LASI 2018 Workshops & Tutorials
We are pleased to present a great LASI'18 program that includes 7 Workshops and 10 Tutorials!
Individuals take a deep dive into one workshop across the entire LASI, while tutorials enable participants to get a flavour of a range of topics.
Each individual can choose 1 workshop and 2 tutorials.
Title: Building predictive models of student success with the Weka toolkit
Craig Thompson, University of British Columbia
Description: In this workshop participants will learn the theory behind several computational methods for machine learning and predictive modelling, including decision trees, naive bayes, k-means clustering and expectation maximization clustering. Participants will be able to use these methods on a supplied set of data using the freely available Weka toolkit. The final session of the workshop will be dedicated to exploring new data (and participants are encouraged to bring datasets they are interested in with them).
Prerequisite Skills/Experience: Participants should be familiar with basic introductory statistics. No programming experience is necessary. This workshop is particularly aimed at individuals who do not have experience with predictive modeling techniques.
Advanced preparation:
- Participants should download and install the latest version of Java for their operating system, as well as the Weka toolkit (version 3.8).
Title: Social Network Analysis: Learning Analytics Perspectives
Sasha Poquet, National University of Singapore
Srećko Joksimović, University of South Australia
Description: Networks of learners, ideas and artefacts are omnipresent in learning settings. Bringing together various elements of the learning process can offer new insights: Are the learners developing social capital? How are they using their social capital to achieve their learning aims? Who are the central actors? What communities of learners exist? How do different participants of the learning process influence the information flow? Do high achievers tend to interact with one another? Can we differentiate between the participants based on their social interactions in the learning system? How can groups of learners be described and compared? These are just some of the questions that could be answered through the combination of network analysis and learning analytics, and this workshop will help you get up to speed with using such a lens for your own data.
This workshop will introduce a suite of conceptual and methodological approaches commonly known as social network analysis (SNA), and their application in the context of learning analytics research and practice. Through a set of hands-on exercises the participants will be stepped through the decisions and techniques needed for the effective use of SNA on various types of learning and interaction data. The participants will learn how to think about constructing networks, will discover various network metrics commonly used in the educational context, and apply them to different sets of interaction data. The participants will conduct network analysis (design, implementation and interpretation) on the datasets collected in both face-to-face and online contexts (synchronous interactions recorded by the wearables; transcript of face-to-face group work; VLE forum interactions; a sample of MOOC log data).
Prerequisite Skills/Experience: To participate in the workshop, you will be required to install R and R-studio prior to the workshop, and bring your computer along. There are no other formal prerequisites. Basic understanding of SNA or some familiarity with basic graph theoretical constructs is an advantage. A few resources to form some common ground will be suggested prior to the workshop. R-scripts helping participants get started with network analysis will form part of the learning materials.
Advanced preparation:
- None.
Title: Text mining for learning content analysis
Jelena Jovanović, University of Belgrade
Description: This workshop will introduce participants to text mining methods and techniques, and enable them to develop working knowledge of text mining in R programming language. During the workshop, we will go through the overall text mining process and examine each of its key phases. In particular, we will start with text preprocessing, then move to the transformation of unstructured textual content to a structured numerical format (i.e., feature creation and selection), thus obtaining a feature set that can serve as the input to a statistical, or machine learning, or a graph-based algorithm for pattern mining or information extraction; finally, we will examine and evaluate the obtained results.
After attending the workshop, the participants should be able to apply text mining methods and techniques to classify or cluster text-based learning content, as well as to extract key terms and topics from such content.
Prerequisite Skills/Experience: Workshop participants should have a basic level of programming experience in R language.
No experience with any form of text mining / text analytics is required.
Advanced preparation:
- Participants should bring their own computer (laptop) to the workshop. They are also expected to have the latest version of R and RStudio installed on their computers.
- In case there are LASI’18 participants who are not familiar with R but would like to attend the workshop, they are encouraged to contact the workshop organizer for advice on developing basic R programming skills required for a productive participation in the workshop.
Title: Unsupervised Machine Learning
Vitomir Kovanović, University of South Australia
Srećko Joksimović, University of South Australia
Description: In this workshop, we will cover unsupervised machine learning (ML), which — alongside supervised ML — represents one of the most fundamental kinds of data mining analyses. We will briefly cover what the different types of unsupervised machine learning methods are, and how are they used in learning analytics research. We will then focus on cluster analysis, which is by far the most popular unsupervised machine learning method, typically used for exploratory data analysis. In this workshop, we will be using Weka, a popular graphical toolkit for machine learning, to analyse several of the real-world learning analytics datasets.
