• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.7
• /
• pp.1868-1887
• /
• 2024

Recommendation systems research is a subfield of information retrieval, as these systems recommend appropriate items to users during their visits. Appropriate recommendation results will help users save time searching while increasing productivity at work, travel, or shopping. The problem becomes more difficult when the items are geographical locations on the ground, as they are associated with a wealth of contextual information, such as geographical location, opening time, and sequence of related locations. Furthermore, on social networking platforms that allow users to check in or express interest when visiting a specific location, their friends receive this signal by spreading the word on that online social network. Consideration should be given to relationship data extracted from online social networking platforms, as well as their impact on the geolocation recommendation process. In this study, we compare the similarity of geographic locations based on their distance on the ground and their correlation with users who have checked in at those locations. When calculating feature embeddings for users and locations, social relationships are also considered as attention signals. The similarity value between location and correlation between users will be exploited in the overall architecture of the recommendation model, which will employ graph convolution networks to generate recommendations with high precision and recall. The proposed model is implemented and executed on popular datasets, then compared to baseline models to assess its overall effectiveness.

• Journal of computational fluids engineering
• /
• v.15 no.2
• /
• pp.79-85
• /
• 2010

Aerodynamic design optimization of rotor airfoil has been performed with advanced design method for improved aerodynamic characteristics of ONERA airfoils. A multiple response surface method is used to consider various requirements in rotor airfoil design. Shape functions for mean camber line are proposed to extend possible design domain. Numerical simulations are performed using KFLOW, a Navier-Stokes solver with shear stress transport turbulence model. The present design method provides favorable configurations for the high performance rotor airfoil. Resulting optimized airfoils give better aerodynamic performance than the baseline airfoils.

• New & Renewable Energy
• /
• v.7 no.4
• /
• pp.50-57
• /
• 2011

A wind turbine simulation program for the coupled dynamics of aerodynamics, elasticity, multi-body dynamics and controls of turbine is newly developed by combining an aero-elastic code and a multi-body dynamics code. The aero-elastic code, based on the blade momentum theory and generalized dynamic wake theory, is developed by NREL(National Renewable Energy Laboratory, USA). The multi-body dynamics code is commercial one which is capable of accounting for geometric nonlinearity and twist deflection. A turbulent wind load case is simulated for the NREL 5-MW baseline wind turbine model by the developed program and FAST. As a result, the two results agree well enough to verify the reliability of the developed program.

• 한국전산유체공학회:학술대회논문집
• /
• 2009.04a
• /
• pp.51-56
• /
• 2009

Optimization with metamodel is one of numerical optimization methods. Response surface method is performed for making metamodel. The Hcks-Henne function is used for designing 2D shape of the airfoil and spring analogy is used to change the grid according to the change in shape of the airfoil. Aerodynamic coefficient required for response surface method are obtained by using Navier-Stokes solver with $\kappa-\omega$ shear stress transport turbulence model. For the baseline airfoils, OA 312, OA 309, and OA 407 airfoils select and optimize to improve aerodynamic performance.

• MALSORI
• /
• v.55
• /
• pp.103-118
• /
• 2005

In this paper, we describe a morpheme-based pronunciation lexicon useful for Korean LVCSR. The phonemic-context-dependent multiple pronunciation lexicon improves the recognition accuracy when cross-morpheme pronunciation variations are distinguished from within-morpheme pronunciation variations. Since adding all possible pronunciation variants to the lexicon increases the lexicon size and confusability between lexical entries, we have developed a lexicon pruning scheme for optimal selection of pronunciation variants to improve the performance of Korean LVCSR. By building a proposed pronunciation lexicon, an absolute reduction of $0.56\%$ in WER from the baseline performance of $27.39\%$ WER is achieved by cross-morpheme pronunciation variations model with a phonemic-context-dependent multiple pronunciation lexicon. On the best performance, an additional reduction of the lexicon size by $5.36\%$ is achieved from the same lexical entries.

