Adaptive Learning

AI technique (Almohammad et al.,2017)

Bayesian networks (Dziuban et al., 2016)

Genetic Algorithms

Decision tree

Hidden Markov model

Fuzzy logic (FL)

Downside

Core elements (Dziuban et al., 2016)

Definition(ELI, 2017; Komar & Troka, 2015)

System

Purpose

time (Dziuban et al., 2016)

detailed mapping (ELI, 2017)

cost

content development

lack of familiarity (Daines et al., 2016)

vague algorithm effectiveness (ELI, 2017)

cultural differences between online and F2F

multiple intervention

rapport: instructor, student (importance of 1st and 2nd classes)(Daines et al., 2016)

interaction

media

technologies

classroom training

feedback (O'Connell, 2017; Straumsheim, 2017)

syllabus change

assessment

incentives

Algorithm- "same aspect of learning will never be measured with any algorithm..."

instructor's assessment/intervention

faculty active preparation

training

analysis & interpretation

intervention

strategies for the data use (real-time info in dashboard)

Data
(Komar & Troka, 2015)

Report

Variables

Analysis

Assessment quality

Delivery method

Effectiveness of content

monitoring student's progress

instructor's action

Weekly analysis to modify content and assessment (Komar & Troka, 2015)

avg final score, ave weekly learning growth rate..

total amount of time spent in the platofrm

percentage of assigned learning activities completed

num of interactions in the platform course retention rate

course persistence rate

decrease dropout rate (Milano, 2017)

increase retention rate (Milano, 2017)

improving performance/outcome & learning process (Dziuban et al., 2016)

improve pedagogy

reduce gaps in understanding (ELI, 2017)

interactive support when problem solving (dreambox, na)

great flexibility & multiple paths to achievement with reduced time (Dziuban et al., 2016)

customized presentation

build strong foundation skills (e.g. "unit zero") (Komar & Troka, 2015

evidence of knowledge

customized learning path

competency-based

data-driven

self-paced

real-time information

personalized learning

knowledge progress/process

non-linear

Learning path

User profile(Almohammad et al.,2017; Howlin & Lynch, 2014; Komar & Troka, 2015)

desired learning objectives

prior knowledge/experience

psychometric cognitive information/ability (perceptual speed, processing speed, working memory capacity, reasoning ability, verbal ability, spatial ability and other cognitive abilities)(Forsyth et al., 2016)

characteristics

"intellipath"(Daines et al., 2016)

startpoint, "zero unit"

"learning map"

Learning model (Almohammad et al.,2017)

contents & assessment development

effective instruction based on learner behavior

learning nodes & map

Platform (Howlin & Lynch, 2014)

content

curriculum

Faculty dashboard

Individual performance

data-driven decision

lecture driven

content layout

weakness & strength

classroom activities

assignment

1:1 interaction

personnel (Dziuban et al., 2016; ELI, 2017)

Project manager

Content experts

Data Analysis, personalized learning team

Technologiest

Faculty training team

Business manager

ID

Faculty

Vendor (Dziuban et al., 2016)

CogBooks

SmartSparrow

Knewton

Realizeit

...

Acrobatiq

Type (Dziuban et al., 2016)

Open

Hybrid

Closed

Structure

Interactive content

Curriculum - hierarchical structure of knowledge

Assessment

Challenges

limitation in course, entry-level (ELI, 2017)

impatience with technology (Daines et al., 2016)

learning style (Surjono, 2014)

personality, attitude, behaviour

Visual, auditory, kinesthetic, global, sequential

big data, detailed design, quick adaptation, Feedback mechanisms, AI (Holz, 2017)

click to edit

'granular' level (Kerr, 2017)

preferences for specific learning materials (Forsyth et al., 2016)

click to edit

click to edit

Benefit (ref)

Students

Reseacher and designer

Instructor

Revising/adding lessons for student's mastery

Save time and enables more meaningful interaction with students

Empowering teachers with granular level information

More productive interactions with instructor

Reducing the risk of falling behind and giving up

Produce scientific data on learner's behavior

Provides focused remediation based on each student's performance

Personalized learning experience

Maximizing student's learning efficiency

Prepares students more effectively for lectures and course exams

Various methods and media

Enhance engagement with the material