• Journal of Information Technology Applications and Management
• /
• v.26 no.6
• /
• pp.89-101
• /
• 2019

Many researchers have been focused on designing beauty product recommendation system for a long time because of increased need of customers for personalized and customized recommendation in beauty product domain. In addition, as the application of the deep neural network technique becomes active recently, various collaborative filtering techniques based on the deep neural network have been introduced. In this context, this study proposes a deep neural network model suitable for beauty product recommendation by applying Neural Collaborative Filtering and Generalized Matrix Factorization (NCF + GMF) to beauty product recommendation. This study also provides an implementation of web API system to commercialize the proposed recommendation model. The overall performance of the NCF + GMF model was the best when the beauty product recommendation problem was defined as the estimation rating score problem and the binary classification problem. The NCF + GMF model showed also high performance in the top N recommendation.

• Journal of the Microelectronics and Packaging Society
• /
• v.29 no.2
• /
• pp.81-89
• /
• 2022

The in-situ electromigration(EM) and annealing test were performed at 110, 130, and 150℃ with a current density of 1.3×10⁵ A/cm² conditions to investigate the effect of non-conductive film (NCF) on growth kinetics of intermetallic compound (IMC) in Cu/Ni/Sn-2.5Ag microbump. As a result, the activation energy of the Ni₃Sn₄ IMC growth in the annealing and EM conditions according to the NCF application was about 0.52 eV, and there was no significant difference. This is because the growth rate of Ni-Sn IMC is much slower than that of Cu-Sn IMC, and the growth behavior of Ni-Sn IMC increases linearly with the square root of time, so it has the same reaction mechanism dominated by diffusion. In addition, there is no difference in the activation energy of the Ni₃Sn₄ IMC growth because the EM resistance effect of the back stress according to the NCF application is not large.

• Restorative Dentistry and Endodontics
• /
• v.42 no.4
• /
• pp.301-308
• /
• 2017

Objectives: It was aimed to compare the cyclic fatigue resistances of ProTaper Universal (PTU), ProTaper Next (PTN), and ProTaper Gold (PTG) and the effects of sterilization by autoclave on the cyclic fatigue life of nickel-titanium (NiTi) instruments. Materials and Methods: Eighty PTU, 80 PTN, and 80 PTG were included to the present study. Files were tested in a simulated canal. Each brand of the NiTi files were divided into 4 subgroups: group 1, as received condition; group 2, pre-sterilized instruments exposed to 10 times sterilization by autoclave; group 3, instruments tested were sterilized after being exposed to 25%, 50%, and 75% of the mean cycles to failure, then cycled fatigue test was performed; group 4, instruments exposed to the same experiment with group 3 without sterilization. The number of cycles to failure (NCF) was calculated. The data was statistically analyzed by using one-way analysis of variance and post hoc Tukey tests. Results: PTG showed significantly higher NCF than PTU and PTN in group 1 (p < 0.05). Sterilization significantly increased the NCF of PTN and PTG (p < 0.05) in group 2. PTN in group 3 had significantly higher cyclic fatigue resistance than PTN group 4 (p < 0.05). Also, significantly higher NCF was observed for PTG in group 2 than in groups 3 and 4 (p < 0.05). Conclusions: PTG instrument made of new gold alloy was more resistant to fatigue failure than PTN and PTU. Autoclaving increased the cyclic fatigue resistances of PTN and PTG.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.18 no.1
• /
• pp.59-64
• /
• 2019

In recent years, the HP-CRTM method, which has the ability to produce carbon fiber-reinforce plastic composites at high speeds, has come into the spotlight in the automotive parts industry, which demands high productivity. Multi-axial carbon fabric, an intermediate material used in this HP-CRTM molding process, consists of layered fibers without crimp, which makes it better in terms of tensile and shear strength than the original woven fabrics. The NCF (non-crimp fabric) can form the layers of the carbon fiber, which have different longitudinal and lateral directions, and ${\pm}{\theta}$ degrees, depending on the product's properties. In this research, preforms were made with carbon fibers of ${\pm}45^{\circ}$ and $0/90^{\circ}$, which were lamination structures under seven different conditions, in order to create the optimal laminated structure for automobile reinforcement center floor tunnels. Carbon fiber composites were created using each of the seven differently laminated preforms, and polyurethane was used as the base material. The specimens were manufactured in accordance with the ASTM D3039 standards, and the effect of the NCF lamination structure on the mechanical properties was confirmed by a tensile test.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.173-178
• /
• 2022

Recently, a recommendation system using deep learning in social network services has been actively studied. However, in the case of a recommendation system using deep learning, the cold start problem and the increased learning time due to the complex computation exist as the disadvantage. In this paper, the user-tailored exercise routine recommendation algorithm is proposed using the user's metadata. Metadata (the user's height, weight, sex, etc.) set as the input of the model is applied to the designed model in the proposed algorithms. The exercise recommendation system model proposed in this paper is designed based on the neural collaborative filtering (NCF) algorithm using multi-layer perceptron and matrix factorization algorithm. The learning proceeds with proposed model by receiving user metadata and exercise information. The model where learning is completed provides recommendation score to the user when a specific exercise is set as the input of the model. As a result of the experiment, the proposed exercise recommendation system model showed 10% improvement in recommended performance and 50% reduction in learning time compared to the existing NCF model.

