• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.95-95
• /
• 2023

Common hydrological problems of developing countries include poor data management, insufficient measuring devices and ungauged watersheds, leading to small or unreliable data availability. This has greatly affected the adoption of artificial intelligence techniques for flood risk mitigation and damage control in several developing countries. While climate datasets have recorded resounding applications, but they exhibit more uncertainties than ground-based measurements. To encourage AI adoption in developing countries with small ground-based dataset, we propose data augmentation for regression tasks and compare performance evaluation of different AI models with and without data augmentation. More focus is placed on simple models that offer lesser computational cost and higher accuracy than deeper models that train longer and consume computer resources, which may be insufficient in developing countries. To implement this approach, we modelled and predicted streamflow data of the Asa River Watershed located in Ilorin, Kwara State Nigeria. Results revealed that adequate hyperparameter tuning and proper model selection improve streamflow prediction on small water dataset. This approach can be implemented in data-scarce regions to ensure timely flood intervention and early warning systems are adopted in developing countries.

• Asia pacific journal of information systems
• /
• v.33 no.3
• /
• pp.737-767
• /
• 2023

With the availability of big customer data and advances in machine learning techniques, the prediction of customer behavior at the session-level has attracted considerable attention from marketing practitioners and scholars. This study aims to predict customer purchase conversion at the session-level by employing customer profile, transaction, and clickstream data. For this purpose, we develop a multimodal deep learning fusion model with dynamic and static features (i.e., DS-fusion). Specifically, we base page views within focal visist and recency, frequency, monetary value, and clumpiness (RFMC) for dynamic and static features, respectively, to comprehensively capture customer characteristics for buying behaviors. Our model with deep learning architectures combines these features for conversion prediction. We validate the proposed model using real-world e-commerce data. The experimental results reveal that our model outperforms unimodal classifiers with each feature and the classical machine learning models with dynamic and static features, including random forest and logistic regression. In this regard, this study sheds light on the promise of the machine learning approach with the complementary method for different modalities in predicting customer behaviors.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_3
• /
• pp.1209-1219
• /
• 2019

For consistent vegetation monitoring, it is necessary to generate time-series vegetation index datasets at fine temporal and spatial scales by fusing the complementary characteristics between temporal and spatial scales of multiple satellite data. In this study, we quantitatively and qualitatively analyzed the prediction accuracy of time-series change information extracted from spatio-temporal fusion models of multiple satellite data for vegetation monitoring. As for the spatio-temporal fusion models, we applied two models that have been widely employed to vegetation monitoring, including a Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) and an Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM). To quantitatively evaluate the prediction accuracy, we first generated simulated data sets from MODIS data with fine temporal scales and then used them as inputs for the spatio-temporal fusion models. We observed from the comparative experiment that ESTARFM showed better prediction performance than STARFM, but the prediction performance for the two models became degraded as the difference between the prediction date and the simultaneous acquisition date of the input data increased. This result indicates that multiple data acquired close to the prediction date should be used to improve the prediction accuracy. When considering the limited availability of optical images, it is necessary to develop an advanced spatio-temporal model that can reflect the suggestions of this study for vegetation monitoring.

• Korean Journal of Ecology and Environment
• /
• v.46 no.1
• /
• pp.1-9
• /
• 2013

Algal blooms in potable water supplies are becoming an increasingly prevalent and serious water quality problem around the world. In addition to precipitating taste and odor problems, blooms damage the environment, and some classes like cyanobacteria (blue-green algae) release toxins that can threaten human health, even causing death. There is a recognized need in the water industry for models that can accurately forecast in real-time algal bloom events for planning and mitigation purposes. In this study, using data for an interconnected system of rivers and reservoirs operated by a New Jersey water utility, various ANN models, including both discrete prediction and classification models, were developed and tested for forecasting counts of three different algal classes for one-week and two-weeks ahead periods. Predictor model inputs included physical, meteorological, chemical, and biological variables, and two different temporal schemes for processing inputs relative to the prediction event were used. Despite relatively limited historical data, the discrete prediction ANN models generally performed well during validation, achieving relatively high correlation coefficients, and often predicting the formation and dissipation of high algae count periods. The ANN classification models also performed well, with average classification percentages averaging 94 percent accuracy. Despite relatively limited data events, this study demonstrates that with adequate data collection, both in terms of the number of historical events and availability of important predictor variables, ANNs can provide accurate real-time forecasts of algal population counts, as well as foster increased understanding of important cause and effect relationships, which can be used to both improve monitoring programs and forecasting efforts.