• Korean System Dynamics Review
• /
• v.9 no.2
• /
• pp.5-25
• /
• 2008

This paper analysed the impact of return policy as a coordination mechanism for decentralized supply chain with one capacitated supplier and two competing retailers under random demand distribution. In this study, Shortage gaming also considered to reflect a competing environment of two retailers. System dynamics approach was used to model the baseline two echelon supply chain and return policy on it. Given each of 4 experiment settings being used for 100 simulations with different random seeds, 400 random samples were used in a t-test. The result show that return policy significantly enhance the supply chain profits and fillrates. The analysis suggest that the supply chain performance can be build up by implementing a return policy even though under consideration of a capacitated supplier and competing retailers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1704-1720
• /
• 2020

Recognizing food from photographs presents many applications for machine learning, computer vision and dietetics, etc. Recent progress of deep learning techniques accelerates the recognition of food in a great scale. We build a hierarchical structure composed of deep CNN to recognize and classify food from photographs. We build a dataset for Korean food of 18 classes, which are further categorized in 4 major classes. Our hierarchical recognizer classifies foods into four major classes in the first step. Each food in the major classes is further classified into the exact class in the second step. We employ DenseNet structure for the baseline of our recognizer. The hierarchical structure provides higher accuracy and F1 score than those from the single-structured recognizer.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.139-142
• /
• 2004

훈련 과정과 인식 과정사이의 주변 잡음과 채널 특성으로 인한 환경의 불일치는 음성 인식 성능을 급격히 저하시킨다. 이러한 차이를 극복하기 위해 다양한 전처리 방법이 제안되어 왔으며, 최근에는 스테레오 데이터와 잡음 음성의 Gaussian Mixture Model(GMM)을 이용하여 보상벡터를 구하는 SPLICE 방법이 좋은 성능을 보여주고 있다. 하지만 차원별로 특징벡터를 보상해주는 추정된 보상벡터는 underestimation되는 경향이 있으며, 그 정도가 각각의 차원마다 달라짐이 관찰되었다. 본 논문에서는 SPLICE 방법에 기반하여 추정된 보상벡터와 실제 보상벡터 사이의 관계를 관찰하여 차원별로 다른 가중치를 적용하는 차원별 가중 보상 방법을 제안하였다. 제안한 방법은 Aurora2 Clean-condition인 경우 baseline 실험 결과에 비해 $68\%$의 높은 상대적인 인식 향상율을 얻었다.

• MALSORI
• /
• no.44
• /
• pp.105-113
• /
• 2002

In Korean, each digit is monosyllable and some pairs are known to have high confusability, causing performance degradation of connected digit recognition systems. To improve the performance, in this paper, we employ various discriminant analyses (DA) including Linear DA (LDA), Weighted Pairwise Scatter LDA WPS-LDA), Heteroscedastic Discriminant Analysis (HDA), and Maximum Likelihood Linear Transformation (MLLT). We also examine several combinations of various DA for additional performance improvement. Experimental results show that applying any DA mentioned above improves the string accuracy, but the amount of improvement of each DA method varies according to the model complexity or number of mixtures per state. Especially, more than 20% of string error reduction is achieved by applying MLLT after WPS-LDA, compared with the baseline system, when class level of DA is defined as a tied state and 1 mixture per state is used.

• MALSORI
• /
• no.57
• /
• pp.165-174
• /
• 2006

In this paper we use a training algorithm, MCE (Minimum Classification Error), to improve the performance of a text-independent speaker identification system. The MCE training scheme takes account of possible competing speaker hypotheses and tries to reduce the probability of incorrect hypotheses. Experiments performed on a small set speaker identification task show that the discriminant training method using MCE can reduce identification errors by up to 54% over a baseline system trained using Bayesian adaptation to derive GMM (Gaussian Mixture Models) speaker models from a UBM (Universal Background Model).