1,551 research outputs found
확률적 신경망을 이용한 예측치 결합 모형
No full textMany studies have focused considerable attention on choosing a model that represents the underlying process of a time series and on using that model to forecast the future. In the real applications, however, there may be cases in which a single model cannot represent all relevant characteristics of the original time series. In such circumstances, combining the forecasts from several models may yield better performance. The most popular methods for combining forecasts involve taking a weighted average of multiple forecasts. These weights, however, are usually unstable. When the assumptions of normality and unbiasedness of forecast errors are satisfied, a Bayesian method can be used to update the weights. In applications, however, there are many circumstances in which the Bayesian method is not appropriate. This paper proposes a PNN (Probabilistic Neural Network) approach to combining forecasts that can be applied when the assumptions of normality or unbiasedness of the forecast errors are not satisfied. The PNN method has traditionally been used in pattern recognition. It is similar to the Bayesian approach and we suggest its use as an updating method for unstable weights when combining forecasts. Unlike the Bayesian approach, it does not require the assumption of a specific prior distribution because it estimates the probability distribution from given data. Empirical results reveal that the PNN method offers superior predictive capabilities
Detection of aminoglycosidic antibiotics using amino-functionalized polydiacetylene sensor
No full text
A study on the adhesion and brightness of electrophoretically deposited ZnS:Ag, Cl phosphor by heat treatment condition
No full textMagnetic properties of gamma-Fe2O3 nanoparticles made by coprecipitation method
No full textWe have synthesized maghemite (gamma-Fe2O3) nanoparticles using chemical coprecipitation technique through a typical pipette drop method (pipette diameter: 2000 mum) and a piezoelectric nozzle method (nozzle size: 50 pm). The size distribution of the maghemite nanoparticles prepared by the pipette drop method is from 5 nm to 8 nm. However, the nanoparticles made by the piezoelectric nozzle method show smaller size and very narrow size distribution from 3 nn to 5 nm. Zero-Field-Cooled (ZFC)/Field-Cooled (FC) magnetization and magnetic hysteresis measurement were performed using a superconducting quantum interference device (SQUID) magnetometer from 5 K to 300 K to investigate the magnetic properties of nanoparticles. The SQUID measurements revealed superparamagnetism of nanoparticles with the blocking temperature of 119.5 K and 94.3 K for the nanoparticles made by the pipette drop method and the piezoelectric nozzle method, respectively. (C) 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.This work was supported by the Korean Ministry of Science & Technology through the
Creative Research Initiatives Project and Strategic National R&D Program
Criteria-Aware Graph Filtering: Extremely Fast Yet Accurate Multi-Criteria Recommendation
No full textMulti-criteria (MC) recommender systems, which utilize MC rating information for recommendation, are increasingly widespread in various e-commerce domains. However, the MC recommendation using training-based collaborative filtering, requiring consideration of multiple ratings compared to single-criterion counterparts, often poses practical challenges in achieving state-of-the-art performance along with scalable model training. To solve this problem, we propose CA-GF, a training-free MC recommendation method, which is built upon criteria-aware graph filtering for efficient yet accurate MC recommendations. Specifically, first, we construct an item–item similarity graph using an MC user-expansion graph. Next, we design CA-GF composed of the following key components, including 1) criterion-specific graph filtering where the optimal filter for each criterion is found using various types of polynomial low-pass filters and 2) criteria preference-infused aggregation where the smoothed signals from each criterion are aggregated. We demonstrate that CA-GF is (a) efficient: providing the computational efficiency, offering the extremely fast runtime of less than 0.2 seconds even on the largest benchmark dataset, (b) accurate: outperforming benchmark MC recommendation methods, achieving substantial accuracy gains up to 24% compared to the best competitor, and (c) interpretable: providing interpretations for the contribution of each criterion to the model prediction based on visualizations