• Journal of the Korea Safety Management & Science
• /
• v.4 no.2
• /
• pp.11-22
• /
• 2002

In this study, an asymmetric lifting posture prediction model was developed, which was a three-dimensional model with 12 links and 23 degrees of freedom open kinematic chains. Although previous researchers have proposed biomechanical, psychophysical, or physiological measures as cost functions, for solving redundancy, they lack in accuracy in predicting actual lifting postures and most of them are confined to the two-dimensional model. To develop an asymmetric lifting posture prediction model, we used the resolved motion method for accurately simulating the lifting motion in a reasonable time. Furthermore, in solving the redundant problem of the human posture prediction, a moment weighted Joint Range Availability (JRA) was used as a cost function in order to consider dynamic lifting. However, it is known that the moment weighted JRA as a cost function predicted the lower extremity and L5/S1 joint motions better than the upper extremities, while the constant weighted JRA as a cost function predicted the latter better than the former. To compensate for this, we proposed a hybrid moment weighted JRA as a new cost function with moment weighted for only the lower extremity. In order to validate the proposed cost function, the predicted and real lifting postures for various lifting conditions were compared by using the root mean square(RMS) error. This hybrid JRA reduced RMS more than the previous cost functions. Therefore, it is concluded that the cost function of a hybrid moment weighted JRA can be used to predict three-dimensional lifting postures. To compare with the predicted trajectories and the real lifting movements, graphical validations were performed. The results also showed that the hybrid moment weighted cost function model was found to have generated the postures more similar to the real movements.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.6
• /
• pp.763-774
• /
• 2023

The water availability in a river is related to the return flow of residential water. However it is still difficult to determine the exact return flow. In this study, the residential water-cycle system is defined as a process consisting of water inflow, water transfer and water outflow. The study area is Hampyeong-gun, Jeollanam-do, and is set as a single inflow to a single outflow through the water-cycle system after classification of complete and incomplete measurement points. The time-series prediction models(ARIMA model and TFM) are established with daily inflow and outflow data for 6 years. Inflow and outflow are predicted by dividing into training and test periods. As a result, both models show the feasibility of short-term prediction by deriving stable residuals and securing statistical significance, implementing the preliminary form of the water-cycle system. As a further study, it is suggested to predict the actual return flow of the target basin and efficient water operation by adding input factors and selecting the optimal model.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1196-1200
• /
• 2003

The degree of supersaturation is an important measure for the operation of crystallization processes, because it is directly related to the control of crystal size distribution and shape. A conventional technique utilizing solution composition and temperature has a variety of problems caused from the measurement error and the handling of analyzing samples. A monitoring system of the supersaturation using a quartz crystal sensor is proposed here, and its performance is examined applying different manipulations of coolant temperature. The experimental outcome and photographic examination indicate that the measurements of resonant frequency and resistance of the sensor can be used for the prediction of the formation and growth of solid crystal from the crystallization process. The monitoring system eliminates the intrinsic error source of the conventional system to give the improved measurement and on-line application availability.

• The Magazine of the IEIE
• /
• v.18 no.2
• /
• pp.46-54
• /
• 1991

현대에 개발되는 시스팀은 고성능, 다기능이 요구되며 그 요구가 사용자 관점에서 볼 때 고신뢰성이 한층 요구 되어지고 있다. 연구 개발 단계의 초기 단계에서 설정된 품질 목표치, 즉 신뢰도(reliability), 가용도(availability), 정비도(mailtailability)를 기준으로 연구 개발 단계별로 품질 및 신뢰성 활동들이 구체화 된다. 특정 시스팀이 하나의 building block의 개념으로 구체화 되고 시스팀의 계층 구조를 물리적 구조로 분류되는 시스팀 설계 단계에서 시스팀 블록 단위의 신뢰도 배분(reliability allocation)이 이루어진다. 시스팀의 설계가 완료된 상태에서 상세 설계되어 제품이 실현되며, 시제품에 대한 신뢰도 예측(reliability prediction)업무가 착수된다. 시스팀의 품질목표치를 겨냥한 실질적인 품질 및 신뢰성 활동들이 신뢰도 배분치 및 시스팀 신뢰도 목표치로 bottom-up 방식으로 접근하게 됨에 따라, 본고에서는 시스팀의 품질 목표치를 달성하기 위해 가장 원천적으로 기본이 될 수 있는 전자부품의 품질 수준을 분석하고, 신뢰성 관련 제반 시험 기술을 분석 기술하고자 한다.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.354-356
• /
• 2011

Renewable energy resources such as wind, wave, solar, micro hydro, tidal and biomass etc. are becoming importance stage by stage because of considering effect of the environment. Solar energy is one of the most successful sources of renewable energy for the production of electrical energy following solar energy. And, the solar/photovoltaic cell generators depend on the solar radiation, which is a random variable so this poses difficulty in the system scheduling and energy dispatching, as the schedule of the photovoltaic cell generators availability is not known in advance. This paper proposes to use the two-layered artificial neural networks for predicting the actual solar radiation from the previous values of the same variable.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.2
• /
• pp.193-200
• /
• 2002

A study was conducted to investigate the availability of Fitts' Law to Hangul input systems on mobile phones. Three different Hangul input systems were experimented to measure the performance time to evaluate the physical interface of all. The measured performance time was found to be well fitted with the modified Fitts' Law by Hangul input systems on mobile phones. As a result, the physical interfaces for Hangul input systems could be evaluated quantitatively with the prediction of the performance time by Fitts' Law.