• Journal of Society of Preventive Korean Medicine
• /
• v.28 no.1
• /
• pp.109-118
• /
• 2024

Objectives : This study aimed to investigate the relationship between cold feet and sleep quality using polysomnography (PSG) and Pittsburgh Sleep Quality Index (PSQI). Methods : We divided 11 adults (6 females, 5 males) with Insomnia Severity Index score below 21 into cold feet (CF) and a non-cold feet (NCF) group based on the median feet temperature (Taichong, LR3). PSG and PSQI were administered to assess sleep characteristics and subjective sleep quality. Results : CF group exhibited significantly lower time in bed, sleep period time, and total sleep time compared to NCF group. While there were no significant group differences in sleep latency, wakefulness after sleep onset, or total arousal index, NCF group had significantly lower minimum oxygen saturation and apnea-hypopnea index in REM (rapid eye movement) sleep compared to CF group. Although the PSQI score and the proportion of poor sleepers were both higher in the CF group (7.40 and 80%) compared to the NCF group (5.50 and 50%), these differences did not reach statistical significance. Conclusions : This study showed that foot temperature affects sleep characteristics and suggests the need to utilize PSG in sleep research in Korean medicine.

• Proceedings of the Korean Fiber Society Conference
• /
• 1993.06b
• /
• pp.237-242
• /
• 1993

• Restorative Dentistry and Endodontics
• /
• v.38 no.1
• /
• pp.31-35
• /
• 2013

Objectives: This study compared the cyclic fatigue resistance of nickel-titanium (NiTi) files obtained in a conventional test using a simulated canal with a newly developed method that allows the application of constant fatigue load conditions. Materials and Methods: ProFile and K3 files of #25/.06, #30/.06, and #40/.04 were selected. Two types of testing devices were built to test their fatigue performance. The first (conventional) device prescribed curvature inside a simulated canal (C-test), the second new device exerted a constant load (L-test) whilst allowing any resulting curvature. Ten new instruments of each size and brand were tested with each device. The files were rotated until fracture and the number of cycles to failure (NCF) was determined. The NCF were subjected to one-way ANOVA and Duncan's post-hoc test for each method. Spearman's rank correlation coefficient was computed to examine any association between methods. Results: Spearman's rank correlation coefficient (${\rho}$ = -0.905) showed a significant negative correlation between methods. Groups with significant difference after the L-test divided into 4 clusters, whilst the C-test gave just 2 clusters. From the L-test, considering the negative correlation of NCF, K3 gave a significantly lower fatigue resistance than ProFile as in the C-test. K3 #30/.06 showed a lower fatigue resistance than K3 #25/.06, which was not found by the C-test. Variation in fatigue test methodology resulted in different cyclic fatigue resistance rankings for various NiTi files. Conclusions: The new methodology standardized the load during fatigue testing, allowing determination fatigue behavior under constant load conditions.

• Knowledge Management Research
• /
• v.21 no.3
• /
• pp.27-44
• /
• 2020

In order to solve the user's information overload problem, recommender systems infer users' preferences and suggest items that match them. The collaborative filtering (CF), the most successful recommendation algorithm, has been improving performance until recently and applied to various business domains. Visual information, such as book covers, could influence consumers' purchase decision making. However, CF-based recommender systems have rarely considered for visual information. In this study, we propose VizNCS, a CF-based deep learning model that uses visual information as additional information. VizNCS consists of two phases. In the first phase, we build convolutional neural networks (CNN) to extract visual features from image data. In the second phase, we supply the visual features to the NCF model that is known to easy to extend to other information among the deep learning-based recommendation systems. As the results of the performance comparison experiments, VizNCS showed higher performance than the vanilla NCF. We also conducted an additional experiment to see if the visual information affects differently depending on the product category. The result enables us to identify which categories were affected and which were not. We expect VizNCS to improve the recommender system performance and expand the recommender system's data source to visual information.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2009.10a
• /
• pp.689-690
• /
• 2009