Comparison of PLSR and PCR techniques in terms of dimension reduction: an application on internal migration data in Turkey

Şamkar, Hatice; Güven, Gamze

doi:10.21833/ijaas.2016.08.002

Cited by 2 publications

(2 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The contribution of different climate factors to TWS anomalies are also conducted based on the partial least squares regression (PLSR) so as to validate the evaluation results based on the entropy method. Compared with traditional multivariate techniques, PLSR is a better choice for analyzing high‐dimensional data because of its robustness and adaptability, and detailed algorithm can be seen in Samkar and Guven (2016). Limited by the short time series of in situ records, the anthropogenic impact is not considered in the PLSR model.…”

Section: Methodsmentioning

confidence: 99%

Assessing Underground Water Exchange Between Regions Using GRACE Data

Yin

Zheng

et al. 2020

JGR Atmospheres

View full text Add to dashboard Cite

Quantifying the underground water exchange between adjacent regions is a common question in field investigations and groundwater flow models. This paper proposes an innovative approach to assess underground water exchange patterns based on Gravity Recovery and Climate Experiment (GRACE) observations, the water balance method, and Darcy's law. The dynamic patterns of flux changes are investigated over four boundaries of an arid plain. Moreover, the causes of anthropogenic and climate factors are evaluated using the entropy method and the partial least squares regression (PLSR) model. The results indicate that terrestrial water storage (TWS) in the study area shows a decreasing trend at a rate of −1.77 ± 0.21 mm/yr from 2003 to 2015. The water resources recharging the study area from its southern boundary show a noticeable increasing trend, which is mainly attributed to the warming climate and increasing precipitation on the Qilian Mountains. The Ecological Water Diversion Project and the imbalance between precipitation and evapotranspiration are the main factors inducing the decline of flux changes for the northern, western, and eastern boundaries. Under the constrains of empirical values of hydrogeological parameters, GRACE‐based methods find that increment of the flux changes in the southern boundary may account for approximately 0.43% of the investigated flux in 2000, and the decrement of flux changes in the eastern boundary may be about 0.01% to the previous measured results. Anthropogenic activities are the dominant driving force for the changes in water resources, with the weight of 88.70%. The presented method can provide insights into the patterns of flux change in data‐poor areas.

show abstract

Section: Methodsmentioning

confidence: 99%

Assessing Underground Water Exchange Between Regions Using GRACE Data

Yin

Zheng

et al. 2020

JGR Atmospheres

View full text Add to dashboard Cite

show abstract

“…, y q ] l×q are obtained. The aim of PLSR is to form the structure between X and Y, and to predict Y via X [30]. The first latent vector t 1 and u 1 are extracted from X and Y respectively, and the principles of the extraction are shown as follows.…”

Section: Plsr For the Prediction Of The Dvl Measurementsmentioning

confidence: 99%

A Novel Hybrid Approach to Deal with DVL Malfunctions for Underwater Integrated Navigation Systems

Zhu

Cheng

et al. 2017

Applied Sciences

View full text Add to dashboard Cite

Abstract:As a common device for underwater integrated navigation systems, Doppler velocity log (DVL) has the risk of malfunction. To improve the reliability of navigation systems, a hybrid approach is presented to forecast the measurements of the DVL while it malfunctions. The approach employs partial least squares regression (PLSR) coupled with support vector regression (SVR) to build a hybrid predictor. As the current and past calculating velocities of strapdown inertial navigation system (SINS) are taken as the predictor's inputs, PLSR is applied to cope with the situation where there exists intense relativity among independent variables. Since PLSR is a linear regression, SVR is used to predict the residual components of the PLSR prediction to improve the accuracy. When the DVL works well, the hybrid predictor is trained online as a backup, whereas during malfunctions, the predictor offers the estimation of the DVL measurements for information fusion. The performance of the proposed approach is verified with simulations based on SINS/DVL/MCP/pressure sensor (PS) integrated navigation system. The comparison results indicate that the PLSR-SVR hybrid predictor can correctly provide the estimated DVL measurements and effectively extend the tolerance time on DVL malfunctions, thereby improving the navigation accuracy and reliability.

show abstract

Comparison of PLSR and PCR techniques in terms of dimension reduction: an application on internal migration data in Turkey

Cited by 2 publications

References 30 publications

Assessing Underground Water Exchange Between Regions Using GRACE Data

Assessing Underground Water Exchange Between Regions Using GRACE Data

A Novel Hybrid Approach to Deal with DVL Malfunctions for Underwater Integrated Navigation Systems

Contact Info

Product

Resources

About