• Journal of Digital Contents Society
• /
• v.18 no.4
• /
• pp.739-745
• /
• 2017

This paper presents and discusses an implementation of the GPU shifted sorting method to find approximate k nearest neighbors which executes within "warp", the minimum execution unit in GPU parallel architecture. Also, this paper presents the comparison results with other two common nearest neighbor searching methods, GPU-based kd-tree and ANN (Approximate Nearest Neighbor) library. The proposed implementation focuses on the cases when k is small, i.e. 2, 4, 8, and 16, which are handled efficiently within warp to consider it is very common for applications to handle small k's. Also, this paper discusses optimization ways to implementation by improving memory management in a loop for the CUB open library and adopting CUDA commands which are supported by GPU hardware. The proposed implementation shows more than 16-fold speed-up against GPU-based other methods in the tests, implying that the improvement would become higher for more larger input data.

• Journal of KIISE
• /
• v.42 no.2
• /
• pp.169-176
• /
• 2015

The atomic write technique is a good solution to solve the problem of the double write buffer. The atomic write technique needs modified I/O subsystems (i.e., file system and I/O schedulers) and a special SSD that guarantees the atomicity of the write request. In this paper, we propose the writing unit aligned block allocation technique (for EXT4 file system) and the merge prevention of requests technique for the CFQ scheduler. We also propose an atomic write-supporting SSD which stores the atomicity information in the spare area of the flash memory page. We evaluate the performance of the proposed atomic write scheme in MariaDB using the tpcc-mysql and SysBench benchmarks. The experimental results show that the proposed atomic write technique shows a performance improvement of 1.4~1.5 times compared to the double write buffer technique.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1474-1479
• /
• 2015

With mobile devices showing marked improvement in performance in the age of the Internet of Things (IoT), there is demand for rapid processing of the extensive amount of multimedia big data. However, because research on image searching is focused mainly on increasing accuracy despite environmental changes, the development of fast processing of high-resolution multimedia data queries is slow and inefficient. Hence, we suggest a new distributed image search algorithm that ensures both high accuracy and rapid response by using feature extraction of distributed images based on MapReduce, and solves the problem of memory scalability based on BIRCH indexing. In addition, we conducted an experiment on the accuracy, processing time, and scalability of this algorithm to confirm its excellent performance.

• Journal of The Korean Society of Clinical Toxicology
• /
• v.11 no.1
• /
• pp.28-30
• /
• 2013

A 66-year-old male with chronic alcoholism presented with tremor, gait disturbance, memory impairment, insomnia, decreased appetite, and confusion. The patient had been taking lithium daily for treatment of bipolar disorder. Brain CT showed no specific abnormality, and serum lithium and ammonia levels were 3.63 mEq/L (therapeutic range, 0.6~1.2 mEq/L) and $85{\mu}g/dL$ (reference range: $19{\sim}54{\mu}g/dL$), respectively. Therefore, the initial differential diagnosis included chronic lithium intoxication, hepatic encephalopathy, Wernicke encephalopathy, or alcohol withdrawal syndrome. Even with the provision of adequate hydration, the patient's neurologic status did not show improvement, so that lactulose enema, thiamine replacement, and continuous venovenous hemodiafiltration (CVVHDF) were started on the third admission day. By the fifth admission day he had made a rapid neurologic recovery, and was discharged on the 20th admission day. Therefore, CVVHDF might be a treatment for patients with chronic lithium intoxication, because, even if serum lithium concentration is normal, lithium concentration in the brain may be different from that of the serum.

• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.3
• /
• pp.170-175
• /
• 2009

In this paper, we propose real time automatic speaker verification (ASV) system to protect personal information on personal digital assistant (PDA) device. Recently, the capacity of PDA has extended and been popular, especially for mobile environment such as mobile commerce (M-commerce). However, there still exist lots of difficulties for practical application of ASV utility to PDA device because it requires too much computational complexity. To solve this problem, we apply the method to relieve the computational burden by performing the preprocessing such as spectral subtraction and speech detection during the speech utterance. Also by applying the hidden Markov model (HMM) optimal state alignment and the sequential probability ratio test (SPRT), we can get much faster processing results. The whole system implementation is simple and compact enough to fit well with PDA device's limited memory and low CPU speed.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.53-60
• /
• 2012

Nearest-neighbor classification predicts the class of an input data with the most frequent class among the near training data of the input data. Even though nearest-neighbor classification doesn't have a training stage, all of the training data are necessary in a predictive stage and the generalization performance depends on the quality of training data. Therefore, as the training data size increase, a nearest-neighbor classification requires the large amount of memory and the large computation time in prediction. In this paper, we propose a prototype selection algorithm that predicts the class of test data with the new set of prototypes which are near-boundary training data. Based on Tomek links and distance metric, the proposed algorithm selects boundary data and decides whether the selected data is added to the set of prototypes by considering classes and distance relationships. In the experiments, the number of prototypes is much smaller than the size of original training data and we takes advantages of storage reduction and fast prediction in a nearest-neighbor classification.

