이수원 교수

With the development of the Web, users spend more time accessing information that they seek. As a result, recommendation systems have emerged to provide users with preferred contents by filtering abundant information, along with providing means of exposing search results to users more effectively. These recommendation systems operate based on the user reactions to items or on the various user or item features. It is known that recommendation results based on sparse datasets are less reliable bec

Layered lithium manganese oxides suffer from irreversible phase transitions induced by Mn migration and/or dissolution associated with the Jahn-Teller effect (JTE) of Mn³⁺, leading to inevitable capacity fading during cycling. The popular doping strategy of oxidizing Mn³⁺ to Mn⁴⁺ to relieve the JTE cannot completely eliminate the detrimental structural collapse from the cooperative JTE. Therefore, they are considered to be impractical for commercial use as cathod

Few studies classified and predicted hypertension using blood pressure (BP)-related determinants in a deep learning algorithm. The objective of this study is to develop a deep learning algorithm for the classification and prediction of hypertension with BP-related factors based on the Korean Genome and Epidemiology Study-Ansan and Ansung baseline survey. We also investigated whether energy intake adjustment is adequate for deep learning algorithms. We constructed a deep neural network (DNN) in w

Reconstructing a three-dimensional (3D) environment is an indispensable technique to make augmented reality and augmented virtuality feasible. A Kinect device is an efficient tool for reconstructing 3D environments, and using multiple Kinect devices enables the enhancement of reconstruction density and expansion of virtual spaces. To employ multiple devices simultaneously, Kinect devices need to be calibrated with respect to each other. There are several schemes available that calibrate 3D image

A vector field-based guidance law is proposed for the speed and impact angle control of an unpowered, glide-capable air-to-ground munition. An artificial 3-D space is designed for the glider vehicle to satisfy the terminal constraints. The glider vehicle is guided to the target position by the proposed vector field-based guidance law while satisfying the desired impact angle and final speed constraints. In the numerical simulations, various wind conditions are considered to demonstrate the perfo

In this Letter, the authors present a ‘landmark‐free’ clothes recognition approach. Recent studies have shown that the use of landmark information has achieved great success in the task of clothes recognition. However, the landmark annotation is very labour intensive and time consuming. It also suffers from inter‐ and intra‐individual variability. To overcome these problems, the authors propose a two‐branch feature selective network for category classification and attribute prediction. Note that

A homing guidance law against a high-speed target in the exoatmospheric area is proposed based on the vector field approach. The vector field is designed and utilized to achieve head-on hit-to-kill interception. The trajectory of the target is predicted using the gravity acceleration model. The vector-field-based guidance law makes the missile converge to the predicted trajectory of the target, and therefore, the missile can await the target by flying along the predicted trajectory. The shape of

The possibility of integrating binary features into the bag‐of‐features (BoFs) model is explored. The set of binary features extracted from an image are packed into a single vector form, to yield the bag‐of‐binary‐features (BoBFs). The efficient BoBF feature extraction and quantisation provide fast image representation. The trade‐off between accuracy and efficiency in BoBF compared with BoF is investigated through image retrieval tasks. Experimental results demonstrate that BoBF is a competitive

Personalized recommender systems are used not only in e-commerce companies but also in various web applications. These systems conventionally use collaborative filtering (CF) and content-based filtering approaches. CF operates using memory-based or model-based methods; both methods use a user-item matrix that considers user preferences as items. This matrix denotes information on user preferences, which refers to the user ratings for items. The model-based method exploits the fact that the input

ABSTRACT The persistent coverage control problem is formulated based on cell discretisation of two-dimensional mission space and time-increasing cell ages. A new performance function is defined to represent the coverage level of the mission space, and time behaviour is evaluated by the probabilistic method based on the detection model of agents. For comparison, persistent coverage controllers are designed by a target-based approach and a reactive approach. Both controllers are designed in a dist

In order to extend fair queueing algorithms to wireless networks, we propose a channel error and handoff compensation scheme based on a compensation session with a priority swapping mechanism. The proposed compensation scheme provides a short-term fairness guarantee for an error-free session, long-term fairness guarantee for an erroneous session, fast handoff and traffic-specific control.