Magnetic properties of Fe3O4 nanoparticles encapsulated with poly(D,L lactide-Co-glycolide)
No full textWe have investigated the magnetic properties of Fe3O4 nanoparticles encapsulated with poly(D,L lactide-Co-glycolide) (PLGA) prepared by an emulsification-diffusion technique. The size of nanoparticles was reduced down to 90 nm through the optimization of the preparation condition such as homogenizer and agitator speed. Zero-field-cooled (ZFC)/field-cooled (FC) magnetization, magnetic hysteresis, and relaxation measurement were performed using a superconducting quantum interference device (SQUID) magnetometer from 5 K to 300 K to investigate the magnetic properties of nanoparticles. The SQUID measurements show superparamagnetism of nanoparticles with a blocking temperature of 120 K. By measuring the magnetic relaxation of the magnetization at 5 K, we obtained magnetic viscosity of PLGA-encapsulated Fe3O4 nanoparticles
Magnetic properties of superparamagnetic gamma-Fe2O3 nanoparticles prepared by coprecipitation technique
No full textgamma-Fe2O3 nanoparticles have been synthesized by a chemical coprecipitation technique using the typical pipette drop method and the novel piezoelectric nozzle method. Nanoparticles made by the piezoelectric nozzle method show smaller size and very narrow size distribution compared to the nanoparticles produced by the pipette drop method. The superconducting quantum interference device (SQUID) measurements show superparamagnetism of nanoparticles and reveal that the anisotropy constants of the nanoparticles made by the pipette drop method and the piezoelectric nozzle method are 2.2 x 10(6) and 9.0 x 10(6) erg/cm(3), respectively. By measuring the magnetic relaxation of the magnetization at 5 K, we also obtained magnetic viscosity of gamma-(FeO3)-O-2 nanoparticles. (C) 2004 Elsevier B.V. All rights reserved.This work was supported by Ministry of Science
and Technology through the Creative Research
Initiatives Project and Center for Ultramicrochem-
ical Process Systems Project. The authors would
like to express sincere thanks to Dr. Sang-Jun Oh
in KBSI for helping in SQUID measurements
확률적 신경망을 이용한 예측치 결합 모형
No full text학위논문(석사)- 한국과학기술원 : 경영공학전공, 2001.8, [ v, 46 p. ]Many studies have focused considerable attention on choosing a model that represents the underlying process of a time series and on using that model to forecast the future. In the real applications, however, there may be cases in which a single model cannot represent all relevant characteristics of the original time series. In such circumstances, combining the forecasts from several models may yield better performance. The most popular methods for combining forecasts involve taking a weighted average of multiple forecasts. These weights, however, are usually unstable. When the assumptions of normality and unbiasedness of forecast errors are satisfied, a Bayesian method can be used to update the weights. In applications, however, there are many circumstances in which the Bayesian method is not appropriate. This paper proposes a PNN (Probabilistic Neural Network) approach to combining forecasts that can be applied when the assumptions of normality or unbiasedness of the forecast errors are not satisfied.
The PNN method has traditionally been used in pattern recognition. It is similar to the Bayesian approach and we suggest its use as an updating method for unstable weights when combining forecasts. Unlike the Bayesian approach, it does not require the assumption of a specific prior distribution because it estimates the probability distribution from given data. Empirical results reveal that the PNN method offers superior predictive capabilities.한국과학기술원 : 경영공학전공
sj-docx-1-tag-10.1177_17562848231154103 – Supplemental material for Real-life effectiveness and safety of tofacitinib treatment in patients with ulcerative colitis: a KASID multicenter cohort study
No full textSupplemental material, sj-docx-1-tag-10.1177_17562848231154103 for Real-life effectiveness and safety of tofacitinib treatment in patients with ulcerative colitis: a KASID multicenter cohort study by Seung Hwan Shin, Kyunghwan Oh, Sung Noh Hong, Jungbok Lee, Shin Ju Oh, Eun Soo Kim, Soo-Young Na, Sang-Bum Kang, Seong-Joon Koh, Ki Bae Bang, Sung-Ae Jung, Sung Hoon Jung, Kyeong Ok Kim, Sang Hyoung Park, Suk-Kyun Yang, Chang Hwan Choi and Byong Duk Ye in Therapeutic Advances in Gastroenterology</p
- …
