• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.10
• /
• pp.483-488
• /
• 2016

In this paper, we propose a prototype-based classification learning by using the nearest-neighbor rule. The nearest-neighbor is applied to segment the class area of all the training data with hyperspheres, and a hypersphere must cover the data from the same class. The radius of a hypersphere is computed by the mid point of the two distances to the farthest same class point and the nearest other class point. And we transform the prototype selection problem into a set covering problem in order to determine the smallest set of prototypes that cover all the training data. The proposed prototype selection method is designed by a greedy algorithm and applicable to process a large-scale training set in parallel. The prediction rule is the nearest-neighbor rule and the new training data is the set of prototypes. In experiments, the generalization performance of the proposed method is superior to existing methods.

• Journal of KIISE:Databases
• /
• v.31 no.5
• /
• pp.492-504
• /
• 2004

Recently, as growing of interest for LBS(location-based services) techniques, lots of works on moving objects that continuously change their information over time, have been performed briskly. Also, researches for NN(nearest neighbor) query which has often been used in LBS, are progressed variously However, the results of conventional NN Query processing techniques may be invalidated as the query and data objects move. Therefore, they are usually meaningless in moving object management system such as LBS. To solve these problems, in this paper we propose a new nearest neighbor query processing technique, called CTNN, which is possible to meet accurate and continuous query processing for moving objects. Our techniques include an Approximate CTNN(ACTNN) technique, which has quick response time, and an Exact CTNN(ECTNN) technique, which makes it possible to search nearest neighbor objects accurately. In order to evaluate the proposed techniques, we experimented with various datasets. Experimental results showed that the ECTNN technique has high accuracy, but has a little low performance for response time. Also the ACTNN technique has low accuracy comparing with the ECTNN, but has quick response time The proposed techniques can be applied to navigation system, traffic control system, distribution information system, etc., and specially are most suitable when both data and query are moving objects and when we already know their trajectory.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.5
• /
• pp.259-266
• /
• 2022

In this study, we developed a deep learning and vision sensor-based assembly performance evaluation method isfor prefabricated steel structures. The assembly parts were segmented using a modified version of the receptive field block convolution module inspired by the eccentric function of the human visual system. The quality of the assembly was evaluated by detecting the bolt holes in the segmented assembly part and calculating the bolt hole positions. To validate the performance of the evaluation, models of standard and defective assembly parts were produced using a 3D printer. The assembly part segmentation network was trained based on the 3D model images captured from a vision sensor. The sbolt hole positions in the segmented assembly image were calculated using image processing techniques, and the assembly performance evaluation using the k-nearest neighbor algorithm was verified. The experimental results show that the assembly parts were segmented with high precision, and the assembly performance based on the positions of the bolt holes in the detected assembly part was evaluated with a classification error of less than 5%.

• Journal of Information Processing Systems
• /
• v.5 no.3
• /
• pp.145-150
• /
• 2009

The awareness of boundaries in wireless sensor networks has many benefits. The identification of boundaries is especially challenging since typical wireless sensor networks consist of low-capability nodes that are unaware of their geographic location. In this paper, we propose a simple, efficient algorithm to detect nodes that are near the boundary of the sensor field as well as near the boundaries of holes. Our algorithm relies purely on the connectivity information of the underlying communication graph and does not require any information on the location of nodes. We introduce the 2-neighbor graph concept, and then make use of it to identify nodes near boundaries. The results of our experiment show that our algorithm carries out the task of topological boundary detection correctly and efficiently.

• The Bulletin of The Korean Astronomical Society
• /
• v.40 no.1
• /
• pp.48.1-48.1
• /
• 2015

We investigate the role of small-scale and large-scale environments in triggering nuclear activity of the local galaxies using a volume-limited sample with $M_r$ < -19.5 and 0.02 < z < 0.0685 from the Sloan Digital Sky Survey Data Release 7. To fix the mass of the supermassive black hole in its host galaxy, we limit the central velocity dispersion of the sample galaxies. The active galactic nuclei (AGN) host sample is composed of Type II AGNs identified with flux ratios of narrow emission lines with S/N > 6. In this study, we find that the AGN fraction of late-type host galaxies are commonly larger than of early type galaxies. The AGN fraction of host galaxy with late-type nearest neighbor starts to increase as the host galaxy approaches the virial radius of the nearest neighbor (about a few hundred kpc scale). Our result may support the idea that the hydrodynamic interaction with the nearest neighbor plays an important role in triggering the nuclear activity of galaxy. The early-type galaxies in high density regions show decline of AGN activity compared to ones in lower density regions, whereas the direction of the environmental dependence of AGN activity for late-type galaxies is rather opposite. We also find that the environmental dependence of star formation rate is analogous to one of AGN activity except in the high density region.

