August 16, 2016 at 11:00 am in YR-459
Dehui Li
Advisor: Dr.
Harry Zhou
Abstract
The traditional teaching environment is usually thought to be that of a
classroom: a single teacher giving lectures to a group of students who are
expected to use their notes and textbook to prepare for periodic examinations
and demonstrate that they have learned. An obvious problem with this approach is
that everyone receives the same lecture within a fixed time frame. With so many
students with different levels of understanding, it is impossible for the
teacher to provide tailored lessons to every student. Most of the E-learning
systems lack artificial intelligence and merely present the content materials
without evaluating the students’ comprehension and competence. The lecture
materials in traditional e-learning systems are presented in a predefined order
and within a certain timeframe regardless the student’s understanding of the
topic being discussed. They cannot handle a large and potentially diverse
student population. In responses to these challenges and difficulties, this
dissertation proposes, designs, implements, and tests a course delivery system
that provides individualized lessons to students based on their levels of
comprehension, progress and weakness. By analyzing the student’s statistic data
on his/her background and intellectual ability, and dynamic data collected
during a lecture session in real time, our system is able to provide
personalized lessons with different levels of difficulty. An Intelligent and
Effective E-learning System(IELS) evaluates the student’s real time learning
activity, determines their competency level, analyzes their progress, and
selects appropriate teaching materials. Good students can finish a lecture unit
much faster than others, while the students at the introductory level may take
longer. All students, hopefully, can meet the lecture objective at the end. A
variety of students with different backgrounds and abilities can benefit from
this effective, efficient and individualized pedagogical strategy. IELS
employs three artificial intelligent components in its design: a knowledge base,
a case base and a fuzzy reasoning mechanism. The knowledge base captures the
expertise of domain subject experts and uses it to dynamically construct a
lecture content based on the student’s competency. The case base
enables IELS
to recognize similar situations and recall and adapt its past course content for
students with similar characteristics. The fuzzy reasoning component
allows ITS to conduct approximate reason and handle
vague and imprecise terms. By combing expert’s knowledge, analogical reasoning
and fuzzy reasoning,
IELS
demonstrates its adaptive ability to deliver personalized courses to students.
To show its benefits and feasibility,
IELS
has been tested in the domain of computer science courses, but its design and
structure promise to be domain-independent. Without any structure changes, any
domain subject expert, such as in the fields of SAT, GRE, MCAT or any college
courses, can input their lectures with ease. The potential applications of IELS
are promising and unlimited.