Prerequisite Skills/Experience:None.
Advanced preparation:
- To install Weka data mining toolkit (version 3.8) and download workshop dataset (to be later provided).
Title: Educational measurement and the challenges of inferring learning
Geraldine Gray, Institute of Technology Blanchardstown Dublin
Description: This workshop will focus on the challenges of modeling learning. As a latent construct, learning must be inferred from student activities. We will discuss the inherent challenges of measuring a latent construct, the potential effects of test design and administration choices, and statistical tests of reliability and validity. We will also experiment with, and evaluate, data pre-processing and data modelling choices. Workshop instruction will be conducted in R.
Prerequisite Skills/Experience:Some familiarity with R and R Studio would be beneficial but not required.
Advanced preparation:
- None.
Title: Engineering Learning Analytics Using Learning Science
Phil Winne, Simon Fraser University
Description: Learning science strives to answer at least four complex questions:
- What can learners do to identify and transform information into usable knowledge? How could learners manage their learning environment; find, comprehend and learn information; and recall and apply knowledge?
- What factors – those each learner controls and those instructors can control – modulate learners’ proficiency in these activities?
- What conditions shape learners’ attention to these activities, and their thoughtfulness about and experiments with these conditions? That is, what factors affect self-regulating learning?
- How can learners be guided and supported to self-regulate learning productively?
Developing and testing answers to question 4 is the province of learning analytics. That is, a goal of learning analytics is helping learners better self-regulate learning. This bonds learning analytics and learning science. Importantly, learning analytics spotlights something learning science underemphasizes – successive approximations. Every learning episode has a history and a future. Data can document how this timeline unfolds if the data describe information learners attended to and disregarded, how they managed the learning environment, how they processed select information, and outcomes arising as a result. In short, learning analytics are serial updates about a program of research on learning. Accumulated over the timeline for an individual learner and aggregated wisely across learners, learning analytics draw on historical evidence about what worked and what didn’t as grounds for forecasting options about how to improve future learning activities. Iterative and hopefully productive change is what learning analytics is all about.
Participants in this workshop will:
- explore provisional answers to the four questions posed earlier.
- use nStudy to stimulate thinking about ambient data for learning analytics. (Bring your wifi-enabled laptop pre-stocked with the Chrome browser and the nStudy extension – see Advance Preparation.)
- inventory data describing learning events and patterns in segments of learning timelines.
- engineer and justify learning analytics to guide and support self-regulated learning.
- develop an agenda and form partnerships for integrating learning science and learning analytics.
Prerequisite Skills/Experience:Elementary grounding in learning science or learning analytics.
Advanced preparation:
- Approximately May 01, visit http://www.sfu.ca/edpsychlab/LASI.html for:
- copies of or citations to background materials.
- a sketch of nStudy and instructions for installing it in the Chrome browser.
- Bring to the workshop to share your list of data
- now available or likely accessible “at home” relating to any of the four questions posed earlier.
- you wish you could access.
Title: Multimodal Learning Analytics (MMLA)
Marcelo Worsley, Northwestern University
Xavier Ochoa, Escuela Superior Politécnica del Litoral
Description: Learning does not only occur over Learning Management Systems or digital tools. It tends to happen in several face-to-face, hands-on, unbounded and analog learning settings such as classrooms and labs. Multimodal Learning Analytics (MMLA) emphasizes the analysis of natural rich modalities of communication during situated learning activities. This includes students’ speech, writing, and nonverbal interaction (e.g., movements, gestures, facial expressions, gaze, biometrics, etc.). A primary objective of multimodal learning analytics is to analyze coherent signal, activity, and lexical patterns to understand the learning process and provide feedback to its participants in order to improve the learning experience. This workshop is posed as a gentle introduction to this new approach to Learning Analytics: its promises, its challenges, its tools and methodologies. To follow the same spirit of MMLA, this workshop will include a hands-on learning experience analyzing different types of signals captured from real environments.
Prerequisite Skills/Experience:Basic Python programming and basic learning analytics knowledge.
Advanced preparation:
- None.
