Conference Session

DSA Technical Session 2

Collection

2024 ASEE Annual Conference & Exposition

Authors

Ben D Radhakrishnan, National University; James Jay Jaurez, National University; Nelson Altamirano, National University

Tagged Topics

Data Science & Analytics Constituent Committee (DSA), Diversity

understand and anticipate the potentialimplications of actions on environmental elements surrounding practical engineering challengesthrough trend analysis and descriptive statistics. Solar energy output and a state-by-state analysisof the potential for renewable energy sources in the United States provided the essential data forthe initial case study, allowing for interpolating and extrapolating data from real-world examples(Aginako & Guraya, 2021). Using the Excel toolset, students could see the statistical outcomesand forecasts using regression, which allowed them to visualize the potential benefits andchallenges of various decisions. Students could also dynamically evaluate the impact of specificresource decisions and engineering

Conference Session

DSA Technical Session 1

Collection

2024 ASEE Annual Conference & Exposition

Authors

Betul Bilgin, The University of Illinois at Chicago; Naomi Groza, The University of Illinois at Chicago

Tagged Topics

Data Science & Analytics Constituent Committee (DSA), Diversity

understand and apply data to real-world situations,transforming it from abstract numbers to valuable information. • "Useless if you do not know how to utilize it. Scary and overwhelming sometimes." This response reflects the apprehension and difficulty students experience when they cannot contextualize data. • "Applying data into specific applications" further emphasizes the struggle to translate data analysis into practical applications, highlighting a gap between technical skills and real-world implementation.Working with Large DatasetsHandling large volumes of data is another significant challenge reported by students, particularlyin terms of analysis and deriving meaningful conclusions. • "Working with large

Conference Session

DSA Technical Session 2

Collection

2024 ASEE Annual Conference & Exposition

Authors

Xiang Zhao, Alabama A&M University; Mebougna L. Drabo, Alabama A&M University

Tagged Topics

Data Science & Analytics Constituent Committee (DSA), Diversity

programs incorporated lectures,hands-on labs, group projects and/or national lab intern experience. In the last three-year’simplementation, the student assessment and project completion result all showed theeffectiveness of the approach in enhancing students’ ability to understand the science foundation,identify real-world problems, analyze data and develop data-driven solutions in nuclear energyand security areas. The feedback from student surveys is also satisfactory and positive. Thisresearch is sponsored by Department of Energy/NNSA and intends to share the project team’sexperience and lessons learned with the STEM education community.KeywordsData Science, Workforce Development, STEM Education, Nuclear Energy and SecurityIntroductionData

Conference Session

DSA Technical Session 4

Collection

2024 ASEE Annual Conference & Exposition

Authors

Fengbo Ma, Northeastern University; Xuemin Jin, Northeastern University

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

evaluated and compared against the baselines.We found that the reinforcement learning models’ performances are on par with the othersupervised machine learning models, establishing their potential in speech emotion recognition.Moreover, the reinforcement learning models are effective for continuous real-time speechemotion recognition. We also noticed that accurate audio segmentation plays a crucial role inreal-time speech emotion recognition.We conclude that reinforcement learning’s ability to continually integrate feedback greatlyenhances speech emotion recognition tasks in practical settings. However, the current data,derived from controlled lab environments, differs significantly from more complex and noisyreal-world data. Future work should

Conference Session

DSA Technical Session 6

Collection

2024 ASEE Annual Conference & Exposition

Authors

Emily Nutwell, The Ohio State University; Thomas Bihari, The Ohio State University; Thomas Metzger, The Ohio State University

Tagged Topics

Data Science & Analytics Constituent Committee (DSA), Diversity

., nominal, ordinal…) and what operations can be done on different types of data. • Data Models - (e.g., relational, graph, column…) and their pros and cons for use in solving particular problems. • Computation and Complexity – Why are some problems so hard to solve (big O notation)? • Parallel Processing and Cloud Computing – How can we apply parallelism to deal with computational complexity? • Frameworks: Hadoop, Map/Reduce, Spark – How can we apply frameworks to simplify application of parallelism?The main goal of the Data-Driven Problem-Solving Workflows thread in this course is thatlearners understand the value and the characteristics of defined workflows, and that theseworkflows help us solve real-world

