higher ed 2.src verticals

Dell Technologies

I love college—especially the learning part. Somewhere near the middle of my technology marketing career, I decided to earn a degree in science. I can’t say that navigating the various paper and electronic admissions and onboarding processes was easy, nor that everyone in my online classes was ready for online learning. But it worked for me, and now I find myself longing for yet another education adventure. But the choices — how will I decide which institution and how will they choose on me? Do I still have the discipline to succeed?  

Administrators have grappled for decades with how to attract and retain students. And today, academic leaders often find that they need to increase their efforts in this area: Industry research shows that enrollments in U.S. higher education institutions have fallen steadily for the past eight years, and are now down by more than 2 million students since 2011.1  
 

One clear path to growth is through online programs, but that too brings challenges. While shifts to online teaching enable institutions to have a national and even international presence like never before, the selection process itself may become more complex. And even the most independent and self-driven students accustomed to classroom learning sometimes find that the path to online learning has a steep curve. Educators and administrators need better methods for selecting candidates and understanding and removing roadblocks to their success.   

At a broader level, all universities want to rise to the highest reputation levels. For many institutions, this means striving to become a “Tier 1” institution known for excellence in research. To achieve or maintain this status, university staff and administrators need to identify and eliminate gaps in their research programs and capabilities. 

These are among the challenges that can be addressed with systems and applications that leverage artificial intelligence (AI). For universities, AI is rapidly emerging as a powerful tool that helps attract and retain students, understand faculty and student needs, streamline processes, enhance the campus experience, and more. 

AI use case examples 

Let’s look at how AI-driven applications help colleges and universities address some of these challenges. 

Student acquisition 

Universities use AI to identify students who are most likely to succeed in a higher education environment. In their efforts to attract and enroll new students, universities are using AI and data analytics to determine which students might need additional guidance and assistance to navigate the available resources and apply for enrollment and financial aid. 
  

Some are delivering AI-driven chatbot services to answer questions and help students find their way through complex enrollment and transition processes, then using the data from these chatbots to improve the student experience even more.  

Student retention 

AI-driven applications and data analytics techniques are helping universities understand how to help current students succeed. They do this in part by identifying students who are having trouble academically and might be at risk of dropping out. Over time, the algorithms used in these machine learning applications get steadily better at zeroing in on students who need assistance. 
 

Another tactic for retaining students tracks the number of courses a student drops and the number of times a student changes majors. These metrics can help counselors determine if a student needs one-on-one academic advice and new prerequisites. Similarly, algorithms might track the number of times a student seeks assistance from online teaching tools that supplement classroom work. 

Raising academic reputation 

Retaining students helps elevate academic reputation, as does conducting ground-breaking research in areas such as life sciences, chemistry materials, molecular dynamics, and natural language processing. As a result, institutions can rise higher in national and international rankings important to student and faculty recruitment and ongoing grants and other funding.  

Gaining insights into network operations and student life 

Universities increasingly use AI to gain insights from data on network operations and student life. For example, IT teams use AI and network analytics to pinpoint dormitories with high numbers of online gamers who require more network bandwidth. This same information can help universities identify students who may want to participate in collegiate eSports, a rapidly emerging field that gives students another avenue of extracurricular success. 

This focus on analytics also extends to the playing field, where athletic departments and teams now use AI to analyze game films showing their performance and that of opposing teams. These applications can help coaches and players recognize patterns, such as players who are often out of position or formations that an opposing team uses for specific plays.  

Optimizing the campus experience 

One way that higher education institutions enhance student experience is by using AI to analyze network traffic so they can remove bottlenecks that impact the user experience. Similarly, AI-driven analytics can help university administrators understand the need for additional facilities located strategically on campus. What food options are students are students choosing–including where and when? Why is one coffee shop so much busier than the others? AI can help answer questions like these to enrich the student experience.  

 
Considerations for capitalizing on AI 

While the use cases vary widely, they share similar challenges. To better understand these challenges and how institutions can address them, I turned to Wynell Jenkins, a Dell Technologies HPC solution architect focused on higher education. Wynell outlines several considerations for colleges and universities that want to capitalize on the opportunities made possible through AI and advanced data analytics. 

Cultivate the big vision.  

“Work to understand your desired outcomes,” Wynell advises. “What problems are you trying to solve? What conclusions are you trying to reach? How fast do you need answers?” 

Wynell suggests starting with a focus on outcomes and then working your way backward to design algorithms that will answer the right questions and help you see the actions you need to take.  

“Start with the big picture, and then get tactical,” he says. “Think about the data you need to capture and analyze and how you are going to train your algorithms.” 

Bring data together. 

AI applications invariably leverage data from different systems and collection points. That’s why it’s essential to think about your data strategy from the outset, and how you’re going to leverage data from different silos spread across the university environment, Wynell says.  

“Consider building a centralized data lake, where different applications can access the data from different systems,” he says. “Develop a strategy for data curation and management. Determine what data gets analyzed at the edge and what data gets analyzed in an on-campus private cloud or off-campus public cloud.” 
 

Leverage HPC infrastructure. 

AI applications leverage data-intensive and compute-intensive machine learning and deep learning algorithms. It takes a lot of high performance computing to develop and train these algorithms, so it’s essential to put the right compute, storage, and network infrastructure in place. This same infrastructure supports the needs of the growing numbers of academic researchers who require access to HPC clusters. 

“Academic research often needs HPC systems with the latest CPUs, optional accelerators, fast interconnects, high-performance storage, and enterprise-class solutions for data protection, data security, and data privacy — along with management tools that make it all easy to use,” Wynell says. “AI development projects require these same resources to train the neural networks and machine learning models that enable AI applications.” 

Cultivate your expertise. 

Beyond technology, it’s essential to focus on the people side of the AI journey, Wynell advises.  

“Make sure you have access to competent, qualified data scientists who understand how to put the technologies and frameworks for AI in place to gain insights from data across the campus,” he says. “You can’t get there without the right people and skillsets.” 
 

Key takeaways 

AI offers benefits across academic institution operations. It helps digitally-driven universities address problems that can’t be solved easily with conventional, manual approaches. 

Going forward, AI will increasingly be one of the keys to success in higher education, in terms of acquiring and retaining students and faculty, increasing funding for research, improving administrative processes, and enabling a richer campus experience for all stakeholders. 

To learn more 

For another view on AI in higher education, see the Dell Technologies article “Using AI to Identify & Help Struggling Students.” This article explores using AI to analyze and find patterns in massive datasets and identify students who are struggling and likely to drop out.  

  1. National Student Clearinghouse, “Fall Enrollments Decline For 8th Consecutive Year,” December 16, 2019.

Copyright © 2020 IDG Communications, Inc.