The Open Academic Analytics Initiative (OAAI) is a collaborative, multi-year grant program aimed at researching issues related to the scaling up of learning analytics technologies and solutions across all higher education. paper describes goals objectives OAAI, depicts process challenges collecting, organizing mining student data predict academic risk, report results on predictive performance those models, their portability pilot programs partner institutions, interventions at-risk students.

This project explored perceptions of ChatGPT in higher education among students and faculty to assess teaching learning implications this Generative Artificial Intelligence (Generative AI)–based novel tool. Two theoretical frameworks inspired the project, including Diffusion Innovation theory (Rogers, 1962) Technology Acceptance Model (Davis, 1989). An online survey was completed by 380 participants ( N = 380). Participants indicated that they would not use plagiarize but believed others...

In this paper we report ongoing research on the Open Academic Analytics Initiative (OAAI), a project aimed at increasing college student retention by performing early detection of academic risk using data mining methods. The describes goals and objectives OAAI, lays out methodological framework to develop models that can be used perform inferential queries performance open source course management system records. Preliminary results initial model development several algorithms for...

In this global information age, schools that teach public affairs and administration must meet the needs of students. Increasingly, means providing students in online classrooms to help them reach their highest potential. The acts teaching learning generate data, but date, has remained largely untapped for assessing student performance.Using data generated by an Master Public Administration program, drawn from Marist College Open Academic Analytics Initiative,1 we identify analyze...

<title>Abstract</title> This study employs Bayesian hierarchical generalized linear models to investigate predictors of freshmen student attrition using Marist University data from nine academic years (n=10921) and across its six schools. The proposed framework builds binary logistic regression estimate the posterior probability distributions models' parameters derived metrics. paper shows how formulate (Bernoulli) implement them in a probabilistic programming platform compute Markov chain...

This paper describes the results on research work performed by Open Academic Analytics Initiative, an on-going project aimed at developing early detection system of college students academic risk, using data mining models trained student personal and demographic data, as well course management data. We report initial findings predictive performance those models, their portability across pilot programs in different institutions interventions applied pilots.

Practitioners and researchers regularly refer to error rates or accuracy percentages of databases. The former is the number cells in divided by total cells; latter correct cells. However, databases may have similar (or percentages) but differ drastically complexity their problems. A simple percent does not provide information as whether errors are systematic randomly distributed throughout database. We expand metric include a randomness measure probability distribution value. proposed check...

This article analyzes the data quality issues that emerge when training a shrinkage-based classifier with noisy data. A statistical text analysis technique based on variation of multinomial naive Bayes is applied to set free-text discharge diagnoses occurring in number hospitalizations. All these were previously coded according Spanish Edition ICD9-CM. We deal issue analyzing predictive power and robustness machine learning algorithm proposed for ICD-9-CM classification. explore effect...

Over the last years, use of peripheral blood-derived big datasets in combination with machine learning technology has accelerated understanding, prediction, and management pulmonary critical care conditions. The goal this article is to provide readers an introduction methods applications blood omics other multiplex-based technologies medicine setting better appreciate current literature field. To accomplish that, we essential concepts needed rationalize approach introduce types molecules...

In this paper, we present our research in applying statistical machine learning methods for network intrusion detection. With the advent of online distributed services, issue preventing and other forms information security failures is gaining prominence. work, use two different algorithms classification (decision trees naive Bayes classifier) to build predictive models capable distinguishing between 'bad' TCP/IP connections, called intrusions attacks, 'good' normal connections. We...

In this paper, the authors present a quantitative model for estimating security risk exposure firm. The includes formulation optimization of controls as well determining sensitivity assets to different threats. uses series matrices organize data groups assets, vulnerabilities, threats, and controls. are then linked such that is aggregated in each matrix cascaded across other matrices. computations reversible transparent allowing analysts answer what-if questions on data. based Annualized...

This paper synthesizes some of the technical decisions, design strategies & concepts developed during execution Open Academic Analytics Initiative (OAAI), a research program aimed at improving student retention rates in colleges, by deploying an open-source academic early alert system to identify students risk. The explains prototype demonstration system, detailing several dimensions data mining analysis such as: integration, predictive modelling and scoring with reporting. should be...

This poster synthesizes the design features of a visualization layer applied on Open Academic Analytics Initiative (OAAI), an open source academic early alert system based predictive analytics. The explores ways to convey model outputs and benchmark student performances using visually intuitive radar plots.

The present paper addresses the issue of learning underlying structure a discrete binary Bayesian network, expressed as directed acyclic graph, which includes specification conditional independence assumptions among attributes model; and given model, probability distributions that quantify those dependencies. approach followed in this work heuristically searches space network structures using scoring function based on Minimum Description Length Principle, takes into account volume model...

Technology management education emerged in order to enable technology- driven firms link strategic goals their technological capabilities and requirements. the broadest sense has become a driver for individual entire industries, as well being key component increasingly complex global economy. It is therefore given that technology programs need provide thorough understanding of how impacts economy, vice versa, manage globalized technology. This paper provides framework development courses...