Data analytics: one call for papers, free research papers, and a free paper on libraries

BayLAN conference, UC Berkeley

Submission deadline December 15th, 2017 (submit here)
Conference date: February 24, 2018
Conference fee: students for free, professionals: 15 $
Can't make it to LAK'18 in Sydney? Want to meet Bay Area LA practitioners and researchers on your way south? BayLAN is a local network of researchers and professionals in the field of learning analytics. The BayLAN conference is a regional event designed to  facilitate the exchange of information, case studies, ideas, and early stage research in the field of learning analytics broadly construed.

The Bay Area Learning Analytics Network (BayLAN), in co-operation with SoLAR, is hosting the third annual BayLAN conference on February 24, 2018 at the University of California, Berkeley. The BayLAN conference brings together thought leaders from both industry and academia. Presentations and discussions will focus on current research at the intersection of education, data science, and technology.

Registration is free for students and $15 for professionals. To register click here:
https://www.eventbrite.com/e/the-bay-area-learning-analytics-conference-2018-baylan-tickets-38191297198

BayLAN is currently accepting abstract submissions for the conference. Abstracts should report on research in the broad area of learning analytics.  Presentations may include technical work that applies data science or other quantitative methods to improve education, as well as interventions, methodologies, tools or technology that are intended to improve learning outcomes.

Suggested topics: 

• Theoretical topics: cognitive science models about education, data science methods applied to learning, novel theories about learning
• Lessons learned: After going through the learning analytics implementation process, share insights that have surfaced that affect the completion of the project
• Innovative new tools/techniques: Share newly developed tools or approaches to learning analytics that have been implemented at an institution.
• Application of standards: A project making use of data/analytics standards and illustrating the benefits of such an approach.
• Collaboration and sharing: How are groups of institutions/practitioners partnering to solve shared problems in the learning analytics space?

The deadline for submission is December 15, 2017. Submit here 

Free papers on Big data available until 4 December 2018 (the free access is available until then)

Special Issue: Big Data in Robotics from Liebert publishing
This issue was guest edited by Jeannette Bohg, Matei Ciocarlie, Javier Civera, and Lydia E. Kavraki

FREE ACCESS through December 4, 2017. Read Now:

Big Data on Robotics
Jeannette Bohg, Matei Ciocarlie, Javier Civera, and Lydia E. Kavraki  Read Now

Recent Data Sets on Object Manipulation: A Survey
Yongqiang Huang, Matteo Bianchi, Minas Liarokapis, and Yu Sun  Read Now

Leveraging Large-Scale Semantic Networks for Adaptive Robot Task Learning and Execution
Adrian Boteanu, Aaron St. Clair, Anahita Mohseni-Kabir, Carl Saldanha, and Sonia Chernova  Read Now

The KIT Motion-Language Dataset
Matthias Plappert, Christian Mandery, and Tamim Asfour  Read Now

DOOMED: Direct Online Optimization of Modeling Errors in Dynamics
Nathan Ratliff, Franziska Meier, Daniel Kappler, and Stefan Schaal  Read Now

Detection of Subtle Context-Dependent Model Inaccuracies in High-Dimensional Robot Domains
Juan Pablo Mendoza, Reid Simmons, and Manuela Veloso  Read Now

Free paper on libraries from paper to cloudbased

With the open access discussion going strong as ever (great reflective article here on the goal of open access from Stephen Downes, in reply to Willey's view on open access).
IGI global has just released a free paper (downloading free papers seems like a good strategy) on library history from paper to cloudbased, with examples.

The library paper can be downloaded here (you must register for IGI global)

The abstract of this paper: A Library is an organized collection of resources made accessible to a defined community for reference or borrowing. It provides physical or digital access to material, and may be a physical building, or a virtual space, or both. During the last decays, the Libraries had witnessed a continuous revolution and still do. This paper reviews the main milestone of such revolution starting from the classical up to the current Cloud-based era passing by the intermediate digital transformation period. It reviews the library types, services, problems and drive of changes from the classical form. The paper then tackles the transformation of the library to the digital form. It discusses the characteristics of the digital library, the web-based library, and the library 2.0 through their advantages and limitations. The paper finally focuses on the current Cloud-based ear, where most of the library cloud platforms, services management, innovative products and opened environments are addressed through their features, add values, pros and cons. The paper also provides a comparative study on such solutions coming up with opened research issues. Hereby, the paper provides a comprehensive overview on the development of the library till now. 

xAPI case studies available #xapi yeah!

