Search for: All records

This work-in-progress (WIP) research-to-practice paper describes a work in progress by the authors to integrate appreciation of privacy, ethics, regulatory compliance, and research into Senior Project capstone experiences for Electrical and Computer Engineering. The student work focused on data quality assurance and de-identification topics to enhance quality, accuracy, completeness, consistency, and timeliness. Real-world data protection regulations grounded projects to meet ABET EAC Criterion 3 requirements for Student Outcome 2. Students explored the topics in a Project-Based Learning (PBL) format as a part of their senior project. In addition to implementing PBL, our focus for the senior project capstone is securing as many industrially sponsored projects as possible. This paper focuses on a few senior projects that are PBL, sponsored by industry, and emphasize data quality assurance and privacy protection techniques. We present a framework that meets assessment needs and uses project-based learning on a current topic of interest. The student findings offer insights into the theoretical and practical challenges and opportunities of implementing data quality assurance and de-identification techniques across different domains. 

This is a work-in-progress paper. The flipped classroom (FC) model is a well established teaching strategy dating to 1970’s practices in the Soviet Union. FC has two decades of use in post-secondary education since it was proposed by Lage et al. However, breaking studies find no academic improvement with FC model among minority students. Rather, it distances at-risk students. Indeed, certain demographics prefer authoritative over dialogic instruction style. We are motivated to determine FCs effectiveness with students at a medium-sized Hispanic Serving Institution (HSI) and Minority Serving Institution (MSI). For one of our NSF grant activities, we piloted two variations of the flipped classroom model. The key idea is that literature finds that FC classes need better regulation of underperforming students. Generally, the FC models in our work included peer-instruction, active learning, recorded lectures, and pre-assessment quizzes. There were no post-assessment assignments or traditional homework. Some sections employed Just-in-Time-Teaching, and careful selection of groups according to skill (within-class homogenous grouping). Other sections experimented with diversity and inclusion-based grouping and project-based learning. Students at the university are non-traditional, a term used to describe individuals who meet some of the following criteria: having a significant gap between post-secondary education and high-school graduation, being financially independent from their parents, having dependents, and working twenty or more hours per week. 60% of the individuals at our campus are Pell eligible. We study an intersectional inequality: wage-based work is disinclined to accommodate students attending lecture during the work day, and minorities may not prefer dialogic instruction. We analyze student attitudes since Fall 2020, among tens of class sections and hundreds of students. Class sections in the study are upper-division core courses in Computer Science, Computer Engineering and Electrical Engineering. Data is collected from mostly online sections during the COVID-19 pandemic. A pre- and post-surveys were administered collecting demographic information and student attitudes. Hispanic/Latino(a) students found videos to be a complete study medium—that it was not required to seek out third-party materials to prepare for class. They found the class to be more engaging, and self-identified that they could identify previous concepts important to the task at hand. Results were surprising because there were no statistically significant differences with a general population’s exposure to FC. Hispanic/Latino(a)s find the FC model described in our work engaging and effective. 

In the oil and gas industry, exploration is largely dependent on the study of the subsurface hundreds or thousands of feet below. Most of the data used for this purpose is collected using borehole logging tools. Although sophisticated, these tools are limited as to how precisely they can measure the subsurface in terms of vertical resolution. There is one method of studying the subsurface that provides unlimited vertical resolution – core samples. Although core samples provide scientists the opportunity to generate a full, continuous data set, lab analysis work is normally done at one-foot intervals, as anything more would be prohibitively expensive. This means at best, a representative data set is generated. However, if the subsurface is not homogeneous, it is difficult to generate a representative data set with lab analysis done at one-foot intervals. This is a void that artificial intelligence can fill. More specifically, a properly trained neural network can analyze high-resolution core images continuously from top to bottom and generate a continuous analysis. It is also important to note that geologic interpretation tied to core analysis can introduce human error and subjectivity. Here too, a properly trained neural network can generate results with extreme levels of accuracy and precision. One core analysis expert believes that core analysis done manually is flawed about 70% of the time. This flawed analysis can result from lack of experience and or a lack of knowledge of the geologic formation. We are not the first to attempt to analyze core samples with vision algorithms. A group of Stanford researchers used micro-computed tomography (micro-CT) and Scanning Electron Microscopy (SEM) images of core samples to characterize the porous media. While promising, SEM and micro-CT imaging is expensive, and more importantly it is not a standard practice in the oil and gas industry to collect these types of images, making these images rare. One other work applied convolutional neural networks to a GIS based regional saturation system, but our work is significantly different. It is well known that training a neural network requires abundant data, thankfully with the method of core analysis we are proposing that will not be a problem. Through industrial partnerships we’ve obtained hundreds to thousands of core images sufficient to train a neural network, as well as core interpretations tied to those images coming from a core analysis expert with over 40 years of experience. We are the first to propose automatic hydrocarbon saturation as well as lithology prediction from core slab images. We propose the use of convolutional neural networks to analyze core samples at a single site. We plan to conduct experiments using a variety of neural networks to determine the best practices, and explore how such a service can be offered to the industry via the software-as-a-service paradigm. In the past, automated analysis through core slab images has not been possible simply because images of the required resolution were not common, but that has changed. If implemented successfully, this proposed method could become the new standard for core evaluation. 

Our work describes the best-practices and findings for a recent NSF IUSE HSI HRD grant. Its overarching goal is to drive an institutional change where the University proactively places students in internships with local industry partners. Students at the University are non-traditional, minority and low-income. They often working twenty to forty hours a week in non-curricular jobs. The Grant Program fully or partially subsidizes compensation for interns through financial aid scholarships. It aims to replace non-curricular work with relevant, real-world engineering experiences. This in turn improves their prospects to find jobs post-graduation. Modern students work while going to school. A small amount of work—less than fifteen hours a week—is beneficial. However, beyond twenty hours a week has a negative impact. Hispanic/Latino(a) students work twenty to forty hours a week, more than any other demographic. This workload affects attendance, GPA, and utility, resulting in poor workforce placement. Academia must concede that work comes first for under-represented students. Universities must take steps to supplant irrelevant work experience with industry internships. Participants of this program received relevant internship/work experience, had better retention rates due to perceived utility of their degree. In the long term we expect timely to graduation due to participants taking internship units as credit toward their degree. Students learned of the internships from faculty members soliciting applications to the program, supported by the grant. Executing the MOU between the University and industry partners took considerable effort and is a major barrier to executing formal partnerships between internship hosts. One MOU is still in negotiation since the start of the program. Despite some student participants reporting prior internship experiences, no one involved in the program would have found an internship this academic year without help from the Grant Program. Some students claimed to have submitted from twenty to fifty applications and the Grant Program was the only internship that called for an interview. Quality of internship varied from corporation to corporation. Universities must carefully monitor the feedback of participants to ensure that the individual goals of the participants are being met. Finding corporations that are willing to invest time in mentorship of students is a critical component to ensure student satisfaction. Even so, regardless of internship quality, participants would not have found internships if not for the Grant Program. According to the participants, internships are an opportunity to network and build lasting professional connections. While students may be unable to turn every internship into a full-time position, each experience will give them something much more valuable and long-lasting: relationships with professionals and co-workers. The connections they make during their time at an organization can be stepping stones to their next opportunity.