• Journal of the Korea Safety Management & Science
• /
• v.11 no.1
• /
• pp.27-33
• /
• 2009

A study of fracture to material is getting interest in nuclear and aerospace industry as a viewpoint of safety. Acoustic emission (AE) is a non-destructive testing and new technology to evaluate safety on structures. In previous research continuously, all tensile tests on the pre-defected coupons were performed using the universal testing machine, which machine crosshead was move at a constant speed of 5mm/min. This study is to evaluate an AE source characterization of SM45C steel by using k-nearest neighbor classifier, k-NNC. For this, we used K-means clustering as an unsupervised learning method for obtained multi -variate AE main data sets, and we applied k-NNC as a supervised learning pattern recognition algorithm for obtained multi-variate AE working data sets. As a result, the criteria of Wilk's $\lambda$, D&B(Rij) & Tou are discussed.

• Journal of Communications and Networks
• /
• v.14 no.5
• /
• pp.566-577
• /
• 2012

In randomly deployed networks, such as sensor networks, an important problem for each node is to discover its neighbor nodes so that the connectivity amongst nodes can be established. In this paper, we consider this problem by incorporating the physical layer parameters in contrast to the most of the previous work which assumed a collision channel. Specifically, the pilot signals that nodes transmit are successfully decoded if the strength of the received signal relative to the interference is sufficiently high. Thus, each node must extract signal parameter information from the superposition of an unknown number of received signals. This problem falls naturally in the purview of random set theory (RST) which generalizes standard probability theory by assigning sets, rather than values, to random outcomes. The contributions in the paper are twofold: First, we introduce the realistic effect of physical layer considerations in the evaluation of the performance of logical discovery algorithms; such an introduction is necessary for the accurate assessment of how an algorithm performs. Secondly, given the double uncertainty of the environment (that is, the lack of knowledge of the number of neighbors along with the lack of knowledge of the individual signal parameters), we adopt the viewpoint of RST and demonstrate its advantage relative to classical matched filter detection method.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.1
• /
• pp.34.3-34.3
• /
• 2019

We report our discovery of observational evidence for the coherence between galaxy rotation and the average motion of neighbors. Using the Calar Alto Legacy Integral Field Area (CALIFA) survey data analyzed with the Python CALIFA STARLIGHT Synthesis Organizer (PyCASSO) platform, and the NASA-Sloan Atlas (NSA) catalog, we estimate the angular momentum vectors of 445 CALIFA galaxies and build composite maps of their neighbor galaxies on the parameter space of velocity versus distance. The composite radial profiles of the luminosity-weighted mean velocity of neighbors show striking evidence for dynamical coherence between the rotational direction of the CALIFA galaxies and the average moving direction of their neighbor galaxies. The signal of such dynamical coherence is significant for the neighbors within 800 kpc distance from the CALIFA galaxies with a confidence level of $3.5{\sigma}$, when the angular momentum is measured at the outskirt ($Re<R{\leq}2Re$) of each CALIFA galaxy. We also find that faint or kinematically misaligned galaxies show stronger coherence with neighbor motions than bright or kinematically well-aligned galaxies do. Our results show that the rotation of a galaxy, particularly at its outskirt, may be significantly influenced by recent interactions with its neighbors.

• Journal of KIISE:Databases
• /
• v.36 no.5
• /
• pp.405-410
• /
• 2009

This paper proposes an efficient path finding scheme capable of searching the paths to k static objects from a given query point, aiming at both improving the legacy k-nearest neighbor search and making it easily applicable to the road network environment. To the end of improving the speed of finding one-to-many paths, the modified A* obviates the duplicated part of node scans involved in the multiple executions of a one-to-one path finding algorithm. Additionally, the cost to the each object found in this step makes it possible to finalize the k objects according to the network distance from the candidate set as well as to order them by the path cost. Experiment results show that the proposed scheme has the accuracy of around 100% and improves the search speed by $1.3{\sim}3.0$ times of k-nearest neighbor searches, compared with INE, post-Dijkstra, and $na{\ddot{i}}ve$ method.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.5
• /
• pp.139-144
• /
• 2007

The mobile terminal's frequent changes to the access point introduce significant network overhead in mobile IP networks. To solve this problem, we introduce a hierarchical structure with consideration given to the dynamic value of neighbor scope in IP regional registration[1]. When a mobile terminal moves within the neighbor given by the scope value, it makes registration locally without registration with its home agent. We analyze the algorithm mathematically and show the numerical results. As a result, optimization of the scope value for the localized registration under the hierarchical structure makes the proposed scheme outperform the standard mobile IP protocol[2]. This can be explained from the fact that there is only local registration for terminal's movement within the scope region. Moreover, as the signaling cost for home agent increases, the proposed scheme becomes more advantageous.