In the last few months I have been trying to keep up with xAPI (in vain mostly due to phd writing, I admit, but still some blogposts and reflections did happen at my end). Luckily Aaron Silvers and Megan Bowe are keeping me updated by generously sending out all that is available for xAPI, including shared case studies.

This last batch of xAPI case studies offer short (average 15 min) videos covering xAPI in a variety of settings. So real stories on how people in EdTech are using Experience API in their context.

The videos were taped during the Orlando happening, and they include wonderful experts:

• Art Werkenthin on tracking PDF annotations with xAPI
• Andy Johnson as well as Mike Rustici on LRS (learning record store) 
• Ben Betts on enabling personal data ownership
• Chad Udell on implementing xAPI in mobile learning
• David Rogers on using xAPI in medicine
• Marty Rosenheck on driving innovation in teacher education
• Megan Bowe on designing analytics with xAPI
• ... 

A great set of learning videos there, and of interest to everyone with an urge to dig into xAPI.

Part4 #xAPI seminar Andrew Downes on #learningdesign #eln

Almost groggy here after information overload and process... hoping still to make sense a bit. Great seminar!

Part4 #xAPI seminar Andrew Downes on #learningdesign. Wonderful resource: http:// http://tincanapi.co.uk/

And if you want to get an xAPI UK project going, have a look here, at the end there is link to get in touch and start a conversation

http://epiclearninggroup.com/uk/tin-can-predictions-for-2014/

If you want to start designing xAPI statements, here is that start:

http://epiclearninggroup.com/uk/knowledge-base/how-to-design-tin-can-statements/

And a few pointers on how to start tracking xAPI and real world

http://epiclearninggroup.com/uk/files/2013/05/H2_tin_can_tracking.pdf

And an iPhone xAPI statements viewer (free) here:
https://itunes.apple.com/gb/app/experience-api-xapi-statement/id550133878?mt=8

Looking at how Tin Can design affects learning design

He is key UK lead for xAPI at Epic

A xAPI mindset

The xAPI looks at events, not necessarily status of events (latter typical of scorm)

Which results in different learning designs.

I	Did	This
A learner	Succeeded at	A work task
		Some elearning
		Their personal goal
Computer	completed	me

Think about

What different experiences do or could make up your blend of learning?

What needs to happen in experiences X to trigger a change in experience Y

What is a natural flow for your learners?

Challenges on informal xAPI stuff

Reliability of self-reporting

I did this => prove it!

Will learners report their learning?

Please complete this form => possibly NO (so what can we do to avoid this, or work around. What might be a benefit?) – e.g. training courses only made available to those who do send feedback.

Privacy concerns

We are tracking everything you do => Ummmmm (big brother issue, communicated well, open on it)

Interoperability

It’s not working, is it? => Have you tried turning it off and on again? (So communicating on this is crucial, both with learners, as with developers/experts)

Too much data

Joe moved this mouse 1 pixel => Which direction?

Correlation is not causation. 

#LAK11 analyse your learner/worker interactions with SNAPP to increase #elearning outputs, pedagogy and #performance

Many of us use learning management systems, and some of us even track students to support learning processes, but let's face it: tracking does not always result in clear visualization of the interactivity that is taken place amongst learners. Here is where SNAPP comes in.

There are other visual analyzing tools, but SNAPP is simple to use and easy to install, which is why I really like it.

SNAPP is developed by the University of Wollongong (who I think are coming up with a great deal of interesting educational innovations and thoughts). SNAPP allows you to transcript interactions in Moodle, Blackboard (and webCT, Desire2Learn into maps that immediately show who is a hyperconnected learner or participant, and who is more of a lurker, or even to pinpoint learners that at the core of the interactions.

SNAPP is a software tool that allows users to visualize the network of interactions resulting from discussion forum posts and replies. The network visualisations of forum interactions provide an opportunity for teachers to rapidly identify patterns of user behaviour – at any stage of course progression. SNAPP has been developed to extract all user interactions from various commercial and open source learning management systems (LMS) such as BlackBoard (including the former WebCT), and Moodle. SNAPP is compatible for both Mac and PC users and operates in Internet Explorer, Firefox and Safari.

By mapping interactions, you have a better overview of what people are doing and how they learn/perform. As such I see two fields of interest: education/training and knowledge management in corporations.

