• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.1
• /
• pp.125-134
• /
• 2021

With the rapid development of the film industry, the number of films is significantly increasing and movie recommendation system can help user to predict the preferences of users based on their past behavior or feedback. This paper proposes a movie recommendation system based on the latent factor model with the adjustment of mean and bias in rating. Singular value decomposition is used to decompose the rating matrix and stochastic gradient descent is used to optimize the parameters for least-square loss function. And root mean square error is used to evaluate the performance of the proposed system. We implement the proposed system with Surprise package. The simulation results shows that root mean square error is 0.671 and the proposed system has good performance compared to other papers.

• Journal of the KSME
• /
• v.22 no.6
• /
• pp.456-462
• /
• 1982

재래의 공학적 해석과정에서는, 주어진 시스템에 대한 정량적 해석을 위해 우선 시스템에 대한 수학적 모형을 가정.도입하고, 미지계수에 대한 결정은 시스템으로부터 관측된 자료에 의존하게 된다. 반면 DDS에 의한 모형화는 시스템에 대한 사전 지식이 없어도 시스템으로부터의 자료에 대한 체계적인 정량적 해석을 통해서 시스템을 특성화시키는 미분방정식을 유도할 수 있게 하며 오히려 시스템에 대한 이해를 돕게 한다. 이 때 결정되는 모형은 최소오차자승법(least square error method)에 의한 통계학적으로 가장 적합한 근사모형으로 결정된 모형의 계수로부터 시 스템의 물리적 특성을 규명(identify)할 수 없는 경우에도 그 모형으로 표현되는 추상시스템을 바탕으로 특성화, 예측 내지 제어 목적으로 활용할 수 있다. DDS는 단일변수(univariate)및 다 변수(multivariate) 자료에 모두 적용 가능하며 전달 함수 규명(Transfer Function Identification) 및 닫힘 루우프 시스템(closedloop system)의 모형화 및 해석에도 이용되고 있다.

• Journal of the Korean Geotechnical Society
• /
• v.15 no.4
• /
• pp.153-166
• /
• 1999

Since the reliability of results by the existing analyzing method is low, in the case of for excavation performed in urban area whose stability is of great importance, construction control based on field monitoring is always necessary. But the field monitoring reflects only the behavior of construction process that has already been carried out, and it has limitations in predicting the behavior of the expected construction process, which is practically more important for construction control. In this study, construction control system for excavation which can predict the behavior of the expected processes during construction with high degree of accuracy, is developed by adopting inverse analysis. The inverse analied applied field monitoring results to excavation analysis can improve the reliability of predicted results. The developed system uses an elasto-plastic soil spring model for the excavation analysis and the minimization of least squared errors between measured displacements and calculated displacements for the inverse analysis. All the required processes for construction control can be performed as an integrated work within the system reflecting real time application and user's convenience. Their applicabilitis are confirmed by two case studies.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2010.05a
• /
• pp.1274-1278
• /
• 2010

우리나라는 국토의 60% 이상이 산지로 구성되어 있다. 현재 국내에서는 홍수유출 해석 시 집중형 모형을 주로 이용하고 있다. 집중형 모형은 대개 유역 최하류 지점의 유출구를 기준으로 홍수유출 해석 모형의 매개변수 추정 및 검증이 이루어지며, 유역의 매개변수를 소유역별로 동일하게 가정하여 입력 자료를 구성한다. 따라서 산지하천 유역의 홍수유출 해석 및 예측 시 경사가 급하고 고도가 높으며 집중시간이 빠른 산지하천의 지형적 요소 및 특징을 적절히 고려하지 못하여 정확한 예측 및 해석을 하는데 어려움이 발생한다. 분포형 모형은 하나의 유출구가 아닌 임의의 지점에서 홍수유출 해석이 가능하며, 강우자료 입력 시 유역 평균강우가 아닌 분포형 강우, 즉 역거리자승법, 크리깅 기법 등을 사용하여 분포형 강우로 변환한 지점강우와 레이더 강우를 사용하여 보다 정확한 홍수유출 해석이 가능하다. 그리고 분포형 모형은 입력하는 모든 매개변수를 지형 자료에서 추출하여 사용하기 때문에 인공적인 해석을 배제할 수 있어 인위적인 오차를 줄일 수 있다. 본 연구에서는 평창강 상류유역을 시험유역으로 선정하여 연구를 수행하였으며, 분포형 모형의 하나인 $Vflo^{TM}$를 사용하여 홍수유출해석을 수행하였다. 지형자료만을 사용하여 특정 지점이 아닌 유역 내 임의 지점의 홍수유출량과 집중시간, 홍수위를 산정할 수 있어 산지하천에서 돌발적으로 발생하는 홍수를 신속하게 예측할 수 있었다. 또한 임의의 지점에서의 설계홍수량을 손쉽게 산정하여 수공구조물 설계 시 이용할 수 있으므로 홍수에 의한 인적 물적 피해를 최소할 할 수 있을 것으로 기대된다.