Conference Session

DSA Technical Session 5

Collection

2024 ASEE Annual Conference & Exposition

Authors

Mehmet Ergezer, Wentworth Institute of Technology; Mark Mixer, Wentworth Institute of Technology; Weijie Pang, Wentworth Institute of Technology

Tagged Topics

Data Science & Analytics Constituent Committee (DSA), Diversity

the DS program in the fourth year encompasses a compulsory Machine Learning class,complemented by a diverse array of electives within the Data Science program. Further enriching theirskill set, students embark on their second Co-op work experience during the Spring semester, gaininghands-on exposure to real-world Data Science applications. The program concludes in the Summer with acombination of additional electives and a mandatory Senior Design class. In this concluding phase,students collaborate on a group project throughout the semester, applying their accumulated knowledgeand skills to address complex challenges in the field of Data Science.D. Educational ObjectivesIn this program design, we envision that integrating Computer Science

Conference Session

DSA Technical Session 8

Collection

2024 ASEE Annual Conference & Exposition

Authors

Neha Kardam, University of Washington; Denise Wilson, University of Washington

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

indirectimpressions, necessitate an approach capable of handling unstructured data effectively.Unsupervised learning methods, especially NMF, emerged as promising for heterogeneousdatasets, uncovering latent themes without being confined by predefined categories [16]. Theresults suggest that integrating unsupervised learning methods with domain expert interactioneffectively contributes to the creation of a robust dataset capable of accurately predicting themesin student responses using supervised learning. However, the observed decrease in performancemetrics on the testing data highlights the ongoing need for optimization and validation to ensurethe models' robustness, generalizability, and applicability in real-world educational

Conference Session

DSA Technical Session 1

Collection

2024 ASEE Annual Conference & Exposition

Authors

Ahmad Slim, The University of Arizona; Gregory L. Heileman, The University of Arizona; Husain Al Yusuf, The University of Arizona; Yiming Zhang, The University of Arizona; Asma Wasfi; Mohammad Hayajneh; Bisni Fahad Mon, United Arab Emirates University; Ameer Slim, University of New Mexico

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

students who have completed all required courses, signifiedby the absorbing state in the MDP network. Subsequent sections will demonstrate the appli-cation of this MDP model in real-world scenarios, showing how minor curricular changes cansignificantly influence graduation rates. This approach aligns with current research emphasizingstrategic course sequencing and bottleneck reduction as key factors in enhancing student successthrough curriculum structure.4 A Case StudyThe versatility of the Markov Decision Processes (MDP) model allows it to be applied across adiverse range of scenarios, catering to the specific requirements of different users such as faculty, (a

Conference Session

DSA Technical Session 5

Collection

2024 ASEE Annual Conference & Exposition

Authors

Duo Li, Shenyang Institute of Technology; Elizabeth Milonas, New York City College of Technology; Qiping Zhang, Long Island University

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

Data Management competency isrequired in 23-50% of jobs. The Machine Learning competency is required in about 30% of alljobs. Second, low demand competency group: Interestingly, competencies (ranging from 1%-13%) including Data Mining, Big Data, and Data Visualization are not widely required. TheData Science in Context competency is only required for 8-13% of jobs. Few jobs require theData Mining, Big Data (3-7%) and Data Visualization (1-3%) competencies. Third, long-term foundational competency group: Though the Math & Statisticscompetency is not as practical as programming and management, it is a foundational skillrequired for people to solve real-world problems. The Math and Statistics competency is requiredfor 27-30% of

Conference Session

DSA Technical Session 5

Collection

2024 ASEE Annual Conference & Exposition

Authors

Nicolas Leger, Florida International University; Maimuna Begum Kali, Florida International University; Stephanie Jill Lunn, Florida International University