Education and training benefits
you can localize students or learners that are at risk (of not understanding, of having decreasingly less time...). The SNAPP-people have listed some great benefits of network diagrams:
* identify disconnected (at risk) students;
* identify key information brokers within your class;
* identify potentially high and low performing students so you can plan interventions before you even mark their work;
* indicate the extent to which a learning community is developing in your class;
* provide you with a “before and after” snapshot of what kinds of interactions happened before and after you intervened/changed your learning activity design (useful to see what effect your changes have had on student interactions and for demonstrating reflective teaching practice e.g. through a teaching portfolio)
* allow your students to benchmark their performance without the need for marking.

SNAPP also allows you to pinpoint lurkers (who might be a good target audience for research, e.g. did they get something out of the course, and why did they only lurk?
Additionally, you can filter out those students that might be good future facilitators. For students that are in the middle of the interactions, are clearly motivated and strong in reaching out and communicating.
Another educational benefit is for looking at peer-2-peer interactions. It might be that you do not get much feedback from a student, but after analyzing the interactions, you might see that that same student is a real peer-2-peer knowledge node, and as such has great course value.

Performance and knowledge management benefits
This SNAPP approach can also be used for knowledge management. Let's say you have a internal, central 'help' forum inside your company. And there is a person responsible for giving help on that forum. It might well be that after you screen the forum with SNAPP, you can see that someone else in your organization offers just as much help or support, just because they know the subject. As such you can give a bonus, or a different function to that active employee and stimulate knowledge transfer in your organization.

Interpretation of visual network maps
SNAPP does not simply provide the software, they also give an overview of how to interpret the visual data that you get from analyzing the network interactions.
So how does it work?
You download SNAPP and you get cracking with it!
I will take Moodle as an example:
Moodle

1. Visit a Forum in Moodle. Ensure that “Display replies in nested format” is selected.
2. Click on the SNAPP Bookmarklet (from where you placed it in the Bookmarks menu in Firefox or the Favorites menu in IE). You may need to enable the Menu bar in IE 7, to view the Favorites menu. The SNAPP interface will be inserted at the bottom of the page.
3. The SNAPP interface consists of five tabs: Visualization, Statistics, Export, Help and Credits. Within the Export tab, social graph data can be saved as GraphML or to the .vna format. The .vna format can be opened in NetDraw for additional analysis and visualization.

What will you see on your screen: the program will go through all the threads, one-by-one (you see the screen flip from one thread to another), and eventually a Java-icon will be displayed. You just wait and the graphic will appear, together with 5 tabs.

Netdraw - also a free tool developed by Steve Borgatti a nice bearded bowling professor in Kentucky, allows you to fine-tune the graphs you get with SNAPP.

The picture depicted here is a representation of a welcoming forum in Moodle.

#lak11 getting #learning #analytics on track demands interdisciplinary teams and connectedness

Grasping all the factors of learning analytics is at the core of LAK11 course. In the first week John Fritz took the mic of the Elluminate session (you can see the recording here) and got all the LAK11 participants introduced to learning analythics.

In order for learning analytics to take off, strategies must be in place that clearly recommend teacher or trainer interventions to pick-up those learners who are at risk of dropping out of courses or trainings.

Need for interdisciplinary teams
While listening to his presentation, I immediately got the idea that the mere fact of learning analytics is interdisciplinary. You cannot analyze learner metrics and hope to take educational action, if the educational intervention team is not connecting different fields together.

John mentioned in this presentation that the biggest question for learning analytics is: how do we move on, where do we go from here? John mentioned that most educational and training institutes are still at the analysis stage, but they do not move on to an intervention stage to get their learners on track. I think this is part of the fact that within institutions interdisciplinary teams are scarce. Learning analytics demands more than the data, it demands true understanding of learning theories, of sociology, of instructional design, of psychology, IT (if not more). So if an institute wants to cross the barrier of mere analyzing their students, they will have to build a framework, or at least see if framework builders are willing to work together in their institutes.

Is there such a thing as zone of proximal development for researcher teams?
I feel that if a new research that expands across many fields of interest has any chance of succeeding, it must connect experts from all of these fields. That is easier said then done, as the diversity of fields might put the experts so far away, that there is no common ground on which to stand and understand themselves. So, getting together a mix of disciplines is not enough. In a way all of these experts must be within their own zones of proximal development, their knowledge must be close enough to be able to link to the newly to construct knowledge in an understandable way.
If an interdisciplinary task-force to work on learning analytics and its strategies needs to be put together, I think the profiles of this team must be overlapping just enough to allow cross-understanding.