• Sleep Medicine and Psychophysiology
• /
• v.8 no.2
• /
• pp.138-143
• /
• 2001

Sleepiness is associated with many different conditions and, as a neglected topic, it can be the cause of serious psychological and social disadvantages. In the aspect of learning, additional problems may arise from poor progress in school caused by the effect of sleepiness on concentration, memory, and other cognitive functions. Narcolepsy is by no means the most common cause of excessive sleepiness. Nonetheless, it is not a rarity, especially in young people. The non-specific nature of early features of narcolepsy, combined with very limited awareness that the condition can start in various ways, leads to many misinterpretations. Misinterpretation of narcolepsy symptoms is not confined to the medical profession. Teachers may well be critical of a student with narcolepsy because of their perception of narcolepsy symptoms as laziness, poor motivation, or difficult behavior and dull learning ability. Inappropriate reactions by parents, teachers, and peers, based on misinterpretation of narcolepsy symptoms or the patient's reactions to them, make a difficult situation worse. Especially in Korea, where schooling is focused on college entrance examinations, the problem is very serious and intensified by inappropriate or delayed diagnosis and treatment. Therefore, psychiatrists should be aware that narcolepsy in young adolescents is not rare and that they need to be familiar with its clinical features in both its classic and less obvious forms. Narcolepsy should be suspected if a adolescent's excessive sleepiness can not be explained in other ways. Therefore, we report on two patients who portray the tendency of dull learning ability and are mistaken as idle students. We diagnosed narcolepsy through polysomnography and multiple sleep latency testing. We treated the students with methylphenidate and pemolin. The students showed improvement in learning ability and were able to adapt better to school.

• The KIPS Transactions:PartC
• /
• v.10C no.7
• /
• pp.879-884
• /
• 2003

Generally we use web cache in order to increase performance of web base system, and a replacement technique has a great influence on performance of web cache. A web cache replacement technique is different from a replacement technique of memory scope, and a unit substituted for is web object Also, as for the web object, a variation of user reference characteristics is very great. Therefore, a web cache replacement technique can reflect enough characteristics of this web object. But the existing web caching techniques were not able to reflect enough these object reference characteristics. A principal viewpoint of this study is reference characteristic analysis, an elevation of an object hit rate, an improvement of response time. First of all we analyzed a reference characteristics of an web object by log analysis. And we divide web cache storage scope using the result of reference characteristics analysis. In the experiment result, we can confirm that performance of an object-hit ratio and a response speed was improved than a conventional technique about a proposal technique.

• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.51 no.5
• /
• pp.185-190
• /
• 2002

Capacitor material utilized in the downsizing passive devices and dynamic random access memory(DRAM) requires the physical and electrical properties at given area such as capacitor thickness reduction, relative dielectric constant increase, low leakage current and thermal stability. Common capacitor materials, $SiO_2$, $Si_3N_4$, $SiO_2$/$Si_3N_4$,TaN and et al., used until recently have reached their physical limits in their application to several hundred angstrom scale capacitor. $Ta_2O_{5}$ is known to be a good alternative to the existing materials for the capacitor application because of its high dielectric constant (25 ~35), low leakage current and high breakdown strength. Despite the numerous investigations of $Ta_2O_{5}$ material, there have little been established the clear understanding of the annealing effect on capacitance characteristic and conduction mechanism, design and fabrication for $Ta_2O_{5}$ film capacitor. This study presents the structure-property relationship of reactive-sputtered $Ta_2O_{5}$ MIM capacitor structure processed by annealing in a vacuum. X-ray diffraction patterns skewed the existence of amorphous phase in as-deposited condition and the formation of preferentially oriented-$Ta_2O_{5}$ in 670, $700^{\circ}C$ annealing. On 670, $700^{\circ}C$ annealing under the vacuum, the leakage current decrease and the enhanced temperature-capacitance characteristic stability. and the leakage current behavior is stable irrespective of applied electric field. The results states that keeping $Ta_2O_{5}$ annealed at vacuum gives rise to improvement of electrical characteristics in the capacitor by reducing oxygen-vacancy and the broken bond between Ta and O.

• YAKHAK HOEJI
• /
• v.51 no.6
• /
• pp.463-468
• /
• 2007

Chong Myung Tang is consisted of three medicinal herbs (Acori Graminei Rhizoma, Polygalae Radix and Hoelen cum Radix). It has been used as a medicine for the purpose of learning and memory improvement. In this paper, Chong Myung Tang was screened the biological activities for Alzheimer's disease. The extract (70% ethanol) of Acari Graminei Rhizoma (1 mg/ml) showed that acetylcholinesterase (AChE) and amyloid beta ($A{\beta}$) peptide aggregation inhibitory potency are 43.1% and 76.5%, respectively. The extract of Polygalae Radix showed inhibitory activity against $A{\beta}_{1-42}$ peptide aggregation (51.5%). To predict the drug-likeness, oral absorption ability; blood-brain barrier (BBB) penetraion rate, mutagenecity and carcinogenicity; in silico screening was performed against 16 compounds in the three medicinal herbs. According to the results, all compounds have appropriate chemical structures as medicines. The six compounds in Acori Graminei Rhizoma and the five compounds in Hoelen cum Radix showed excellent oral absorption rate and BBB penetration rate. The four compounds in Polygalae Radix showed excellent oral absorption rate, but their BBB penetration was presented low rate. And, the extract of Hoelen cum Radix didn't show AChE and $A{\beta}_{1-42}$ peptide aggregation inhibitory activities in vitro. Therefore, their activity in brain may be other mechanism. According to all of the results, in silico prediction technology is convenient and effective to determine biological active compounds in medicinal herbs.