LR B-splines to approximate bathymetry datasets: An improved statistical criterion to judge the goodness of fit

Skytt, Vibeke; Kermarrec, Gaël; Dokken, Tor

doi:10.1016/j.jag.2022.102894

Cited by 4 publications

(8 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A grade matrix G is constructed by T semester student-course data in the input layer, where G (i) is the grades achieved by students in each course, the G j is the grades of each student in each course, (i, j) ∈ Ω. Because the constructed grade matrix is very sparse, the data dimension can be reduced, and the data sparsity can be alleviated through the embedding-layer mapping [28]. For a given score matrix G ij , forms of grade matrix are obtained as follows:…”

Section: Deep Matrix Factorizationmentioning

confidence: 99%

Study on Score Prediction Model with High Efficiency Based on Deep Learning

Yang

Bai

2022

Electronics

View full text Add to dashboard Cite

In the problem of unified classroom performance prediction, there is a certain lag in the prediction, and there are also problems such as the data sparsity and single feature in the data. In addition, feature engineering is often carried out manually in modeling, which highly depends on the professional knowledge and experience of engineers and affects the accuracy of the prediction to a certain extent. To solve the abovementioned gaps, we proposed an online course score prediction model with a high time efficiency that combines multiple features. The model uses a deep neural network, which can automatically carry out feature engineering and reduce the intervention of artificial feature engineering, thus significantly improving the time efficiency. Secondly, the model uses a factorization machine and two kinds of neural networks to consider the influence of first-order features, second-order features, and higher-order features at the same time, and it fully learns the relationship between the features and scores, which improves the prediction effect of the model compared to using only single feature learning. The performance of the model is evaluated on the learning analysis dataset from Fall 2015 to Spring 2021 and includes 412 courses with 600 students. The experimental results show that the performance of the prediction model based on the feature combination proposed in the present study is better than the previous performance prediction model. More importantly, our model has the best time efficiency of below 0.3 compared to the other models.

show abstract

Section: Deep Matrix Factorizationmentioning

confidence: 99%

Study on Score Prediction Model with High Efficiency Based on Deep Learning

Yang

Bai

2022

Electronics

View full text Add to dashboard Cite

show abstract

“…A subset of the point cloud containing approximately 1 million points has been chosen to perform the mathematical surface approximation with LR B-splines, see Figure 9 (left). We refer to [13] for more details on the procedure. The surface fitting algorithm chosen is called locally adaptive as it compares the difference in absolute value between the mathematical surface and the parametrized points inside a cell of a mesh using a given tolerance (the higher the tolerance, the coarser the approximation).…”

Section: Application: Surface Fittingmentioning

confidence: 99%

“…Here, new solutions based on mathematical surface approximation with local refinement address the challenges of deformation analysis within the context of landslide monitoring with smooth surfaces and allow the efficient analysis of the information contained in the point clouds (see [11,12] for TLS point clouds, ref. [13] for bathymetry data set, or [14] for turbine blade design, to cite but a few). The main advantages of mathematical surfaces is to reduce the point clouds to a few parameters while simultaneously allowing spatially continuous and parametric deformation analysis [15,16].…”

Section: Introductionmentioning

confidence: 99%

“…The corresponding algorithms will be tested on a data set from a mountainous region in Austria where specific measurements with TLS were performed for landslide monitoring. Without going into mathematical details, we will show an application of the classification results for performing adaptive surface approximation with locally refined (LR) B-splines, which are best suited for approximating smooth terrains [13]. Other fields of application of such a comparison are the deformation analysis of near continuous monitoring of snow or sand dune movements from TLS point clouds, as well as for bathymetry.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Classification of Terrestrial Laser Scanner Point Clouds: A Comparison of Methods for Landslide Monitoring from Mathematical Surface Approximation

Kermarrec

Yang²,

Czerwonka‐Schröder

2022

Remote Sensing

Self Cite

View full text Add to dashboard Cite

Terrestrial laser scanners (TLS) are contact-free measuring sensors that record dense point clouds of objects or scenes by acquiring coordinates and an intensity value for each point. The point clouds are scattered and noisy. Performing a mathematical surface approximation instead of working directly on the point cloud is an efficient way to reduce the data storage and structure the point clouds by transforming “data” to “information”. Applications include rigorous statistical testing for deformation analysis within the context of landslide monitoring. In order to reach an optimal approximation, classification and segmentation algorithms can identify and remove inhomogeneous structures, such as trees or bushes, to obtain a smooth and accurate mathematical surface of the ground. In this contribution, we compare methods to perform the classification of TLS point clouds with the aim of guiding the reader through the existing algorithms. Besides the traditional point cloud filtering methods, we will analyze machine learning classification algorithms based on the manual extraction of point cloud features, and a deep learning approach with automatic extraction of features called PointNet++. We have intentionally chosen strategies easy to implement and understand so that our results are reproducible for similar point clouds. We show that each method has advantages and drawbacks, depending on user criteria, such as the computational time, the classification accuracy needed, whether manual extraction is performed or not, and if prior information is required. We highlight that filtering methods are advantageous for the application at hand and perform a mathematical surface approximation as an illustration. Accordingly, we have chosen locally refined B-splines, which were shown to provide an optimal and computationally manageable approximation of TLS point clouds.

show abstract

“…Inner model testing and outside model testing are two types of model testing. The two test models each had their own goodness-of-fit indicators (Skytt, Kermarrec, & Dokken, 2022). Average Variance Extracted (AVE), Square Roots AVE, Cross Loadings, Cronbach Alpha (CA), and Composite Reliability (CR) are used as indicators in the outer model testing.…”

Section: Hypothesis Testingmentioning

confidence: 99%

The Effect Of Kopi Kenangan Online Menus On Consumers’ Purchase Intention During The Covid-19 Pandemic

Sofyan

Hartono

2022

Eduvest

View full text Add to dashboard Cite

This study examined consumers purchase intentions during the pandemic in relation to the online food menu under the extended S-O-R framework, particularly with Kopi Kenangan as the object. Kopi Kenangan is one of fastest-growing grab-andgo coffee chains in Indonesia. The purpose of this study is to discover the factors driving people intention to purchase Kopi Kenangan products through online food ordering platforms. In the data collection process, 250 samples from Kopi Kenangan consumers who had previously purchased them online were used, which involved distributing online questionnaires using Google forms. The data was then further analyzed using the PLS-SEM approach. The result shows that menu visual appeal and menu informativeness positively affected desire for food. The perception of COVID-19 risks positively affected the perceived convenience of online food ordering as well. While, desire for food and perceived convenience also positively affected purchase intention

show abstract

LR B-splines to approximate bathymetry datasets: An improved statistical criterion to judge the goodness of fit

Cited by 4 publications

References 17 publications

Study on Score Prediction Model with High Efficiency Based on Deep Learning

Study on Score Prediction Model with High Efficiency Based on Deep Learning

Classification of Terrestrial Laser Scanner Point Clouds: A Comparison of Methods for Landslide Monitoring from Mathematical Surface Approximation

The Effect Of Kopi Kenangan Online Menus On Consumers’ Purchase Intention During The Covid-19 Pandemic

Contact Info

Product

Resources

About