Title: Python Bootcamp for Learning Analytics Practitioners
Alfred Essa, McGraw-Hill Education
Ani Aghababyan, McGraw-Hill Education
Description: The workshop will provide a rigorous introduction to the Python language for learning analytics practitioners. Python is the de facto language for scientific computing and one of the principal languages, along with R, for data science. Along with foundational concepts such as data structures, functions, and iteration we will also cover intermediate concepts such as generators, collections, comprehensions, map/filter/reduce, and object orientation. Special emphasis will be given to coding in “idiomatic Python”. We will touch upon Python data science libraries including numpy, pandas, matplotlib, scipy, and seaborn. As a capstone, we will conclude by building a deep learning model for an educational dataset using Python APIs for Keras and TensorFlow.
Prerequisite Skills/Experience:Participants should be familiar with basic programming concepts. The workshop is aimed at individuals who have no previous experience with Python or need a rigorous refresher.
Advanced preparation:
- Participants will need to download and install the Anaconda distribution of Python.
- Course materials and instructions will be available in January 2018 at: https://github.com/alfredessa/lasi2018
Title: Introduction to R programming language
Vitomir Kovanović, The University of South Australia
Description: In this tutorial, we will provide an introduction to R programming language, which is one of the most popular and widely-used technologies for learning analytics research. The focus of the tutorial is to provide the solid foundation of basic R knowledge so that you can then continue your journey through numerous R libraries and capabilities. We will first cover the basic building blocks of R programming (variable declarations, function definitions, control structures) and development within R studio environment. We will then take a closer look at how you can use R to perform some of the most widely used statistical analysis such as t-test, chi-square test, or ANOVA. Finally, we will show how you can use R for running simple linear regression analyses.
Prerequisite Skills/Experience: Basic knowledge of statistics, and some prior programming experience in any programming language.
Advanced preparation:
- To install R programming language and RStudio development environment.
This tutorial session has been cancelled
Title: Learning analytics: Planning for the Future
Rebecca Ferguson, The Open University
Description: Developing and deploying analytics takes time. Meanwhile, courses and institutions change and cohorts of students move on. In this workshop, we shall look at what is on the horizon for education and for learning analytics so that participants can be confident they are working on analytics that will remain relevant for learners and educators.
Prerequisite Skills/Experience: None.
Advanced preparation:
- None.
Title: Developing and Implementing Institutional Data Governance Policies for Educational Data
Grace Lynch, Royal Melbourne Institute of Technology University (RMIT)
Stephanie Teasley, University of Michigan
Description: Data governance is a set of roles and policies that are combined to improve how data assets are handled within an organisation. Data policy should establish a set of rules defining how decisions are made, who is accountable for data management, who can access the data and for what purposes, and how data is extracted, obtained and stored. This policy needs to be integrated with current organisational processes and practices– and local governmental policies– in order to translate into tangible improvements in how data assets are managed and used across the institution.
The interdependency between institutional data and its business processes and applications increases the need for clear guidelines regarding how such data is managed. This tutorial will discuss how to:
- define roles and responsibilities in relation to the governance of institutional data;
- identify the best practices in data management to facilitate its use within the institution,
- provide a secure environment for data access and analysis whilst ensuring the privacy of all stakeholders;
- define the organisational structure of roles, access rights and responsibilities;
- establish clear lines of accountability; and
- assure that the University complies with the current laws, regulations and standards about data management.
Prerequisite Skills/Experience: None.
Advanced preparation:
- None.
Title: Feature Engineering
Ryan Baker, University of Pennsylvania
Description: This tutorial will discuss how to distill and engineer features for data mining. We will cover the process of feature engineering and distillation, including brainstorming features, deciding what features to create, and criteria for selecting features. We will go through examples of feature prototyping in Microsoft Excel and discuss how to scale-up feature engineering to other tools.
Prerequisite Skills/Experience: Familiarity with Microsoft Excel.
Advanced preparation:
- Installation of Microsoft Excel.
Title: Social Media Mining in Learning Contexts
Bodong Chen, University of Minnesota
Description: Learning – of various kinds – takes place through mediated interaction on social media. This tutorial will engage participants in both theoretical considerations and analytical techniques pertinent to social media mining. This tutorial will provide hands-on experience with applying network analysis and text mining on a sample dataset.
