1 implemented in this paper is 0.720 and 0.616 in the learning and the evaluation respectively. ETF trading yields a yield of 8.386 to 16.324 %. The proposed models show the better ETF trading success rate and yield than the neural network models predicting KOSPI.

• Korean Chemical Engineering Research
• /
• v.60 no.2
• /
• pp.300-307
• /
• 2022

Time-resolved serial femtosecond crystallography (TR-SFX) is a powerful technique for determining temporal variations in the structural properties of biomacromolecules on ultra-short time scales without causing structure damage by employing femtosecond X-ray laser pulses generated by an X-ray free electron laser (XFEL). The mixing rate of reactants and biomolecule samples, as well as the hit rate between crystal samples and x-ray pulses, are critical factors determining TR-SFX performance, such as accurate image acquisition and efficient sample consumption. We here develop two distinct sample delivery systems that enable ultra-fast mixing and on-demand droplet injecting via pneumatic application with a square pulse signal. The first strategy relies on inertial mixing, which is caused by the high-speed collision and subsequent coalescence of droplets ejected through a double nozzle, while the second relies on on-demand pneumatic jetting embedded with a 3D-printed micromixer. First, the colliding behaviors of the droplets ejected through the double nozzle, as well as the inertial mixing within the coalesced droplets, are investigated experimentally and numerically. The mixing performance of the pneumatic jetting system with an integrated micromixer is then evaluated by using similar approaches. The sample delivery system devised in this work is very valuable for three-dimensional biomolecular structure analysis, which is critical for elucidating the mechanisms by which certain proteins cause disease, as well as searching for antibody drugs and new drug candidates.

• The Journal of the Korea Contents Association
• /
• v.8 no.3
• /
• pp.190-196
• /
• 2008

In these days we can often see examples of suicide. Nowaday suicide becomes a serious social problem in Korean society. The number of suicides increased continuously from 3,157 in 1990 to 10,688 in 2006. The suicide rate per 100,000 person increased progressively from 9.1 in 1991 to 26.1 in 2005 and hit the highest level in 2005. In 2006, The suicide rate decreased after the lapse of 6 years. However, It is still high actual condition with 23.0.

• Journal of Advanced Navigation Technology
• /
• v.15 no.3
• /
• pp.355-363
• /
• 2011

Transcoding in the wireless mobile network is an important mechanism that reduces the delay time and improves the stream processing capacity. Wireless mobile streaming media services, however, have such problems as congestion, interference and delay due to narrow network bandwidth and limited resources. These problems degrade not only Quality of Service (QoS) but also responsiveness of the streaming media service. To solve this problem, this paper proposes a object version transcoding method. The proposed method analyzes the object versions to construct the transcoding graph. This paper utilizes a reference rate-based control function for an efficient streaming, and measures MVDS(Multiple Version Delay Saving) for an efficient delay savings. The simulation results show that the proposed mechanism achieves improved performance in delay rate and cache hit rate compared with those of other existing methods.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.12
• /
• pp.3150-3158
• /
• 1997

In the ATM networks, there are two method in traffic control as schemes to improve the quality of service; one is the reactive control after congestion and the other is the preventive control before congestion. The preventive control include the CAC(Connection Admission Control), the UPC(Usage Parameter Control), the NPC(Network Parameter Control) and the PC(Priority co ntrol). In this paper, we propose an efficient UPC algorithm that has a complex structure using the Jumping window algorithm within the Leaky Bucket algorithm. The proposed algorithm controls peak hit rate by the Leaky Bucket algorithm, then it does the traffic control to evaluate by the Jumping Window whether violates mean bit rate or not. As we assume On/Off traffic source model, our simulation results showed cell loss rate less than the pre-existential Leaky Bucket algorithm method, and it could decrease the demanded Bucket size.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1065-1085
• /
• 2020

