Search

Background: Teachers often rely on the use of open‐ended questions to assess students' conceptual understanding of assigned content. Particularly in the context of mathematics; teachers use these types of questions to gain insight into the processes and strategies adopted by students in solving mathematical problems beyond what is possible through more close‐ended problem types. While these types of problems are valuable to teachers, the variation in student responses to these questions makes it difficult, and time‐consuming, to evaluate and provide directed feedback. It is a well‐studied concept that feedback, both in terms of a numeric score but more importantly in the form of teacher‐authored comments, can help guide students as to how to improve, leading to increased learning. It is for this reason that teachers need better support not only for assessing students' work but also in providing meaningful and directed feedback to students. Objectives: In this paper, we seek to develop, evaluate, and examine machine learning models that support automated open response assessment and feedback. Methods: We build upon the prior research in the automatic assessment of student responses to open‐ended problems and introduce a novel approach that leverages student log data combined with machine learning and natural language processing methods. Utilizing sentence‐level semantic representations of student responses to open‐ended questions, we propose a collaborative filtering‐based approach to both predict student scores as well as recommend appropriate feedback messages for teachers to send to their students. Results and Conclusion: We find that our method outperforms previously published benchmarks across three different metrics for the task of predicting student performance. Through an error analysis, we identify several areas where future works may be able to improve upon our approach. Lay Description: What is already known about this topic: Open‐ended questions are used by teachers in the domain of mathematics to assess their students understanding but automated support for these types of questions are limited in online learning platforms. Recent advancements in areas of machine learning and natural language processing have led to promising results in a range of domains and applications. What this paper adds: Emulating how teachers identify similar student answers can be used to build tools that support teachers in assessing and providing feedback to student open‐ended work. Implications for practice: Developing better automated supports can increase the amount of direct feedback students receive to guide their learning. [ABSTRACT FROM AUTHOR]

Economic growth, air pollution, and forest fires in some states in the United States have increased the concentration of particulate matter with a diameter less than or equal to 2.5 μm (PM2.5). Although previous studies have tried to observe PM2.5 both spatially and temporally using aerosol remote sensing and geostatistical estimation, they were limited in accuracy by coarse resolution. In this paper, the performance of machine learning models on predicting PM2.5 is assessed with linear regression (LR), decision tree (DT), gradient boosting regression (GBR), AdaBoost regression (ABR), XGBoost (XGB), k‐nearest neighbors (K‐NN), long short‐term memory (LSTM), random forest (RF), and support vector machine (SVM) using PM2.5 station data from 2017 to 2021. To compare the accuracy of all the nine machine learning models, the coefficient of determination (R2), root mean square error (RMSE), Nash‐Sutcliffe efficiency (NSE), root mean square error ratio (RSR), and percent bias (PBIAS) were evaluated. Among all nine models, the RF (100 decision trees with a max depth of 20) and support vector regression (SVR; nonlinear kernel, degree 3 polynomial) models were the best for predicting PM2.5 concentrations. Additionally, comparison of the PM2.5 performance metrics displayed that the models had better predictive behavior in the western United States than that in the eastern United States. Key Points: We present the prediction of PM2.5 concentrations over the United States using various machine learning (ML) algorithmsWe show ML as a new approach for analyzing large data sets due to the computational speed and easy implementation for massive amounts of dataThe study is important for improving our understanding of the differences among ML algorithms for Earth Science research [ABSTRACT FROM AUTHOR]

Manganese (Mn) concentrations and the probability of arsenic (As) exceeding the drinking‐water standard of 10 μg/L were predicted in the Mississippi River Valley alluvial aquifer (MRVA) using boosted regression trees (BRT). BRT, a type of ensemble‐tree machine‐learning model, were created using predictor variables that affect Mn and As distribution in groundwater. These variables included iron (Fe) concentrations and specific conductance predicted from previously developed BRT models, groundwater flux and age estimates from MODFLOW, and hydrologic characteristics. The models also included results from the first airborne geophysical survey conducted in the United States to target an entire aquifer system. Predictions of high Mn and As occurred where Fe was high. Predicted high Mn concentrations were correlated with fraction of young groundwater (less than 65 years) computed from MODFLOW results. High probabilities of As exceedance were predicted where groundwater was relatively old and airborne electromagnetic resistivity was high, typically proximal to streams. Two‐variable partial‐dependence plots and sensitivity analysis were used to provide insight into the factors controlling Mn and As distribution in groundwater. The maps of predicted Mn concentrations and As exceedance probabilities can be used to identify areas where these constituents may be high, and that could be targeted for further study. This paper shows that incorporation of a selected set of process‐informed data, such as MODFLOW results and airborne geophysics, into a machine‐learning model improves model interpretability. Incorporation of process‐rich information into machine‐learning models will likely be useful for addressing a wide range of problems of interest to groundwater hydrologists. Article impact statement: Machine‐learning models trained on output from geophysical surveys, MODFLOW, and water‐quality predictions can be used to map trace metals. [ABSTRACT FROM AUTHOR]

