Multi-Class Assessment Based on Random Forests

Berriri, Mehdi; Djema, Sofiane; Rey, Gaëtan; Dartigues-Pallez, Christel

doi:10.3390/educsci11030092

Cited by 6 publications

(3 citation statements)

References 14 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Diagram depicting random forest modeling (modified from Berriri et al, 2021). (a) The full set of observations are bagged (i.e., subsetted) to create (b) random subsets of the data (also called “in‐bag” samples) to construct each regression tree of the random forest.…”

Section: Methodsmentioning

confidence: 99%

Random forest models to estimate bankfull and low flow channel widths and depths across the conterminous United States

Doyle

Hill

Leibowitz

et al. 2023

J American Water Resour Assoc

View full text Add to dashboard Cite

Channel dimensions (width and depth) at varying flows influence a host of instream ecological processes, as well as habitat and biotic features; they are a major consideration in stream habitat restoration and instream flow assessments. Models of widths and depths are often used to assess climate change vulnerability, develop endangered species recovery plans, and model water quality. However, development and application of such models require specific skillsets and resources. To facilitate acquisition of such estimates, we created a dataset of modeled channel dimensions for perennial stream segments across the conterminous United States. We used random forest models to predict wetted width, thalweg depth, bankfull width, and bankfull depth from several thousand field measurements of the National Rivers and Streams Assessment. Observed channel widths varied from <5 to >2000 m and depths varied from <2 to >125 m. Metrics of watershed area, runoff, slope, land use, and more were used as model predictors. The models had high pseudo R2 values (0.70–0.91) and median absolute errors within ±6% to ±21% of the interquartile range of measured values across 10 stream orders. Predicted channel dimensions can be joined to 1.1 million stream segments of the 1:100 K resolution National Hydrography Dataset Plus (version 2.1). These predictions, combined with a rapidly growing body of nationally available data, will further enhance our ability to study and protect aquatic resources.

show abstract

Section: Methodsmentioning

confidence: 99%

Random forest models to estimate bankfull and low flow channel widths and depths across the conterminous United States

Doyle

Hill

Leibowitz

et al. 2023

J American Water Resour Assoc

View full text Add to dashboard Cite

show abstract

“…Accuracy, precision, recall, and F1-score are four classic measurements often used to describe the classification model [34][35][36]. They are defined as Formulas ( 11)-( 14), in which, TP indicates the number of true positives, TN indicates the number of true negatives, FP indicates the number of false positives, and FN indicates the number of false negative [37].…”

Section: Evaluatingmentioning

confidence: 99%

An Improved Wood Recognition Method Based on the One-Class Algorithm

Sun

et al. 2022

Forests

View full text Add to dashboard Cite

Wood recognition is necessary for work in the wood trade activities. The advantage of the one-class wood classification method is more generalization, and it only needs positive samples and does not need negative samples in the training phase, so it is suitable for rare wood species inspection. This paper proposed an improved method based on the one-class support vector machine (OCSVM) for wood species recognition. It uses cross-section images acquired with a magnifying glass, which uses a pre-trained VGG16 model for feature extraction, a normal distribution test for key features filtering, and OCSVM to determine the wood species. The results showed that the approach achieved a mean recall of 0.842 for both positive and negative samples, which indicates this method has good performance for wood recognition. In a negative public dataset, the negative recall reached as high as 0.989, which showed that this method has good generalization.

show abstract

“…The number of features used to train ML models in the reviewed studies varied significantly e.g. one study (Beaulac & Rosenthal, 2019) only used 7 features to train the model, another study (Tenpipat & Akkarajitsakul, 2020) used 81 features, whereas, another study (Berriri et al, 2021) used 150 features. The features of the datasets in the reviewed studies are based on demographic and socio-economic background, pre-university, and university academic records, LMS interaction attributes.…”

Section: Figure 7: Top Abstract Appearance Across the Range Of Public...mentioning

confidence: 99%

Application of machine learning in higher education to assess student academic performance, at-risk, and attrition: A meta-analysis of literature

Fahd

Venkatraman²,

Miah

et al. 2021

Educ Inf Technol

View full text Add to dashboard Cite

Recently, the field of machine learning (ML) has evolved and finds its application in higher education (HE) for various data analysis. Studies have shown that such an emerging field in educational technology provides meaningful insights into several dimensions of educational quality. An in-depth analysis of the application of ML could have a positive impact on the HE sector. However, there is a scarcity of a systematic review of HE literature to gain from the overarching trends and patterns discovered using ML. This paper conducts a systematic review and meta-analyses of research studies that have reported on the application of ML in HE. The differentiating factors of this study are primarily vested in the meta-analyses including a specific focus on student academic performance, atrisk, and attrition in HE. Our detailed investigation adopts an evidence-based framework called PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) for reporting the findings of our systematic review and meta-analyses of literature on the use of ML models, algorithms, evaluation metrics, and other criteria including demographics for assessing student academic performance, at-risk and attrition in HE. After undergoing the PRISMA steps such as selection criteria and filtering, we arrive at a narrowed down dataset of 89 relevant studies published from 2010 to 2020 for an in-depth analysis. The results not only show the outcomes of the quantitative analysis of the application of ML types, models, evaluation metrics, and other related demographics but also provide quality insights of publication patterns and future trends towards predicting and monitoring student academic progress in HE.

show abstract

Multi-Class Assessment Based on Random Forests

Cited by 6 publications

References 14 publications

Random forest models to estimate bankfull and low flow channel widths and depths across the conterminous United States

Random forest models to estimate bankfull and low flow channel widths and depths across the conterminous United States

An Improved Wood Recognition Method Based on the One-Class Algorithm

Application of machine learning in higher education to assess student academic performance, at-risk, and attrition: A meta-analysis of literature

Contact Info

Product

Resources

About