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

give some clear examples of how data scienceprinciples can be used and potentially implemented in the traditional “numerical methods orapplied computing course [7, p. 1413].” In this work, we define non-computing engineers asthose individuals who are not pursuing computer science or computer engineering-specificformal education or degrees.However, despite the rapid growth and increasing adoption of data science in industry, thereremain challenges in incorporating it into engineering learning settings that traditionally have notheavily utilized data science computing techniques. These issues arise from the interdisciplinarycharacter of data science, as educators struggle to integrate the application of professionalknowledge to real-world

Conference Session

DSA Technical Session 1

Collection

2024 ASEE Annual Conference & Exposition

Authors

Ahmad Slim, The University of Arizona; Gregory L. Heileman, The University of Arizona; Melika Akbarsharifi, The University of Arizona; Kristina A Manasil, The University of Arizona; Ameer Slim, University of New Mexico

Tagged Topics

Data Science & Analytics Constituent Committee (DSA), Diversity

visually represent the direction of causality between different vari-ables, enabling us to investigate how changes in one variable, the causal factor, impact anothervariable. This implementation of causality not only facilitates predictive tasks, like other conven-tional machine learning models (i.e., hypothetical causation), but also enables us to conduct objec-tive “what-if” analyses (i.e., counterfactual causation) within the research context. In this study, weleverage real-world student data from 30 universities across the United States. The richness anddiversity of our dataset empower us to draw robust insights into the causal relationships amongvarious factors that influence student performance, particularly the complexity of the

Conference Session

DSA Technical Session 8

Collection

2024 ASEE Annual Conference & Exposition

Authors

Paula Francisca Larrondo, Queen's University; Brian M Frank P.Eng., Queen's University; Julian Ortiz, Queen's University

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

Design and PracticeSequence. This sequence prepares students for real-world, open-ended design problems [42].Part of the student summative evaluations includes team reports at different stages of the designprocess, such as Scoping and Preliminary Design Concept, Full Design Proposal, or FinalDesign, following Dym et al.[17] framework. Our analysis focuses on the Problem Definitionsections of these reports, which include identifying the problem goal, stakeholders, safetyconsiderations, functions, attributes, and constraints, among other possibilities that can be addedas the students progress in their project design.Feedback Mechanism and Model Fine-tuning:A multi-class classification approach using a fine-tuned LLM is the base for the feedback

Conference Session

DSA Technical Session 8

Collection

2024 ASEE Annual Conference & Exposition

Authors

Amirreza Mehrabi, Purdue Engineering Education; Jason Morphew, Purdue University, West Lafayette

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

systematically explorethe influence of factors like cognitive fatigue on test performance and ability estimates. Thesesimulation approaches, deemed promising models in psychometrics studies, underscore thesimulation's utility as an effective method for modeling and studying various aspects of the model.Indeed, the accuracy and Root Mean Square Error of Approximation (RMSEA) of the IRT modelserve as metrics, offering insights into the model's efficacy in the real world [42], [43], [48]. Theprobability of students answering correctly was simulated with and without CF. In the "no CF"condition, probability was simulated using the 4PL IRT model (equation 1). If the probability wascomputed below the guessing parameter (c), then the probability was set equal

Conference Session

DSA Technical Session 5

Collection

2024 ASEE Annual Conference & Exposition

Authors

Karl D. Schubert FIET, University of Arkansas; Shantel Romer, University of Arkansas; Stephen R. Addison, IEEE Educational Activities; Tina D Moore; Laura J Berry, North Arkansas College; Jennifer Marie Fowler, Arkansas State University; Lee Shoultz, University of Arkansas; Christine C Davis

Tagged Topics

Data Science & Analytics Constituent Committee (DSA)

Gen ED, Technical CompositionENGL 1033 HIST or PLSC or PSYC or Elective II SOC Intro to Object-Oriented Analytical Geometry andDASC 1204 MAT 2204 Programming for DASC (JAVA) Calculus I Role of Data Science in Today'sDASC 1223 DVSC 1013 Intermediate Data Science World UofA NorthArk Year 2: Semester 1