• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.707-720
• /
• 2021

To improve the accuracy of the recommendation system, multi-criteria recommendation systems have been widely researched. However, it is highly complicated to extract the preferred features of users and items from the data. To this end, subjective indicators, which indicate a user's priorities for personalized recommendations, should be derived. In this study, we propose a method for generating recommendation candidates by predicting multi-criteria ratings from reviews and using them to derive user priorities. Using a deep learning model based on convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM), multi-criteria prediction ratings were derived from reviews. These ratings were then aggregated to form a linear regression model to predict the overall rating. This model not only predicts the overall rating but also uses the training weights from the layers of the model as the user's priority. Based on this, a new score matrix for recommendation is derived by calculating the similarity between the user and the item according to the criteria, and an item suitable for the user is proposed. The experiment was conducted by collecting the actual "TripAdvisor" dataset. For performance evaluation, the proposed method was compared with a general recommendation system based on singular value decomposition. The results of the experiments demonstrate the high performance of the proposed method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.8A
• /
• pp.1099-1106
• /
• 1999

In this paper, the bit error ratio(BER) performance of multi-carrier CDMA system in multi-user/multi-path environment is analytically derived, and effect of interferences due to multi-user, multi-carrier, multi-path is analyzed. A multi-carrier/multi-user/multi-path joint detection scheme incorporating a decorrelating detector is proposed, and its performance as well as the enhancement of noise due to decorrelating process are analyzed. BER of the proposed joint detection scheme is only slightly degraded compared to the ideal case of single user environment.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.240-242
• /
• 2004

A channel allocation algorithm of multi-user OFDM(orthogonal frequency division multiplexing) system is presented. The proposed algorithm is to reduce the complexity of the system, using the GNN(gradual neural network) with gradual expansion scheme and the algorithm attempts to allocate channel with good channel gain to each user. The method has lower computational complexity and less iteration than other algorithms.

• Current Optics and Photonics
• /
• v.2 no.1
• /
• pp.39-46
• /
• 2018

Visible-light communication (VLC) combined with advanced illumination can be expected to become an integral part of next-generation communication networks. One of the major concerns in VLC implementation is developing resource-allocation schemes in a multi-user scenario. However, the scheduling for heterogeneous quality of service (QoS) traffic has not been studied so far, for the indoor VLC downlink system. In this paper, we creatively introduce effective-bandwidth and effective-capacity theory into the multi-user scheduling (MUS) problem, to guarantee the user's statistical delay QoS. We also take account of the aggregate interference (AI) in the indoor VLC downlink system, and analyze its impact on the user-centric MUS problem for the first time. Simulations show that the AI has a nonnegligible influence on the scheduling result, and that the proposed scheduling scheme could guarantee the user's QoS requirement under the premise of ensuring sum capacity.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.5
• /
• pp.712-717
• /
• 2007

Using home network system including home server, home service robot, a variety of device, it is generally known that application of Multi Agent System for performing variously distributed process that can be occur in home environment, is efficient method. In this system, it is intelligent service robot that a key of human interface and physical service offer Therefore, using application of established multi agent system, we can defined Multi Agent Robot System. In 'open' home environment, between all agent data interaction and cooperation are needed for Multi Agent System offer to user that more efficient service. For this, we focus our attention on define as agent that can autonomic drive and offer to user that physical service robots and, design, suggest the simulator can display that between robot agents communication or between other agents, like home server, and robot agents communication information to user interface.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.10 no.3
• /
• pp.11-28
• /
• 2000

In these days it is rapidly increasing that multi-user, multi-layered commercial software developments for companies or public institutions. Security services are necessary for most of systems and the access control service is the essential of security services. Current access control methods that are used as access control policies are classified as Discretionary Access Control Mandatory Access Control and Role Based Access Control. However there are some inefficiencies when those methods are applied to current multi-user, multi-layered systems. Therefore it is required that a new access control method that takes complex system resources into account from the side of policy. In this paper extending previous Role Based of 'Behavior' and a basic model of the method. And we simply implement the method on the mobile agent based workflow management system that is a representative example of multi-user. multi-layered softwares and shows implementation results to tap possibilities of real-world application.

• Journal of Mechanical Science and Technology
• /
• v.19 no.10
• /
• pp.1864-1874
• /
• 2005

This paper describes a wearable multi-modal user interface design and its implementation for a teleoperated field robot system. Recently some teleoperated field robots are employed for hazard environment applications (e.g. rescue, explosive ordnance disposal, security). To complete these missions in outdoor environment, the robot system must have appropriate functions, accuracy and reliability. However, the more functions it has, the more difficulties occur in operation of the functions. To cope up with this problem, an effective user interface should be developed. Furthermore, the user interface is needed to be wearable for portability and prompt action. This research starts at the question: how to teleoperate the complicated slave robot easily. The main challenge is to make a simple and intuitive user interface with a wearable shape and size. This research provides multi-modalities such as visual, auditory and haptic sense. It enables an operator to control every functions of a field robot more intuitively. As a result, an EOD (explosive ordnance disposal) demonstration is conducted to verify the validity of the proposed wearable multi-modal user interface.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.1
• /
• pp.79-87
• /
• 1999

For its simplicity and cost effectiveness in implementation, the Inverted DGPS is well suited for some specific applications like automatic vehicle location systems, where the monitoring station needs accurate position of the vehicles in the street. In the Inverted DGPS, the user sends its GPS position and associated satellite informations to the reference station, and the corrections are made at the reference station to get differentially corrected user position. A fundamental requirement in IDGPS is that the user and the reference station must use the same satellites when the corrections are made. But in practice, it is not often satisfied. Inverted DGPS is also suffered from performance degradation as the baseline become large like DGPS. IDGPS system using multi-reference station can resolve this problem. In this paper a simple multi-reference IDGPS algorithm is proposed and some experiments and analysis are peformed. Experiment results show that IDGPS can achieve the positioning performance as accurate as the DGPS can provide.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.8
• /
• pp.12-18
• /
• 2012

In this paper, we propose an improved analysis of transmission power of BTHP applied into the downlink multi-user (MU) multi-input multi-output (MIMO) system. On the contrary to the conventional analysis that adopts the strong interference assumption for every users in the system, the proposed analysis approximates the characteristics of the actual interference components so that provides more accurate approximation of the transmission power than that from the conventional analysis. By computer simulations, it is observed that the proposed approximation is more accurate than the conventional one, especially in the case of 4-QAM modulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.10
• /
• pp.1379-1385
• /
• 2018

In the conventional multi-user superposition transmission, the power allocation coefficients of data symbols and the received signal processing of users are determined by the condition of instantaneous channel powers. However, the use of instantaneous channel powers can increase the system complexity. Hence, we consider fixed multi-user superposition transmission using average channel powers. The fixed multi-user superposition transmission can reduce the system complexity because it uses the condition of average channel powers that slowly change over time in order to decide the power allocation coefficients and the received signal processing. In this paper, we analyze the average symbol error rate for the fixed multi-user superposition transmission. In particular, an expression for the average symbol error rate of M-ary Quadrature Amplitude Modulation is derived assuming Rayleigh fading channels. In addition, through the numerical results, we show that the conventional and fixed multi-user superposition transmissions achieve the similar average symbol error rate performances at the user in the severe channel condition.