Computer Science Education Research (CSER) Group

Slides:



Advertisements
Similar presentations
Wednesday, 7 Jan 09 MAA Session Quantitative Literacy across the Curriculum –Kelly E Matthews, PhD student & Student Experience Manager, Faculty of Science,
Advertisements

Active Learning to Classify
A Brief Overview. Contents Introduction to NLP Sentiment Analysis Subjectivity versus Objectivity Determining Polarity Statistical & Linguistic Approaches.
6 th Annual Focus Users’ Conference 6 th Annual Focus Users’ Conference CTE Attendance & Gradebook CTE Attendance & Gradebook Presented by: Natasha Leon.
COMPUTER SKILLS FOR HISTORY TEACHERS * Pressing F1 = HELP *
Student Workload. Questions What is student workload? How does it relate to contact hours and assigned tasks? How does it impact on the way students learn?
Big6™ - Information Skills For Student Achievement
Openings for Progression Dr Lindsay Hewitt. Background to EAL Survey 2009 What we know What we think we know What we don’t know What we wanted to find.
“What’s a Journal?” Ed Foster, Nottingham Trent University.
Using Safari Live Web-conferencing Technology in Technical Writing Courses: Can Nonhuman Software Rehumanize Online Learning Spaces? Dr. David S. Hogsette.
Welcome to Second Grade Parent Night!. My Background Four years experience teaching First Grade and two years teaching Second Grade Received my Professional.
Evaluation and analysis of the application of interactive digital resources in a blended-learning methodology for a computer networks subject F.A. Candelas,
Mixed-level English classrooms What my paper is about: Basically my paper is about confirming with my research that the use of technology in the classroom.
Bruce White Ruth Geer University of South Australia.
1 Milwaukee Mathematics Partnership Program Evaluation Year 4 Results Carl Hanssen Hanssen Consulting, LLC Cindy Walker University of Wisconsin-Milwaukee.
Room 112 Math/Accelerated Math Teacher.  Been in Oxford for 14 years  Have taught math all 14 (other years have taught science and geography)  Went.
Automatic Syllabus Classification JCDL – Vancouver – 22 June 2007 Edward A. Fox (presenting co-author), Xiaoyan Yu, Manas Tungare, Weiguo Fan, Manuel Perez-Quinones,
Mtivity Client Support System Quick start guide. Mtivity Client Support System We are very pleased to announce the launch of a new Client Support System.
DEALER Distance Education And Learning Repository The First Draft Idea Zlatan Sabic Sarajevo, July 1998 Soros Foundations Open Society Fund Bosnia and.
G ROWING R EADERS & W RITERS A T H OME I NSTRUCTOR : M EGAN C ONROY.
Introduction to Information Retrieval Introduction to Information Retrieval Lecture 15: Text Classification & Naive Bayes 1.
Click on “My Courses”. Please note that only summative assignments can be uploaded on the new virtual campus. Formative assignments are now available online.
Can You Enhance Knowledge and Stimulate Excellence One STEM Unit at a Time? AEA – October 16, 2014 Panel: Evaluating STEM Professional Development Interventions.
IR Homework #2 By J. H. Wang May 9, Programming Exercise #2: Text Classification Goal: to classify each document into predefined categories Input:
Learner experiences of online learning in a blended learning situation: Different cohorts, different needs Benjamin Kehrwald University of South Australia.
Lesson 11: Web Services and API's
CSc 120 Introduction to Computer Programing II
Welcome to “Moodle Part Deux”
Who Goes Where? and why
Would you please mark my text for me
Third Grade Curriculum Night
Lilian Yahng Derek Wietelman and Lauren Dula Indiana University
The ability to communicate effectively in the workplace is essential
March 27 – Course introductions; Adts; Stacks and Queues
Links in the Chain: turning learning analytics data into actions
Learning from Our Students’ Summer Experience Survey
Beyond the technology: intervention evaluation based on captured attendance data C.E.B. Freeman University of South Wales, Faculty of Computing, Engineering.
Parents please have a seat
Information Management course
Steps for Curriculum Mapping
September 27 – Course introductions; Adts; Stacks and Queues
Discord Bot Senior Project
Un</br>able’s MySecretSecrets
Hi everyone, for those that don’t know me, I’m Nicole and my colleague Ian and I are responsible for promoting teaching to physicists and engineers. I’m.
Business and Management Research
© The University of Edinburgh
Course Overview - Database Systems
Forensic Science Unit One
Game Design Mr. Gebhart.
Chuck Berryman Professor and Chair Louisiana State University
[insert Module title here]
Helen Jefferis, Soraya Kouadri & Elaine Thomas
Attributes of Information
Imagine Success Engaging Entering Students Innovations 2009
We Survey Companion Indicators – LASS (Morton, Hammond, Merrillville, Lowell) We Teach (N=256) We Learn (N=4,175)
Before you use this presentation…
National Student Survey 2019
Lesson 11: Web Services and API's
Digital experiences with Moodle at Divine Word University
Minority Science and Engineering Improvement Program (MSEIP)
Employability across disciplines:
Business and Management Research
[insert Module title here]
Transferring to a 4-Year University
The Intersection of Literacy and Digital Literacy in the classroom:
Potential Influence of Prior Experience in an Undergraduate-Graduate Level HPC Course Chris Fietkiewicz, Ph.D. Department of Electrical Engineering and.
Using the Dundee Ready Education Environment Measure to understand the allied health undergraduate student journey Eric Chua, Benjamin Soon, May Lim.
Graduate Outreach Services:
Confidence-building assessment for Level 1 Computing and IT students
Back To School Night 2019 – 2020 School Year.
Presentation transcript:

