Learning Analytics with DEEDS simulator Benefits and Challenges of Data Sharing Mehrnoosh Vahdat, ICE PhD student at UNIGE & TU/e Member of LACE Project, Infinity Technology Solutions September 16, 2014
Jun 01, 2015
Learning Analytics with DEEDS simulatorBenefits and Challenges of Data Sharing
Mehrnoosh Vahdat, ICE PhD student at UNIGE & TU/eMember of LACE Project, Infinity Technology Solutions
September 16, 2014
Outline
• Research plan
• Learning Analytics (LA) with DEEDS▫ Goal▫ DEEDS –a shared resource▫ Student’s learning process with
DEEDS ▫ LA role
• What to share?▫ Data▫ Features
• Who benefits from data sharing?▫ Challenges
• References
Research plan
LA/ EDM
Higher Education
Electronic Engineering Courses: Logic Networks
Simulator
Data collection and pre-
processing
Prediction of learning outcome from
interaction data
IndustrySchools
Analyzing learner’s behavior
Goal
• To extract the profiles of students activities, performed during the training sessions of a course of logic networks
• To relate such activities with the students’ performance at intermediate verification tests
• To explore students learning behavior on system while using Deeds simulator
• To extract non-trivial patterns from students’ interaction
• To assist instructors to be aware of students learning process
DEEDS
• Stands for: Digital Electronics Education and Design Suite
• Is an interactive simulation environment for e-learning in digital electronics
• Provides learning materials
• Asks to solve varied-level problems
DEEDS – A freeware and a shared resource
• Deeds is free to use for academics
• It has been and it is used now in several European universities and project assignments have been shared among European schools (within the European Union LeonardoDaVinci NetPro project)
• Deeds educational materials have been translated and published in English, Italian, Turkish, Spanish, Catalan.
Students’ Learning Process with DEEDS Simulator
Components
LA role
• Prediction of students’ learning outcome from their activities in each session
• To understand which learning behavior is effective in the outcome
• To distinguish the students who need more attention in early sessions of the course
• To understand which course content/ exercise is critical
• To provide help in-time
What to share?
• What data level to share concerning the data anonymity?
• Which features are critical in prediction?
• Which prediction method (to predict students’ grades) is effective for these data?
Data
• Activity logs from system
• Data from the questionnaires:
▫ Demographic data: consent, general, motivation, background knowledge, ICT literacy, learning style.
▫ Data from students feedback
• Data from observation and semi-structured interview
• Group assessment per session
• Final grade at the end of the semester
Feature Extraction
Which features are critical in prediction?
• Samples of students’ activities :▫ Text-editor: The time students spend on writing their answers.▫ Image: Students work with images of simulation imported from the
tool.▫ Circuit-simulator : Students work on an exercise with the circuit
simulator.▫ Timing-diagram: Students run the circuit simulator.▫ FSM-Simulator: Students work on Finite State Machine Simulator.▫ Browser-exercise code: Students study the exercise.▫ Warning: they might have taken a wrong action.
Who benefits from data sharing?
• Researchers: ▫ To benefit from simulation-based critical features, and prediction methods,
to develop recommendation engines integrated in digital electronics simulators
• Teachers: ▫ To plan their lessons based on students’ needs and their effective activities,
to help students in-need in early sessions of the course, and help them avoid most frequently mistakes.
• Students: ▫ To get recommendations about activities and resources, receive more
personalized help.• DEEDS/ digital electronics simulator developers:
▫ To improve simulators and adapt it to the students’ needs
Challenges?
• Cost:
▫ Of applications, methods to obtain meaningful data.
• Data
▫ Interoperability: To bring all data levels together
▫ Reliability: user’s role in activity data, trial and error or decision making?
▫ Context and time: needs lots of work to make sense of unorganized information
• Ethical obligations:
▫ Privacy and anonymity(Bienkowski, Feng, & Means, 2012; del Blanco et al., 2013; Gyllstrom, 2009)
These slides are provided under the Creative Commons Attribution Licence: http://creativecommons.org/licenses/by/4.0/. Some images used may have different licence terms.
Learning Analytics with Deeds simulator: Benefits and Challenges of Data Sharing
by Mehrnoosh Vahdat
was presented at Learning Analytics Data Sharing – LADS14 Workshop at EC-TEL.Graz - 16th September 2014
http://goo.gl/ouywVU
@MehrnooshV
References
• Baker, R.S., Corbett, A.T., Koedinger, K.R. (2004). Detecting student misuse of intelligent tutoring systems• Baker, S.J.d., Yacef, K. (2009) The State of Educational Data Mining in 2009: A Review and Future Visions.• Bienkowski, M., Feng, M., Means, B. (2012). Enhancing Teaching and Learning through Educational Data Mining and Learning
Analytics: An Issue Brief • Chatti, M.A., Dyckhoff, A.L., Schroeder, U. and Thüs, H. (2012) ‘A reference model for learning analytics’, Int. J. Technology
Enhanced Learning, Vol. 4, Nos. 5/6, pp.318–331.• del Blanco, A. et al. (2013). E-Learning Standards and Learning Analytics: Can Data Collection Be Improved by Using Standard Data
Models?• Donzellini G., Ponta D. (2007) A Simulation Environment for e-Learning in Digital Design. Trans. on Industrial Electronics, vol. 54,
no. 6: 3078—3085.• Glahn, C., Specht, M., Koper, R. (2007) Smart indicators on learning interactions. In: Duval, E., Klamma, R., Wolpers, M. (eds.) EC-
TEL 2007. LNCS, vol. 4753, pp. 56–70. Springer, Heidelberg.• Gyllstrom, K. (2009). Enriching Personal Information Management with Document Interaction Histories: A Thesis• Gyllstrom, K. (2009) Passages through time: chronicling users' information interaction history by recording when and what they
read, Proceedings of the 14th international conference on Intelligent user interfaces, February 08-11, 2009, Sanibel Island, Florida, USA [doi>10.1145/1502650.1502673]
• Romero, C., Ventura, S. (2007). Educational Data Mining: A Survey from 1995 to 2005.• Romero, C., Ventura, S. (2010) Educational Data Mining: A Review of the State-of-the-Art.• Siemens, G., Baker, S.J.d. (2010). Learning Analytics and Educational Data Mining: Towards Communication and Collaboration