Learning Analytics

Lecture: Learning Analytics (WS 19/20)

Semester: Winter Semester 2019/20

Lecture language: English

Exam language: English

Exam type: Oral examination

Maximum number of participants: 30

About this Course

Learning Analytics (LA) has attracted a great deal of attention as practitioners, institutions, and researchers are increasingly seeing the potential that LA has to shape the future technology-enhanced learning landscape. LA is an emerging data science field that represents the application of big data and analytics in education. It deals with the development of methods that harness educational data sets to support the learning process. LA is an interdisciplinary field involving competences from computer science, cognitive psychology, and pedagogy. It leverages various computer science methods. These include statistics, big data, machine learning, data/text mining, information visualization, visual analytics, and recommender systems. The first part of the course will provide a systematic overview on this emerging field and its key concepts through a reference model for LA based on four dimensions, namely data, environments, context (what?), stakeholders (who?), objectives (why?), and methods (how?). In the second part of this courses, we will discuss various methods and techniques required to develop innovative LA systems, in relation to each dimension of the LA reference model. In the last part of the course, current topics and trends in LA research will be presented and discussed in invited talks. The presented methods and technologies will be further investigated and applied in small student projects carried out throughout the course. The course topics will include:

  • Learning analytics and related areas (e.g. educational data mining)
  • A reference model for learning analytics
  • Big Data (Hadoop ecosystem)
  • Learner modeling
  • Ethics and privacy in learning analytics
  • Assessment and feedback
  • Machine Learning / Data Mining (classification, clustering, association rule mining)
  • Recommender systems
  • Information visualisation and visual analytics
  • Current topics in learning analytics (invited lectures)

Getting credits for this course requires a successful completion of all assignments, project, and oral exam at the end of the semester. The final grade will be calculated as follows: assignments and project (50%) and oral exam (50%).

Target audience

  • Master Applied Computer Science
  • Master ISE

Date and location

Lecture:

  • Wed, 12:00 – 14:00
  • LK 052
  • Starts on October 16, 2019

Lab Session:

  • Thu, 10:00 – 12:00
  • LC 140
  • Starts on October 17, 2019

Prerequisites

  • Interest in data science and/or learning technologies.
  • High motivation and commitment.

Registration

Due to didactical methods, we have a limit of 30 students for this class (first come first serve). To register, please send an email to Dr. Arham Muslim with your contact information, your study program, and if available your knowledge/experience in Data Science and Learning Technologies. If the maximum number of participants is reached, we will use a waiting list.

Organization

Lecturers

Prof. Dr. Mohamed Chatti (Lecturer)

Dr. Arham Muslim (Teaching Assistant)

Student Projects

Open University Learning Analytics

Group name: Data Titans

Group members: Aashish Agarwal, Sameh Frihat, Seyedemarzie Mirhashemi, Shoeb Joarder

Project description:

The project idea is to develop an intelligent grade prediction system called 'Open University Learning Analytics' that analyze students' performance and predict their results (e.g. Distinction, Fail, Pass, Withdrawn) by providing real-time analysis and determining factors affecting the students' results.

Links: GitHub, Live Demo

Student Educational Performance Analysis (grade prediction)

Group name: Orchids

Group members: Farnoosh shahabbaspour, Maral Goudarzi, Meijie Li, Mohaimn Al-Alshekh Alsagara

Project description:

The tool (Student Educational Performance Analysis SEPA) was developed in order to explore and predict the educational performance of students,  by applying machine learning algorithms on a prepared dataset to perform some initial data visualization. 

Links: GitHub, Live Demo

Analysis and prediction of students' academic performance

Group name: Group 3

Group members: Marleen Matjeka, Germán Calleja Rider, Mengfan Liu, Xiaozhong He, Boyuan Han

Project description:

The project idea is to visualize the relationship between some feature data and student performance. Then, determine which feature has a greater impact on student performance. Machine learning algorithms were used to learn and analyze student feature sets. 

Links: GitHub, Live Demo

I-Recommend

Group name: Data Miners

Group members: Furough Zarei, Swarna Sri Teja Rampalli, Sriram Anil Kumar, Elham Valipour, Joyce Kandja

Project description:

This tool allows to identify the most frequent courses taken by the students and provide course recommendations based on the courses chosen by the students. Moreover, it provides effective visualizations to give insights into existing student data.

Links: GitHub, Live Demo

Course Recommender (CoRe)

Group name: H.A.N.A

Group members: Nina Laabs, Hoda Ghanbarzadeh, Anusha Bangaru, Hasan Halacli

Project description:

Course Recommendation is a web application built for the ISE students of the University of Duisburg-Essen to predict their grade and recommend the course based on the grade. This project focus on giving students an opportunity to analyze their study method based on survey questions and improve their methodology to achieve good grades in their desired courses.

Links: GitHub, Live Demo

Tag Recommendation system (TRS)

Group name: IPT

Group members: Aalli Mahmood, Atefeh Safarkhah, Mohammad Armoun, Muhammad Dawar 

Project description:

Asking questions is part of the learning process. Therefore, in this project, we want to make sure that every question will be answered through recommending proper and related tags for a question in order to guide the question to the right points, make the question clearer for the community, and make search more precise and faster.

Links: GitHub, Live Demo