• Journal of the Korea Society of Computer and Information
• /
• v.18 no.2
• /
• pp.107-116
• /
• 2013

In this paper, we propose a hybrid music recommendation system combining users' listening habits and tag information in a social music site. Most of commercial music recommendation systems recommend music items based on the number of plays and explicit ratings of a song. However, the approach has some difficulties in recommending new items with only a few ratings or recommending items to new users with little information. To resolve the problem, we use tag information which is generated by collaborative tagging. According to the meaning of tags, a weighted value is assigned as the score of a tag of an music item. By combining the score of tags and the number of plays, user profiles are created and collaborative filtering algorithm is executed. For performance evaluation, precision, recall, and F-measure are calculated using the listening habit-based recommendation, the tag score-based recommendation, and the hybrid recommendation, respectively. Our experiments show that the hybrid recommendation system outperforms the other two approaches.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.1 no.3
• /
• pp.69-76
• /
• 2008

This paper concerns Fuzzy Radial Basis Function Neural Network (FRBFNN) and automatic rule generation of extraction of the FRBFNN by means of mountain clustering. In the proposed network, the membership functions of the premise part of fuzzy rules do not assume any explicit functional forms such as Gaussian, ellipsoidal, triangular, etc., so its resulting fitness values (degree of membership) directly rely on the computation of the relevant distance between data points. Also, we consider high-order polynomial as the consequent part of fuzzy rules which represent input-output characteristic of sup-space. The number of clusters and the centers of clusters are automatically generated by using mountain clustering method based on the density of data. The centers of cluster which are obtained by using mountain clustering are used to determine a degree of membership and weighted least square estimator (WLSE) is adopted to estimate the coefficients of the consequent polynomial of fuzzy rules. The effectiveness of the proposed model have been investigated and analyzed in detail for the representative nonlinear function.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1054-1061
• /
• 2003

Q-Learning, a representative algorithm of reinforcement learning, experiences repeatedly until estimation values about all state-action pairs of state space converge and achieve optimal policies. When the state space is high dimensional or continuous, complex reinforcement learning tasks involve very large state space and suffer from storing all individual state values in a single table. We introduce Q-Map that is new function approximation method to get classified policies. As an agent learns on-line, Q-Map groups states of similar situations and adapts to new experiences repeatedly. State-action pairs necessary for fine control are treated in the form of rule. As a result of experiment in maze environment and mountain car problem, we can achieve classified knowledge and extract easily rules from Q-Map

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.1
• /
• pp.58-67
• /
• 2000

Todays enterprise networks are composed of multiple types of interconnected networks.Enterprise Networks interoperability between these technologies is needed. To enable interworking, it is necessary to be able to map between the relevant object models and to build on this to provide a mechanism to handle protocol conversion on the domain boundaries.In this paper, we describe a gateway between management application in the CORBA domain and agent in the SNMP domain and various integration methods. The main function of the gateway is to dynamically convert the method invocations on object reference in CORBA domain to SNMP messages for MIB entries at remote agents, We also present translation methods from SNMP MIB to CORBA IDL using Direct translation and Abstract translation. JIDM algorithm has no notion of containment and inheritance relationships between object classes and is difficult to understanding between management attribute and SNMP Action attribute. Abstract translation over come these problems.New superclasses define for common attributes and define explicit CORBA method for SNMP Action.It is a methodology for obtaining the CORBA-compliant management agents from already existing SNMP agents.

• Earthquakes and Structures
• /
• v.13 no.4
• /
• pp.377-386
• /
• 2017

Earthquake-induced pounding damages to building structures were repeatedly observed in many previous major earthquakes. Extensive researches have been carried out in this field. Previous studies mainly focused on the regular shaped buildings and each building was normally simplified as a single-degree-of-freedom (SDOF) system or a multi-degree-of-freedom (MDOF) system by assuming the masses of the building lumped at the floor levels. The researches on the pounding responses between irregular asymmetric buildings are rare. For the asymmetric buildings subjected to earthquake loading, torsional vibration modes of the structures are excited, which in turn may significantly change the structural responses. Moreover, contact element was normally used to consider the pounding phenomenon in previous studies, which may result in inaccurate estimations of the structural responses since this method is based on the point-to-point pounding assumption with the predetermined pounding locations. In reality, poundings may take place between any locations. In other words, the pounding locations cannot be predefined. To more realistically consider the arbitrary poundings between asymmetric structures, detailed three-dimensional (3D) finite element models (FEM) and arbitrary pounding algorithm are necessary. This paper carries out numerical simulations on the pounding responses between a symmetric rectangular-shaped building and an asymmetric L-shaped building by using the explicit finite element code LS-DYNA. The detailed 3D FEMs are developed and arbitrary 3D pounding locations between these two buildings under bi-directional earthquake ground motions are investigated. Special attention is paid to the relative locations of two adjacent buildings. The influences of the left-and-right, fore-and-aft relative locations and separation gap between the two buildings on the pounding responses are systematically investigated.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.48.4-48.4
• /
• 2019

