Article
Chemistry, Multidisciplinary
Komal Naz, Isma Farah Siddiqui, Jahwan Koo, Mohammad Ali Khan, Nawab Muhammad Faseeh Qureshi
Summary: Employee churn analytics is a process that assesses and predicts employee turnover rate in a company, aiming to reduce churn issue and additional costs. This paper proposes an IoT-enabled predictive strategy to identify future churners through analyzing features and performing classification, achieving a high accuracy rate in the experiment.
APPLIED SCIENCES-BASEL
(2022)
Article
Public, Environmental & Occupational Health
Yushuai Yu, Zelin Xu, Tinglei Shao, Kaiyan Huang, Ruiliang Chen, Xiaoqin Yu, Jie Zhang, Hui Han, Chuangui Song
Summary: This study developed a machine learning model to predict the prognosis of patients with primary breast lymphoma. The incidence of primary breast lymphoma increased significantly before 2004 and then began to decrease. The trend of primary breast lymphoma varies by age and race. The prognosis of patients has been improved in recent years.
JMIR PUBLIC HEALTH AND SURVEILLANCE
(2023)
Article
Computer Science, Interdisciplinary Applications
Weike Sun, Richard D. Braatz
Summary: Data analytics tools are transforming decision-making and design processes in manufacturing, but selecting the best method requires expertise. The Smart Process Analytics framework allows users to focus on goals rather than methods, effectively transforming manufacturing data into intelligent information through domain knowledge, data characteristics, and method selection through cross-validation.
COMPUTERS & CHEMICAL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Laith Abu Lekham, Yong Wang, Ellen Hey, Mohammad T. Khasawneh
Summary: This study aimed to build a multi-criteria text mining model for COVID-19 testing reasons and symptoms. The model integrates with a temporal predictive classification model for COVID-19 test results in underserved rural areas. By using look-up wordlists and a multi-criteria mapping process, the text mining model classifies the notes related to testing reasons and reported symptoms into one or more categories.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Biochemistry & Molecular Biology
Zachary S. Bohannan, Frederick Coffman, Antonina Mitrofanova
Summary: This study proposes a machine learning model utilizing interpretable genomic inputs to predict relapse/death in high-risk pediatric B-ALL patients. Through feature engineering and selection, the model effectively predicts patient survival and demonstrates a lower error rate in testing.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2022)
Article
Education & Educational Research
Andreas F. Gkontzis, Sotiris Kotsiantis, Christos T. Panagiotakopoulos, Vassilios S. Verykios
Summary: This article presents a method using big data mining and machine learning techniques to address the issue of student attrition in distance learning. By analyzing and predicting different datasets from the Hellenic Open University, at-risk students can be identified in a timely manner, allowing for personalized intervention.
INTERACTIVE LEARNING ENVIRONMENTS
(2022)
Article
Education & Educational Research
Andreas Gkontzis, S. Kotsiantis, Christos Panagiotakopoulos, Vassilios Verykios
Summary: This study focuses on the issue of student attrition in distance learning and proposes the use of big data and machine learning methods to predict student failure. The research finds that the imbalance problem in minority classes is often overlooked in conventional models, but the proposed algorithms can address this issue. Additionally, early time predictions and the significance of written assignments and specific quizzes are also highlighted.
INTERACTIVE LEARNING ENVIRONMENTS
(2022)
Review
Education & Educational Research
Nabila Sghir, Amina Adadi, Mohammed Lahmer
Summary: In recent years, there has been a rise in the use of Machine and Deep learning models to predict academic outcomes based on student-related data, aiming to improve the learning process. This study reviews the latest research on Predictive Analytics in Higher Education, analyzing outcomes frequently predicted, learning features used, predictive modelling process, and key performance metrics. The study also identifies gaps in current literature and suggests future research directions. It serves as a comprehensive reference for researchers in the field and informs educational stakeholders and decision-makers about potential opportunities.
