• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.371-373
• /
• 2023

factor 들이 많은 데이터의 군집화는 어려움을 요한다. K-means 클러스터링을 사용하여 군집화를 할 때, 각 데이터들이 가진 factor 의 개수가 상이한 경우 비슷한 성향을 가진 데이터임에도 불구하고 클러스터링이 적합하게 되지 않는 현상이 발생한다. 이러한 문제점을 해결하기 위해 최적의 군집화 개수를 결정하는 실루엣 기반 방법을 제안하고 제안기법의 성능을 평가한다.

• Journal of Korean Academy of Nursing
• /
• v.14 no.2
• /
• pp.84-92
• /
• 1984

This article examined relationships between selected variables, such as demographic background, care, treatment variables, environmental characteristics, and patient's daily behavior and mood change. Relationship were determined between independent variabltherapeutic-rapeutie approach, demographic data, environmental management approach-,and dependent variable-patient's daily behavioral and mood change. 35 patients selected within some criteria in a psychiatric ward, were obserbed during 5 weeks by use of Wyatt's Behavior & Mood Rating Scale ac-cording to the object of the study. At the same time, the frequence of the care and treatment were collected. Criteria for sample selection and independent variables as an influential factor to the patient behavioral change, based on a literature revienw and clinical experiences. Pearson's correlation and multiple regression analysis were used to determine the influfntial factors to the patient behavioral change. Systematic reading (r=.8324), Psychiatrist's individual interview (r=.5764), tranquilizer (r=.3441) and hospitalization processing date (r=.4143) were related with patient's behavioral change. That is these 4 variables can be said to influence to the patient's behavior and mood. A stepwise multiple regression analysis of the effect of the independent varibles of systematic reading, psychintrists individual interview, tranquilizer and hospitalization processing date on the dependent variable, patient's behavioral change was carried out. Systematic reading with on R²of. 69 revealed to be the main influential factor to the patient's behavior and mood change, as the next factor psychiatrist individual interview. A total inclusion of these factors revealed a 73% prediction for the patient's behavior and mood change. But the most influential factor was the interaction of the systematic reading and psychiatrist's individual interview.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.1
• /
• pp.19-24
• /
• 2021

In this paper, we study performance factor of machine learning in the distributed environment using Apache Spark and presents an efficient distributed processing method through experiments. This work firstly presents performance factor when performing machine learning in a distributed cluster by classifying cluster performance, data size, and configuration of spark engine. In addition, performance study of regression analysis using Spark MLlib running on the Hadoop cluster is performed while changing the configuration of the node and the Spark Executor. As a result of the experiment, it was confirmed that the effective number of executors was affected by the number of data blocks, but depending on the cluster size, the maximum and minimum values were limited by the number of cores and the number of worker nodes, respectively.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.9
• /
• pp.703-711
• /
• 1991

In this paper, an efficient storage reclamation algorithm for RISC parallel processing in the object orented programming environments is presented. The memory management for the dynamic memory allocation and the frequent memory access in object oriented programming is the main factor that decreases RISC parallel processing performance. The proposed algorithm can be efficiently allocated the memory space of RISCy computer which is required the frequent memory access, so it can be increased RISC parallel processing performance. The proposed algorithm is verified the efficiency by implementing C language on SUN SPARC(4.3 BSD UNIX).

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.1E
• /
• pp.7-13
• /
• 2001

In this paper we propose a new subspace tracking algorithm based on the PASTd (Projection Approximation Subspace Tracking with deflation). The algorithm is obtained via introducing the variable forgetting factor which adapts itself to the time-varying subspace environments. The tracking capability of the proposed algorithm is demonstrated by computer simulations in an abruptly changing DOA scenario. The estimation results of the variable forgetting factor PASTd(VFF-PASTd) outperform those of the PASTd in the nonstationary case as well as in the stationary case.

• Journal of the Korean Society of Industry Convergence
• /
• v.3 no.3
• /
• pp.229-236
• /
• 2000

This paper proposes on the Hanguel character CAI system using image processing. For this, firstly, the characters written by elementary school students or foreigners arc captured by CCD camera. Secondly, Recognition is accomplished by pre-processing, thinning and recognition processes. Thirdly, strokes are separated and beauty evaluation is done by matching feature value of the input image from the similarity measure function. In particular, this paper describe to define the similarity measuring function using extracted factor values after getting the beauty evaluation factor values of the consonant in the entire CAI system. Finally, the effectiveness of the proposed system is demonstrated by experiments.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.9
• /
• pp.97-103
• /
• 2006

This paper described the design and performance evaluation of fiber optic gyro using digital closed-loop processing. For the feedback to null the gyro input rate, digital serrodyne modulation was employed, and for scale factor stabilization, the control circuits of modulation amplitude and optical power are implemented. Performance tests show that prototype fiber optic gyro has bias stability of 0.34 deg/hr, scale factor non-lineality of about 100ppm, and maximum measurement range of ${\pm}500$ deg/sec.

• Journal of the Korean Society of Safety
• /
• v.25 no.6
• /
• pp.161-168
• /
• 2010

A series of fatigue tests have been performed on the non-load carrying fillet welded joints in order to quantitatively assess the fatigue characteristics due to the grindings and TIG(Tungsten Inert Gas) welding treatments. From the results of fatigue tests, it has been shown that the fatigue strengths at $2{\times}106$ load cycles were improved in the case of the grinding sand TIG welding treatments, and we could know that it is satisfying fatigue strength prescribed in fatigue design standard in general. Besides, from the results of fracture mechanics approaches, the geometric shape correction factors were the most dominant factors in the initial fatigue crack growth, but as the fatigue crack develops, the finite plate correction factor were became the most dominant factor, and the fatigue life on non-load carrying fillet welded joints could be relatively exactly estimated by using the relations between fatigue crack growth rate and stress intensity factor obtained from finite element analysis and existing proposed formulae.

• Journal of Information Processing Systems
• /
• v.17 no.1
• /
• pp.163-177
• /
• 2021

Many methods of discovering social networking communities or clustering of features are based on the network structure or the content network. This paper proposes a community discovery method based on topic models using a time factor and an unsupervised clustering method. Online community discovery enables organizations and businesses to thoroughly understand the trend in users' interests in their products and services. In addition, an insight into customer experience on social networks is a tremendous competitive advantage in this era of ecommerce and Internet development. The objective of this work is to find clusters (communities) such that each cluster's nodes contain topics and individuals having similarities in the attribute space. In terms of social media analytics, the method seeks communities whose members have similar features. The method is experimented with and evaluated using a Vietnamese corpus of comments and messages collected on social networks and ecommerce sites in various sectors from 2016 to 2019. The experimental results demonstrate the effectiveness of the proposed method over other methods.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.4
• /
• pp.517-524
• /
• 1998

This paper discusses the development of a reactive power controller using digital signal processing. Digital Signal Processing is the technique of using digital devices to Process continuous signals or data, often in real-time. And DSP algorithms are associated with a discrete time interval between input samples. When one designs a digital filter, one can use a Laplace transform to determine the continuous time frequency response. The corresponding discrete time transform is called Z transform and depends upon discrete samples of the input spaced equally in time. The objectives of this paper are to minimize real power losses and improve the power factor of a given system. Also, the implementation of a direct-form non recursive filter on the TMS320C31 has been described. The application of this microprocessor-based controller using DSP on test system reveals its numerous advantages. Performance and features of the controller for the reactive power control are analyzed.