Classifying images using supervised machine learning (ML) relies on labeled training data—classes or text descriptions, for example, associated with each image. Data‐driven models are only as good as the data used for training, and this points to the importance of high‐quality labeled data for developing a ML model that has predictive skill. Labeling data is typically a time‐consuming, manual process. Here, we investigate the process of labeling data, with a specific focus on coastal aerial imagery captured in the wake of hurricanes that affected the Atlantic and Gulf Coasts of the United States. The imagery data set is a rich observational record of storm impacts and coastal change, but the imagery requires labeling to render that information accessible. We created an online interface that served labelers a stream of images and a fixed set of questions. A total of 1,600 images were labeled by at least two or as many as seven coastal scientists. We used the resulting data set to investigate interrater agreement: the extent to which labelers labeled each image similarly. Interrater agreement scores, assessed with percent agreement and Krippendorff's alpha, are higher when the questions posed to labelers are relatively simple, when the labelers are provided with a user manual, and when images are smaller. Experiments in interrater agreement point toward the benefit of multiple labelers for understanding the uncertainty in labeling data for machine learning research. Plain Language Summary: After hurricanes and storms, pictures taken from a plane can be used to observe how the coast was impacted. A single flight might take thousands of pictures. If a computer could automatically analyze the pictures, then a person would not need to look at them one‐by‐one. To teach a computer to analyze images, we need many pictures and many labels that describe what is visible in each picture. But where do we get those labels? Typically, a coastal scientist labels the pictures by sorting them into folders or typing codes into a spreadsheet. But does every coastal scientist label pictures the same way? Some labeling questions are easy to answer, and scientists mostly agree ("Is this image all water?"). Other labeling questions are harder to answer, and cause disagreement ("Was there damage to buildings?"). This paper is about how well scientists agree when labeling the same pictures, and how we can improve agreement among scientists. We try some experiments and offer a few ideas on how to improve agreement. We suggest writing very clear questions, using smaller images, and having a comprehensive manual. It turns out that having a manual with examples—and reading the manual!—really helps. Key Points: We measure agreement among coastal scientists labeling the same sets of poststorm imagesCoastal scientists agree more when rating landforms, less when labeling inferred processesIterating on questions, providing documentation, and using smaller image sizes all increase agreement [ABSTRACT FROM AUTHOR]

Electricity usage of buildings (including offices, malls, and residential apartments) represents a significant portion of a nation's energy expenditure and carbon footprint. In the United States, the buildings' appliances consume 72% of the total produced electricity approximately. In this regard, cyber‐physical system (CPS) researchers have put forth associated research questions to reduce cyber‐physical building environment energy consumption by minimizing the energy dissipation while securing occupants' comfort. Some of the questions in CPS building include finding the optimal HVAC control, monitoring appliances' energy usage, detecting insulation problems, estimating the occupants' number and activities, managing thermal comfort, intelligently interacting with the smart grid. Various machine learning (ML) applications have been studied in recent CPS researches to improve building energy efficiency by addressing these questions. In this paper, we comprehensively review and report on the contemporary applications of ML algorithms such as deep learning, transfer learning, active learning, reinforcement learning, and other emerging techniques that propose and envision to address the above challenges in the CPS building environment. Finally, we conclude this article by discussing diverse existing open questions and prospective future directions in the CPS building environment research. This article is categorized under:Technologies > Machine LearningTechnologies > Reinforcement Learning [ABSTRACT FROM AUTHOR]

Background: National ranking from the triennial Programme of International Student Assessment (PISA) often serves as a barometer of national performance and human capital. Though excessive student‐ and school‐level covariates (n > 700) may prove intractable for traditional least‐squares estimate procedures, shrinkage methods may be more suitable for subset selection. Aims: With a focus on the United States, this paper proposes sparse regression for PISA 2012 to discover salient student‐ and school‐level predictor variables for mathematical literacy achievement. Sample: The sparse regression analysis was conducted on 10 top‐tiered OECD countries/economies, Canada, and the United States in mathematical literacy on the 2012 PISA. Two‐ and three‐level hierarchical regression analyses were performed on Canadian and US students (N = 26,522) along with five of the ten top‐tiered countries/economies (N = 58,385). Methods: Using the 'least absolute shrinkage and selection operator' (LASSO) technique, the study (1) identified salient predictor variables of mathematical literacy performance for the top‐tiered countries/economies, Canada, and the United States and (2) used these salient variables to perform two‐ and three‐level hierarchical regression on data from Canada and the United States along with five top‐tiered countries/economies. Weights and replicates were used to account for complex sample design. A weighted, two‐level confirmatory factor analysis was performed to identify latent constructs. Missing data were handled through multiple imputation. Results: Separate two‐level hierarchical models accounted for 32–35% student‐level and 58–70% school‐level variance in Canada and the United States, respectively; three‐level models accounted for 33% of level‐one variance, 62–65% level‐two variance, and 13–44% of level‐three variance for the US/Canada and US/Canada/top‐tiered students, respectively. Following top‐tiered countries/economies, Canadian students had high levels of self‐efficacy, were more likely to encounter advanced concepts in class, were less activity/small group‐centred, and were more likely to consider truancy a learning hindrance. Factor analyses revealed a positive relation with rigour and class organization (teacher‐centred) for top‐tiered countries and Canada, though not for the United States. For all countries, there was a strong relation between rigour and self‐beliefs. Conclusion: Compared to top performers, a less rigorous curriculum, coupled with class and school factors, may explain lag in US performance. [ABSTRACT FROM AUTHOR]