• Food Science of Animal Resources
• /
• v.31 no.1
• /
• pp.115-121
• /
• 2011

The objective of this study was to develop models to predict freshness factors (total viable counts (TVC), pH, volatile basic nitrogen (VBN), trimethylamine (TMA), and thiobarbituric acid (TBA) values) and the storage period in beef using a visible and near-infrared (NIR) spectroscopic technique. A total of 216 beef spectra were collected during the storage period from 0 to 14 d at a $10^{\circ}C$ storage. A spectrophotometer was used to measure reflectance spectra from beef samples, and beef freshness spectra were divided into a calibration set and a validation set. Multi-linear regression (MLR) models using the stepwise method were developed to predict the factors. The MLR results showed that beef freshness had a good correlation between the predicted and measured factors using the selected wavelength. The correlation of determination ($r^2$), standard error of prediction (SEP), and ratio of standard deviation to SEP (RPD) of the prediction set for TVC was 0.74, 0.64, and 2.75 Log CFU/$cm^2$, respectively. The $r^2$, SEP, and RPD values for pH were 0.43, 0.10, and 1.10; those for VBN were 0.73, 1.45, and 2.00 mg%; those for TMA were 0.70, 0.19, and 2.58 mg%; those for TBA values were 0.73, 0.13, and 2.77 mg MA/kg; and those for storage period were 0.77, 1.94, and 2.53 d, respectively. The results indicate that visible and NIR spectroscopy can predict beef freshness during storage.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.7
• /
• pp.831-839
• /
• 2012

In this paper, an exponential model of Korea is proposed to minimize an altitude-error of high-altitude target due to atmosphere refraction at radar system. The relation between surface refractivity and refractivity gradient, which is extracted using the least square fit from the measured data at 7 weather stations, is applied to the exponential model. And in order to verify the proposed model, the altitude-errors for a standard atmosphere, a CRPL(Central Radio Propagation Lab.) exponential model, the proposed model are extracted and analyzed using a ray tracing. As a result, the proposed model can improve the altitude estimation performance of radar compared to conventional atmosphere refractive index models.

• Journal of Korea Society of Industrial Information Systems
• /
• v.21 no.5
• /
• pp.11-17
• /
• 2016

In this paper, we propose a method of correcting depth image by the plane modeling and then improving the coding performance. We model a plane by using the least squares method to the horizontal and vertical directions including the target pixel, and then determine that the predicted plane is suitable from the estimate error. After that, we correct the target pixel by the plane mode. The proposed method can correct not only the depth image composed the plane but also the complex depth image. From the simulation result that measures the entropy power, which can estimate the coding performance, we can see that the coding performance by the proposed method is improved up to 80.2%.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.39 no.3
• /
• pp.215-221
• /
• 2015

The elbow reflex, a principal reflex in the upper extremity, plays an important role in the diagnosis of cervical spine syndromes. In this study, the muscle activations of brachial biceps and triceps, and the kinematics of upper extremities were predicted using an activation model for the stretch reflex. The muscle activations that equated the simulation results estimated by the analysis model with the experimental results were obtained first, and the activations obtained from the simulations were compared with the electromyography signals obtained from the experiments, for model validation. The root mean squares error of the joint angles (obtained from experiments and simulation using the suggested model) was 0.056, a value that is half of that obtained using the previous model. This demonstrates that the suggested model corresponded well with the actual reflex.

• Korean Journal of Optics and Photonics
• /
• v.20 no.6
• /
• pp.346-353
• /
• 2009

Glucose and protein in urine are among the important substances for urine analysis and have generally been measured based on a reagent strip test. In this study, these two substances were measured using mid-infrared absorption spectroscopy. Samples were prepared from a commercial synthetic urine product. Glucose and albumin were added as well as red blood cells, which are expected to create the most spectroscopic interference of any substance. Concentrations of these substances were varied independently. Optimal wavelength regions were determined from a partial least squares regression analysis (glucose 980 - 1150/cm, albumin 1400 - 1570/cm). Interference by other substances increased the differences between measured and predicted values. Albumin measurement in particular weres heavily influenced by the presence of glucose and red blood cells. Depending on the inference by other substances, measurement errors were 29.85${\sim}$45.19 mg/dl for a glucose level between 0 and 1000 mg/dl and 14.0${\sim}$93.11 mg/dl for an albumin level of 0 ${\sim}$ 500 mg/dl. Our study proposes an alternative to the chemical test-strip analysis, which shows only discrete concentration levels.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.98-98
• /
• 2017

대부분의 가정과 요식업체, 식품가공업계에서 이용하고 있는 식용유지는 저장 및 가공과정 중에 산패가 빈번하게 일어나게 된다. 기존에는 유지 산패를 측정하기 위해 산가, 과산화물가 등을 측정하는 이화학적인 적정방법을 이용하였는데 실험자의 숙련도에 따라 결과의 오차가 발생할 수 있고, 반복실험으로 인한 시간과 비용이 많이 소모되는 등 여러 제약사항을 포함하고 있어 식용유지의 산패를 실시간 비파괴적으로 분석할 수 있는 기술의 개발에 많은 관심이 모아지고 있다. 따라서, 본 연구에서는 식용유지의 저장조건에 따른 산패정도를 비파괴적으로 평가하기 위한 근적외선 분광분석과 인공신경망 분석기술을 개발하여 그 실효성을 평가하였다. 식물성 식용유지인 들기름을 특정 온도에서 일정한 시간동안 저장하면서 이화학적 적정방법을 통해 산가와 과산화물가를 측정하였으며 동일한 시료의 근적외선 투과스펙트럼을 획득하였다. 수집된 정보를 이용하여 유지 산패 예측 모델을 개발하기 위해 다변량 분석기법 (주성분 회귀분석, 최소자승 회귀분석과 인공신경망 분석)을 적용하였다. 분석 결과, 인공신경망 분석모델이 산가 ($R^2_{tra}:0.9037$, $R^2_{val}:0.8175$, $R^2_{test}:0.8555$)와 과산화물가 ($R^2_{tra}:0.9210$, $R^2_{val}:0.9341$, $R^2_{test}:0.8286$)의 예측 성능이 가장 우수한 것으로 확인되었다. 본 연구의 결과들은 농산물과 식품의 성분 측정뿐만 아니라 다른 산업분야에서도 유용하게 활용될 수 있을 것으로 기대되어진다.