We studied the metal-distribution of isolated Milky-way mass galaxy using various hydrodynamic solvers and investigated the difference of the result between AMR and SPH codes. In particle-based codes, physical quantities like mass or metallicity defined in each particle are conserved unless being injected explicitly by the effect of the supernova, whereas in the Eulerian codes the diffusion is simply accomplished by hydro-equation. Therefore, without including explicit physics of diffusion on the SPH- codes, the metal mixing in the galaxy or CGM only can be accomplished by the direct motion of the particles, however, the standard-SPH codes depress the instability of the turbulent fluid mixing. In this work, we simulated under common initial conditions, common gas-physics like cooling-heating models, and star-formation feedback using ENZO(AMR) GIZMO and GADGET-2 codes. We additionally included a metal-diffusion algorithm on the SPH-codes, which follows the subgrid-turbulent mixing model investigated by Shen et al. (2010) and compared the effect of the metal-outflow on the halo region of the galaxy in different hydro-solvers. We also found that for the implementation of the diffusion scheme in the SPH-codes, the existence of a sufficient number of the gas-particles, which is the carrier of the metals, is necessary. So we tested a new initial condition for proper implementation of the diffusion scheme on the SPH simulations. By comparing the metal-contamination of the circumgalactic medium with different hydrodynamics models, we quantify the diffusion strength of AMR codes using diffusion parameterization of the SPH codes and also suggest the calibration solutions in the different behavior of codes in metal-outflow.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.11-19
• /
• 2024

Multi-agent systems can be utilized in various real-world cooperative environments such as battlefield engagements and unmanned transport vehicles. In the context of battlefield engagements, where dense reward design faces challenges due to limited domain knowledge, it is crucial to consider situations that are learned through explicit sparse rewards. This paper explores the collaborative potential among allied agents in a battlefield scenario. Utilizing the Multi-Robot Warehouse Environment(RWARE) as a sparse reward environment, we define analogous problems and establish evaluation criteria. Constructing a learning environment with the QMIX algorithm from the reinforcement learning library Ray RLlib, we enhance the Agent Network of QMIX and integrate Random Network Distillation(RND). This enables the extraction of patterns and temporal features from partial observations of agents, confirming the potential for improving the acquisition of sparse reward experiences through intrinsic rewards.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.244-251
• /
• 2007

Intention reading technique is essential to provide personalized services toward more convenient and human-friendly services in complex ubiquitous environment such as a smart home. If a system has knowledge about an user's intention of his/her behavioral pattern, the system can provide mote qualified and satisfactory services automatically in advance to the user's explicit command. In this sense, learning capability is considered as a key function for the intention reading technique in view of knowledge discovery. In this paper, ore introduce a personalized media control method for a possible application iii a smart home. Note that data pattern such as human behavior contains lots of inconsistent data due to limitation of feature extraction and insufficiently available features, where separable data groups are intermingled with inseparable data groups. To deal with such a data pattern, we introduce an effective engineering approach with the combination of fuzzy logic and probabilistic reasoning. The proposed learning system, which is based on IFCS (Iterative Fuzzy Clustering with Supervision) algorithm, extract probabilistic fuzzy rules effectively from the given numerical training data pattern. Furthermore, an extended architectural design methodology of the learning system incorporating with the IFCS algorithm are introduced. Finally, experimental results of the media contents recommendation system are given to show the effectiveness of the proposed system.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.53-62
• /
• 1994

This paper presents a new approach of analyzing errors resulting from nonideal general-purpose lighting environment when the Photometric Stereo Method (PSM) is applied to estimate the surface-orientation of a three-dimensional object. The approach introduces the explicit modeling of the lighting environment including a circular-disk type irradiance object plane and the direct simulation of the error distribution with the model. The light source is modeled as a point source that has a certain amount of beam angle, and the luminance distribution on the irradiance plane is modeled as a Gaussian function with different deviation values. A simulation algorithm is devised to estimate the light source orientation computing the average luminance intensities obtained from the irradiance object planes positioned in three different orientations. The effect of the nonideal lighting model is directly reflected in such simulation, because of the analogy between the PSM and the proposed algorithm. With an instrumental tool designed to provide arbitrary orientations of the object plane at the origin of the coordinate system, experiment can be performed in a systematic way for the error analysis and compensation. Simulations are performed to find out the error distribution by widely varying the light model and the orientation set of the object plane. The simulation results are compared with those of the experiment performed in the same way as the simulation. It is confirmed from the experiment that a fair amount of errors is due to the erroneous effect of the general-purpose lighting environment.

• Investigative Magnetic Resonance Imaging
• /
• v.11 no.2
• /
• pp.103-109
• /
• 2007

To develop an advanced non-linear curve fitting (NLCF) algorithm for performing dynamic susceptibility contrast study of the brain. The first pass effects give rise to spuriously high estimates of $K^{trans}$ for the voxels that represent the large vascular components. An explicit threshold value was used to reject voxels. The blood perfusion and volume estimation were accurately evaluated in the $T2^*$-weighted dynamic contrast enhanced (DCE)-MR images. From each of the recalculated parameters, a perfusion weighted image was outlined by using the modified non-linear curve fitting algorithm. The present study demonstrated an improvement of an estimation of the kinetic parameters from the DCE $T2^*$-weighted magnetic resonance imaging data with using contrast agents.