Katrina Le, Hamid Tarmazdi, Rebecca Vivian, Katrina Falkner, Claudia Szabo and Nick Falkner Computer Science Education Research (CSER) Group Directing Teacher Focus in Computer Science Online Learning Environments

Increasing pressures of enrolments Figure: Commencing Domestic IT enrolments in Australia (2001-2015). http://blogs.adelaide.edu.au/cser/2017/02/15/a-look-at-it-and-engineering-enrolments-in-australia-updated/

Increasing pressures of enrolments We are seeing a resurgence in Computer Science enrolments. Increasing pressure on academic staff, and on teaching resources. Academic staff are also under increasing workload pressures, both in expected research output and the breadth of expected curriculum. CRA, Computing Degree and Enrollment Trends, 2012-2013 Taulbee Survey, http://archive2.cra.org/uploads/documents/resources/taulbee/CRA_Taulbee_CS_Degrees_and_Enrollment_2012-13.pdf

Online discussion forums Online discussion forums provide a valuable method for students to engage with their colleagues and academic staff. Asynchronous, and flexible. But can easily get lost amongst the hundreds of messages and requests.

Text Classification “Thanks guys, it helped a lot” “I did an insertion sort but did not use PTList” “These results are also on the website” Non-Critical “Can there be a few more days given for assignment 3?” “Still not working for me” “I cannot access the submission system too” “I am disappointed that no-one from the group has yet responded”. Critical

Research Questions To what extent can a Naive Bayes classifier effectively classify forum posts in Computer Science courses to determine if a post requires intervention? Simple technique, cheap in terms of storage, computationally efficient. But, typically applied for larger discussion items. Assumes conditional independence. Can this still be effective in the context of discussion forums?

Research Questions To what extent can a Naive Bayes classifier effectively classify forum posts in Computer Science courses to determine if a post requires intervention? Can we improve the accuracy by applying negation, parts of speech (POS) tagging, word stemming or removal of stop words ?

Text Classifier Applied tag stripping to remove HTML tags Preprocessing modules able to be applied dependent on experimental conditions, including: Removal of stop words Part of Speech tagging Negation identification Word stemming

Experimental Data Discussion forum data across multiple undergraduate courses. Removed user/timeline information. 4,666 sentences. 1,978 manually identified as critical. 2,688 manually identified as non-critical. Total word count of 83,176.

Results

Analysis Correctly identified as Critical: “I’m quite confused.”, p=0.965 “So what exactly are we supposed to do in the Test file?”, p=0.999 Incorrectly identified as Critical: “Now my code passes example 8.”, p=0.7318

Analysis More importantly, though, is to consider those sentences that are Critical but identified as Non-critical: “Please help us to solve this problem, thank you very much.” , p=0.885

Conclusions Identified as an efficient technique, NB has also been shown to be effective in the classification of critical discussion forum posts. Shows promise to be combined with other indicators of student engagement, such as patterns in resource access and overall participation to direct academic attention.

Computer Science Education Research Group blogs.adelaide.edu.au/cser FIN