EDUCATION AND INFORMATION TECHNOLOGIES
(2023)
Article
Computer Science, Interdisciplinary Applications
Serhat Simsek, Abdullah Albizri, Marina Johnsosn, Tyler Custis, Stephan Weikert
Summary: Predictive analytics and artificial intelligence are crucial for improving organizational performance and managerial decision-making. This study focused on identifying MLB free agents likely to receive a contract, using a design science research paradigm and CAM theory to develop a framework. The research found that a player's statistical performance and factors like age, Wins above Replacement, and last team played for are significant in predicting contract signings.
JOURNAL OF ENTERPRISE INFORMATION MANAGEMENT
(2021)
Article
Obstetrics & Gynecology
Jennifer J. Yland, Taiyao Wang, Zahra Zad, Sydney K. Willis, Tanran R. Wang, Amelia K. Wesselink, Tammy Jiang, Elizabeth E. Hatch, Lauren A. Wise, Ioannis Ch Paschalidis
Summary: Using data from the North American preconception cohort study, the researchers developed models to predict the probability of conception, achieving a performance of around 70% in the area under the receiver operating characteristic curve (AUC).
HUMAN REPRODUCTION
(2022)
Article
Computer Science, Information Systems
Sean C. Yu, Mackenzie R. Hofford, Albert M. Lai, Marin H. Kollef, Philip R. O. Payne, Andrew P. Michelson
Summary: This study proposes a practical terminology system for respiratory support methods, develops heuristics for constructing respiratory support episodes, and evaluates the utility of respiratory support information for mortality prediction. The results show that the addition of respiratory support information significantly improves mortality prediction, and the proposed representation method performs well, with respiratory support features being among the most important.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2022)
Article
Computer Science, Information Systems
Madhuri Gupta, Bharat Gupta, Abdoh Jabbari, Ishan Budhiraja, Deepak Garg, Ketan Kotecha, Celestine Iwendi
Summary: Breast cancer, a leading cause of death for women worldwide, can be detected through genetic markers. However, the high cost of genetic tests in developing countries like India prevents many patients from accessing timely diagnosis. To address this issue, a computer-assisted genetic test method (CAGT) is proposed to predict gene expressions and classify breast cancer in a faster and more cost-effective manner.
HUMAN-CENTRIC COMPUTING AND INFORMATION SCIENCES
(2023)
Article
Health Care Sciences & Services
Ashley J. Housten, Diana S. Hoover, Maggie Britton, Therese B. Bevers, Richard L. Street, Lorna H. McNeill, Larkin L. Strong, Jolyn Hersch, Kirsten McCaffery, Robert J. Volk
Summary: This study examined the knowledge, attitudes, and intentions related to conflicting recommendations for breast cancer screening among racially/ethnically diverse women. The findings suggest that divergent screening recommendations may lead to mistrust and paradoxically reinforce high overall enthusiasm for screening.
JOURNAL OF GENERAL INTERNAL MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Matthias Bogaert, Michel Ballings, Dirk Van den Poel, Asil Oztekin
Summary: The research shows that social media data significantly increases the predictive power of traditional box office prediction models. Facebook data outperforms Twitter data and including user-generated content consistently improves predictive power. Combination variables based on volume and valence of Facebook comments are identified as the most important variables.
DECISION SUPPORT SYSTEMS
(2021)
Article
Chemistry, Multidisciplinary
Murugesan Raju, Krishna P. Shanmugam, Chi-Ren Shyu
Summary: Early detection of glaucoma is crucial for preventing irreversible blindness. A predictive analytic framework was developed using machine learning and logistic regression methods on electronic health records (EHR) from over 650 hospitals and clinics in the USA. The study found that the XGBoost, MLP, and RF methods performed well in predicting glaucoma one year before onset, with an AUC score of 0.81, compared to 0.73 for logistic regression. This suggests that machine learning methods can identify potential pre-glaucoma patients in advance, leading to early intervention and prevention.
APPLIED SCIENCES-BASEL
(2023)