Prerequisite Skills/Experience: Familiarity with popular social media platforms such as Twitter and Github is expected. Familiarity with R programming will be helpful, but is not required. Prior exposure to sociocognitive and sociocultural perspectives of learning will be helpful.
Advanced preparation:
- tutorial website…coming soon.
Title: Designing Dashboards to Impact Educational Practice
June Ahn, New York University
Description: We will discuss the practical and conceptual issues that may arise as we design dashboards for use “in the wild” with our educator partners. I will build from my experiences on an NSF-funded, research-practice partnership, where we are developing data tools and dashboards to support mathematics teachers’ changing pedagogical practices. Tutorial participants will also be invited to share features of their projects to derive design considerations. We’ll utilize various paper, prototyping activities to then ideate on how dashboard designs may relate to improvements in data interpretation, exploration, and ultimately improvements in practitioners’ daily practices (an ultimate hope for learning analytics work).
More information about the NSF project are available here.
Prerequisite Skills/Experience: None.
Advanced preparation:
- None.
Title: Text mining for learning content analysis
Jelena Jovanović, University of Belgrade
Description: In this tutorial, participants will learn about text mining methods and techniques that could be used to extract information and patterns from various kinds of text-based learning content, including both informal content such as posts exchanged in online communication channels and online social media platforms, and more formal content such as lecture notes or student essays. Participants will be introduced to the overall text mining process, and a particular focus will be put on (i) feature creation and selection, that is, transformation of unstructured textual content to a structured numerical format that can be fed to a pattern or information mining algorithm; (ii) different kinds of algorithms, including statistical, machine learning, and graph-based algorithms, that can be used for pattern mining and/or information extraction. A sample of practical, off-the-shelf tools for some sophisticated text analytics tasks, such as entity linking, will be introduced, as well.
Prerequisite Skills/Experience: Basic Machine learning (Data mining) knowledge will be helpful, but is not required. Likewise, some familiarity with R programming will be helpful, but not necessary.
Advanced preparation:
- Participants should bring their own computer (laptop) with R and RStudio installed, if they want to directly work on the practical examples that will be demonstrated during the tutorial.
Title: Education Data: Privacy, Policymaking, and Artificial Intelligence Ethics
Elana Zeide, Seton Hall University School of Law
Description: Learning analytics and educational data mining hold great promise. At the same time, they raise important concerns about security, privacy, and the broader consequences of big data-driven education. This tutorial tackles the regulatory framework governing student data, its neglect of learning analytics and educational data mining, and proactive approaches to privacy. It is less about conveying specific rules and more about relevant concerns and solutions. Traditional student privacy law focuses on ensuring that parents or schools approve disclosure of student information. They are designed, however, to apply to paper “education records,” not “student data.” As a result, they no longer provide meaningful oversight. The primary US federal student privacy statute does not even impose direct consequences for noncompliance or cover “learner” data collected directly from students. Newer privacy protections are uncoordinated, often prohibiting specific practices to disastrous effect or trying to limit “commercial” use. These also neglect the nuanced ethical issues that exist even when big data serves educational purposes. This tutorial will propose a proactive approach that goes beyond mere compliance and includes explicitly considering broader consequences and ethics, putting explicit review protocols in place, providing meaningful transparency, and ensuring algorithmic accountability.
Prerequisite Skills/Experience: None.
Advanced preparation: None
Title: Multimodal Learning Analytics (MMLA)
Marcelo Worsley, Northwestern University
Xavier Ochoa, Escuela Superior Politécnica del Litoral
Description: Learning does not only occur over Learning Management Systems or digital tools. It tends to happen in several face-to-face, hands-on, unbounded and analog learning settings such as classrooms and labs. Multimodal Learning Analytics (MMLA) emphasizes the analysis of natural rich modalities of communication during situated learning activities. This includes students’ speech, writing, and nonverbal interaction (e.g., movements, gestures, facial expressions, gaze, biometrics, etc.). A primary objective of multimodal learning analytics is to analyze coherent signal, activity, and lexical patterns to understand the learning process and provide feedback to its participants in order to improve the learning experience. This tutorial is posed as a gentle introduction to this new approach to Learning Analytics: its promises, its challenges, its tools and methodologies. To follow the same spirit of MMLA, this tutorial will include a hands-on learning experience analyzing different types of signals captured from real environments.
Prerequisite Skills/Experience: None.
Advanced preparation: None