Published: July 10, 2024
Language: Английский
Advances in educational technologies and instructional design book series, Journal Year: 2024, Volume and Issue: unknown, P. 44 - 59
Published: March 4, 2024
Predictive analytics is a crucial tool in changing teaching and learning practices the ever-changing field of educational technology. This study examines dynamic function predictive customizing education, with specific emphasis on its ability to adapt paths improve individual student achievement. The how models might identify distinct patterns demands by assessing many data sources, such as academic achievement, habits, engagement indicators. It showcases capabilities these generating adaptive experiences, thereby providing more focused approach teaching. article investigates facilitates early detection hazards, allowing for timely interventions support students who are at danger underperformance or dropping out.
Language: Английский
Citations
6
Frontiers in Education, Journal Year: 2025, Volume and Issue: 10
Published: Feb. 13, 2025
Introduction The study examines the impact of targeted educational interventions on academic success and retention engineering students identified as high-risk, with a focus two student groups historically underrepresented in STEM: minority (URM) female students. These included an alternative curriculum pathway, co-calculus support course, spatial visualization training. Building our previous work, we evaluated outcomes designed to improve graduation rates among most academically underprepared from these groups, who were consequently categorized high-risk. Methodology We analyzed data 10 cohorts, covering 5 years before after implemented. utilized two-population proportion test compare groups' rates, early STEM courses during pre- post-intervention periods. Additionally, constructed logistic regression models identify key factors influencing on-time graduation. Results Our results show that significantly increased both 4- 6-year for high-risk URM by nearly 20 percentage points. Although improved changes not found be statistically significant. However, their performance foundation courses, particularly Physics I Calculus I, post-intervention. Discussion Logistic indicated shift significance rate predictors post-intervention, demonstrating efficacy tailored strategies. reduced importance grades predicting intervention suggests benefit which decoupled this course mitigated previously significant predictor being non-URM graduation, indicating leveling effect findings highlight potential customized enhance disciplines.
Language: Английский
Citations
0
Discover Sustainability, Journal Year: 2025, Volume and Issue: 6(1)
Published: May 14, 2025
Language: Английский
Citations
0
Published: May 3, 2024
In the evolving landscape of educational research, predictive analysis student performance using data science has garnered significant interest. This study investigates influence diverse factors on academic outcomes, ranging from personal demographics to socioeconomic conditions, enhance strategies and support mechanisms. We employed a ml models analyze information containing records information. The tested include Logistic Regression, Random Forest (RF), Gradient Boosting (GB), Support Vector Machines (SVC), K-Nearest Neighbors (KNN), Gaussian Naive Bayes, Decision Trees. process involved comprehensive preprocessing, exploratory analysis, model training, evaluation based metrics such as precision, recall, accuracy, $F 1$ score. results indicate that ensemble methods, specifically RF GB, demonstrate superior efficacy in accurately predicting categories 'Enrolled,' 'Graduated,' 'Dropped Out.' These excelled handling complex interplay varied predictors affecting success. further underline potential advanced ML techniques significantly outperforming prediction accuracy domain, hence facilitating tailoring interventions foster improved engagement better outcomes for students. provided comparative methods guide future application analytics education.
Language: Английский
Citations
3
Advances in human and social aspects of technology book series, Journal Year: 2024, Volume and Issue: unknown, P. 186 - 202
Published: June 14, 2024
Higher education institutions face a problem with student turnover that has many aspects and affects both students universities in different ways. Using predictive analytics machine learning, this study shows new way to deal problem. The main goal is create predicting algorithms can predict which are most likely drop out, so colleges get involved their lives timely effective way. As part of method, the authors collect preprocess large dataset from university records. This includes information about academic success, socioeconomic background, participation campus activities, psychological health. uses advanced learning methods look at all these data points. It focuses on feature selection engineering find important factors dropout. Rigid validation used test how well model works, making sure it accurately reliably future.
Language: Английский
Citations
1
Advances in educational technologies and instructional design book series, Journal Year: 2024, Volume and Issue: unknown, P. 293 - 316
Published: June 28, 2024
Using a large dataset that includes students' grades, demographic information, and other educational variables from three American high schools, this research work investigates the predictive modeling of mathematical performance. Gender, race/ethnicity, parental education, lunch subsidy status, standardized test results (math, reading, writing), course enrollment in preparation are all part dataset. The purpose study is to examine relationship between socioeconomic status their achievement discover important predictors using sophisticated machine learning algorithms such as ensemble methods, decision trees, linear regression. A more complex picture factors lead can be gained study, which uncovers illuminating relationships across variables, interventions, academic results. highlight promise analytics for developing individualized plans improve experiences. Educators, legislators, future researchers benefit data-driven methods planning decision-making, highlighted paper's examination findings' ramifications.
Language: Английский
Citations
1
Computers and Education Artificial Intelligence, Journal Year: 2024, Volume and Issue: unknown, P. 100352 - 100352
Published: Dec. 1, 2024
Language: Английский
Citations
1
Advances in business information systems and analytics book series, Journal Year: 2024, Volume and Issue: unknown, P. 203 - 224
Published: June 28, 2024
In this study, a wide range of geoFigureical locations are investigated to investigate the complex relationships that exist between agricultural productivity and important environmental parameters. These elements include fluctuations in temperature, patterns rainfall, application pesticides. Through utilization vast dataset encompasses yield measures, meteorological conditions, practices over period several years, we employ sophisticated statistical machine learning techniques order uncover subtle linkages regulate crop output. The findings our study indicate there substantial correlations outcomes yields particular show major impact sustainable farming climate adaptation methods have on efficiency production. highlight significance integrated resource management requirement for precision agriculture
Language: Английский
Citations
0
Published: July 10, 2024
Language: Английский
Citations
0