Self-regulated learning
John also talked about giving the learners the tools to adjust their learning. This is a huge meta-learning challenge, for it demands that learners understand many of the factors that influence their learning, which seems difficult to me. Nevertheless, what John says about this is interesting. It might help some students, but at the same time I feel that if even the researchers do not come up with strong solutions, how can we put most of the burden on the learners? (but I do agree with John that learners must also take responsibility for their learning). John mentions a great resource for research on self-regulated learning: Zimmerman.

#LAK11 what about #privacy when using #data analytics to create #knowledge?

Learning Analytics: EDUCAUSE

View more presentations from gsiemens.

LAK11, the course on learning analytics and knowledge has started and immediately discussions touch a sensitive and complex issue: privacy. While talking about educational data mining, the issue of privacy was raised by many participants.

First a quick note on educational data mining: data mining, also called Knowledge Discovery in Databases (KDD), is the field of discovering novel and potentially useful information from large amounts of data. And with many social media apps gathering massive amounts of data, this field is of growing interests to companies as well as researchers. Educational data mining has a different aim, not to make money, but to map/increase/analyse educational knowledge creation and awareness. What makes educational data mining so interesting, is that it is increasingly using models drawn from the psychometrics literature in education.

The idea behind EDM is a great one: modeling student individual differences in these areas enables software to respond to those individual differences, significantly improving student learning for example. Another key area of application of EDM methods has been in studying pedagogical support (both in learning software, and in other domains, such as collaborative learning behaviors), towards discovering which types of pedagogical support are most effective, either overall or for different groups of students or in different situations. The only thing is, how do you cope with the privacy issues?

Privacy can be good or bad
Gathered data – just like anything else – can be used for the good or the bad. An early form of data gathering was getting all the Jews registered in Germany. Once that was done – for purely administrative purposes at first – it was a small step to gather the Jews themselves for … processing. There is also a recent initiative: in Nepal the third sex has been added to Nepal's next demographic surveys. This is a wonderful thing (I for one do not apply for many 'either/or' categories), but it makes abuse possible.

To me privacy is more relevant to minorities and vulnerable groups, then any other groups in society. If facebook sells gay related data to the Ugandan government, I would not like to be a gay person in Uganda, for you risk being seriously harassed if not killed (at this point in time). No matter when, privacy should be an option that is respected, disconnected from your own online identity, certainly you – as a user – should be able to use a ‘non-traceable’ account for certain reasons?

For instance by using disconnect(ere).

Out of the crowdsourcing cloud, out of the real world?
A way forward to keep your own data private is via: disconnect. But then if all of the people from a specific group use disconnect, their data will no longer be in the crowd, and as such 'non-existent', which makes vulnerable groups even more invisible and less catered to.

perspective of a LAK11 participant
Sarah Haavind, a participant of LAK11, mentions two interesting points on privacy as well:
First: the FB and twitter users no longer clutch to privacy (is this really true and has anyone researched this or the reasons behind it? I do mind and I adjust my privacy settings)
Secondly she adds to the privacy-commercial link: “What if the data-mining marketers know what I like? Isn't it a positive to be bombarded with ads for products I'm interested in rather than a random assortment, given their presence in our lives?”.
She has a point there, but… how can small, very targeted businesses come up through the data analytics that is available?
If I had to choose, I would also like smaller companies to be able to take part in this ad-data-world. I am not interested in starbucks, I am interested in the local coffeeshop burning its own coffee beans, just because I like a variety of tastes. If I go to Ethiopia, I want to listen to local contemporary music, I am not interested in the hashed music that can be found everywhere. So to that effect, I would like it better if I could choose localized data from my profile of ‘local business’ then the big mainstream stuff. But do I need to open up my private life to get access to such time-saving and potentially interesting stuff?

Having said this, gathering and mining educational data might help us a long way to get quick access to some sustainable outcomes (learner preferences, learner critical thinking skills…). But of course, like all data mining, the proof is in the algorithm. If the algorithm is based on a wrong hypothesis, will the results then still be useful? And if data is sold to the highest bidder, then the ethics of that highest bidder better be in sync with my interpretation of ethics.