Emerging Technologies 2019 In Higher Education Administration - PowerPoint PPT Presentation

Emerging Technologies 2019 • In Higher Education Administration Corinne Picataggi and Jason Pully

Trending Topics in HigherEd IT • Blockchain • Chatbots • Machine Learning / Artificial Intelligence – W&M experimentation and application • descriptions, potential use cases, considerations 2

What is Blockchain? • A distributed ledger that validates components by consensus of participants – The chain is an expandable list of securely connected records which significantly reduces the opportunity for unauthorized access or manipulation. – Decentralized management of data – Data ownership transformation 3

Blockchain Flow custom diagram here 4

Blockchain: Use Cases and Benefits • Stackable certificates • Digital diplomas that are verifiable by potential employers and other institutions • Owner autonomy over their records • Globalizing education across multiple providers • Sharing research 5

Blockchain Flow – Degree Completion custom diagram here 6

Blockchain: Considerations and Risks • Identifying appropriate use cases • Protecting endpoints: input and output • Ensuring the security required to write to the chain is sufficient (managing encrypted keys) • Developing solid test cases • Standards and regulation 7

Blockchain in Use: MIT Digital Diploma • 2017 Pilot – 111 graduates received digital diplomas • Students can share diploma immediately with whomever they want Citation: Digital Diploma Debuts At Mit Elizabeth Durant-Alison Trachy- | Office of Undergraduate Education - http://news.mit.edu/2017/mit-debuts-secure-digital-diploma-using- bitcoin-blockchain-technology-1017 • No enrollment verification process to validate diploma 8

What is a Chatbot? • An application that conducts a conversation verbally or through text – Evolution: From Question & Answer to Reasoning and Contextual Decision Making • Old: Siri, Alexa, Google Assitant, Cortana • New: Microsoft's Xiaoice, Retail Markets 9

Chatbot: Use Cases and Benefits • Build student engagement by extending service and outreach • Improve employee service centers Citation: Five Reasons Why Chatbots Are the Future Of Customer Service Aakrit Vaish - https://www.entrepreneur.com/article/325830 • Support prospective students and visitors to website by answering questions at their convenience 10

Chatbot: Considerations and Risks • Outsourcing – delegating the message • Exposing the platform – malicious threats • Protecting sensitive information • Language considerations when supporting a global community • Personalization versus Socialization 11

What is Machine Learning / AI? • Relying on a computer system to complete a specific task using patterns and inference, without providing explicit instructions. 12

All About Data • Different than Analytics – Defined outcomes • Machine Learning vs. Traditional Statistical Modeling – Statistical modeling operates on generally known relationships while machine learning can help discover relationships. 13

ML/AI: Use Cases and Benefits • Predict who would buy meal plans • Student success • Financial Aid needs • Tuition forecasting • Donor data • Aggregating datasets based on previous collection • Event attendance and success • Space planning and management 14

Machine Learning / AI: Risks • Explaining – I don’t know ~ the computer did it! • Errors – Identifying errors and preventing recurrence • Ethics – How should data be used? – What data is appropriate to use? – Relationships between admission and success 15

Can we get by without it ML? • Society, and particularly social networking, is creating tailor-made experiences for people. – This is becoming an expectation in society that we are seeing translate into student expectations. • Managing Student Expectations – 97% of students say technologies that support them outside of class are just as important – 87% of students said the tech “savvy” of schools is important when applying Citation: -- Data Sources: Wakefield Study with Ellucian (2017); DJS Research 16

Data Sources • W&M runs 70+ systems, many of which create unique data that can be used in analysis 17

How will it affect decision making? 18

Where is William & Mary? • Descriptive and diagnostic analytics • IT is experimenting with predictive analytics 19

What Has W&M IT Done? • Work Request Team Assignments • Work Request Estimated Hours – About 3000 tickets a year assigned • Goals from Performance Management – Over 8000 goals from supervisors – Create concise goal library 20

Also Applied at W&M • Machine Learning to match incoming students with a Faculty Advisor – Student collaboration project, IT facilitated • Because of our experimentation with machine learning, we were positioned to take what these students had done and operationalize it. We are now using it to assign advisors. Citation: https://www.wm.edu/news/stories/2018/math- undergrads-deploy-algorithm-to-revamp-student- advising-system.php 21

Where would you start? • Data quality and literacy • Start small with something you know and understand. 22

QUESTIONS? 23