With rapid growth of content demands, device-to-device (D2D) content sharing is exploited to effectively improve the service quality of users. Considering the limited storage space and various content demands of users, caching schemes are significant. However, most of them ignore the influence of the asynchronous content reuse and the selfishness of users. In this work, the user preferences are defined by exploiting the user-oriented content popularity and the current caching situation, and further, we propose the social-aware rate, which comprehensively reflects the achievable contents download rate affected by the social ties, the caching indicators, and the user preferences. Guided by this, we model the collaborative caching problem by making a trade-off between the redundancy of caching contents and the cache hit ratio, with the goal of maximizing the sum of social-aware rate over the constraint of limited storage space. Due to its intractability, it is computationally reduced to the maximization of a monotone submodular function, subject to a matroid constraint. Subsequently, two social-aware collaborative caching algorithms are designed by leveraging the standard and continuous greedy algorithms respectively, which are proved to achieve different approximation ratios in unequal polynomial-time. We present the simulation results to illustrate the performance of our schemes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.9A
• /
• pp.2173-2183
• /
• 1998

We consider a frequency-hopped multiple-access communication system that employs reed-solomon code over GF(Q) and M-ary FSK signaling ($M{\leq}Q$) in rayleigh fading channel. We investigate the tradeoff among the modulation symbol size (M), the number of frequency slots, and the code rate in maximizing the average number of successfully transmitted information bits per unit time and unit bandwidth (called normalized throughput). We find that it is desirabel to use a large M in noise-limited environment. In interference-limited environment, it is more improtant to prevent errors (hits) by increasing the number of frequency slots than to correct them with formward error correction techniques or to reduce the error rate by increasing M.

• Journal of the Korean Society of Safety
• /
• v.28 no.1
• /
• pp.6-11
• /
• 2013

In this study, the effect of damping and elastic nature on the control performance of a safety budget-industrial accident rate model in Korea is examined first. The effectiveness of such dynamic model in establishing safety policies is shown with a simple proportional-integral(PI) feedback control mechanism. Control performance of the safety system model is explained in view of maximizing the effect of IAPF and minimizing the absolute amount of IAFP. Control performance is then evaluated and proved to be effective to prevent and reduce the industrial accidents. Implications in feedback control of a safety system model suggested to optimization of safety policies are also explored. Without proper restructuring of the safety system, it would not be possible to hit the target industrial accident rate. Even if the control objective is met, the amount of industrial accident prevention fund required to reduce the industrial accident rate from the current level to the target level would be far beyond the social consensus.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.369-384
• /
• 2021

In a personalized product recommendation system, when the amount of log data is large or sparse, the accuracy of model recommendation will be greatly affected. To solve this problem, a personalized product recommendation method using deep factorization machine (DeepFM) to analyze user behavior is proposed. Firstly, the K-means clustering algorithm is used to cluster the original log data from the perspective of similarity to reduce the data dimension. Then, through the DeepFM parameter sharing strategy, the relationship between low- and high-order feature combinations is learned from log data, and the click rate prediction model is constructed. Finally, based on the predicted click-through rate, products are recommended to users in sequence and fed back. The area under the curve (AUC) and Logloss of the proposed method are 0.8834 and 0.0253, respectively, on the Criteo dataset, and 0.7836 and 0.0348 on the KDD2012 Cup dataset, respectively. Compared with other newer recommendation methods, the proposed method can achieve better recommendation effect.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1245-1251
• /
• 2017

In this study we present a novel directory architecture that can dynamically allocate a directory entry for a cache block on demand at runtime only when the block is shared by more than one core. Thus, we do not maintain coherence for private blocks, substantially reducing the number of directory entries. Even for shared blocks, we allocate directory entry dynamically only when the block is actively shared, further reducing the number of directory entries at runtime. For this, we propose a new directory architecture called dynamic directory table (DDT), which is implemented as a cache of active directory entries. Through our detailed simulation on PARSEC benchmarks, we show that DDT can outperform the expensive full-map directory by a slight margin with only 17.84% of directory area across a variety of different workloads. This is achieved by its faster access and high hit rates in the small directory. In addition, we demonstrate that even smaller DDTs can give comparable or higher performance compared to recent directory optimization schemes such as SPACE and DGD with considerably less area.