Flexible, Free and Open Data-Driven Learning for the Masses (MOOCs)
Download as PPTX, PDF1 like826 views
The document discusses the Flax language project, which aims to enhance language acquisition through data-driven learning by providing flexible and open-access linguistic resources. It details the development of tools and interfaces for learners and teachers, enabling them to create customized libraries and utilize real-world linguistic data. Further, research findings indicate the effectiveness of Flax in improving specialized terminology acquisition among students engaged in MOOCs compared to traditional learning methods.
Flexible, Free and Open Data-Driven Learning for the Masses (MOOCs)
- 1. Flexible, Free and Open Data- Driven Learning for the Masses Alannah Fitzgerald http://maxpixel.freegreatpicture.com/photo-1742679
- 2. MINING & LINKING OPEN CONTENT FOR DATA DRIVEN LEARNING FLAX Language Digital Library Project, University of Waikato, NZ
- 3. Data-Driven Learning The metaphor that Johns evoked was one where language is treated as empirical data and “every student is a Sherlock Holmes”, investigating the uses of linguistic data directly to assist with language acquisition (Johns, 2002, p. 108).
- 4. flax.nzdl.org Powerful yet simple interfaces for Data-Driven Learning
- 5. The eBook of FLAX “FLAX (Flexible Language Acquisition) is both a vision and a tool that you can use for language learning. The Web contains innumerable language activities, quizzes, and games, but they are fixed: the activities are cast in stone and the material is chosen by others. Our vision is to put the control back where it belongs, in the hands of teachers and learners.”
- 6. WHO ARE WE IN THIS FLAX RESEARCH & DEVELOPMENT COLLABORATION?
- 7. FLAX Language at Waikato University http://flax.nzdl.org FLAX image by permission of non-commercial reuse by Jane Galloway
- 8. FLAX Language Project at the Greenstone Digital Library Lab, Waikato University NZ Professor Ian Witten FLAX Project Lead Dr Shaoqun Wu FLAX Project Lead Researcher & Developer
- 9. Research on Open FLAX Collections http://oerresearchhub.org/ Alannah Fitzgerald Open Fellow with OERRH FLAX Language & Open Education Researcher
- 10. OPEN SOURCE LANGUAGE TOOLS DEVELOPMENT
- 13. Google-esque Interface Designs Designed for the non-expert corpus user, namely: learners, teachers, subject academics, instructional designers and language resource developers. http://flax.nzdl.org/greenstone3/flax?a=fp&sa=collAbout&c=collocations&if=flax
- 14. Link to the Collocation Learning System with the Wikipedia Corpus in FLAX (Wu, Li, Witten & Yu, 2016) http://flax.nzdl.org/greenstone3/flax?a=g&rt=r&sa=CollocationQuery&s=CollocationQuery&s1.title=&c=collocations&s1.threshold= 0.5&s1.startNum=0&s1.perPage=20&s1.sampleNum=10&s1.type=&s1.wordType=&s1.colloType=&s1.query=role&s1.dbName=Wikip edia
- 15. Introducing the Wikipedia Miner Toolkit (Milne & Witten, 2013)
- 16. Building Interactivity into FLAX Language Collections
- 18. FLAX TEAM Apps for Android via GooglePlay http://commons.wikimedia.org/wiki/File:Android_robot_skateboardin g.svg / http://commons.wikimedia.org/wiki/File:Google_Play_Store.svg
- 19. FLAX Team on Google Play https://play.google.com/store/apps/developer?id=FLAX+TEAM&hl=en
- 20. FLAX Across Platforms • FLAX Website flax.nzdl.org for hosting open online language collections • Building directly onto the Web with OER • FLAX multilingual open-source software for download • Set up your own FLAX server online or; • Build collections offline for use on your PC • FLAX Android app for download • Interact with game-based FLAX collections while on the go • FLAX for MOODLE plug-in for download • FLAX for MOOC Platforms? • FLAX in conjunction with translation technologies?
- 21. DOMAIN-SPECIFIC OPEN LANGUAGE COLLECTIONS BUILDING
- 22. The eBook of FLAX “FLAX enables teachers to build bespoke libraries very easily. It is built upon powerful digital library technology, and provides access to vast linguistic resources containing countless examples of actual, authentic, usage in contemporary text. But teachers can also build collections using their own material, focusing on language learning in a particular domain (e.g., business, law) or motivating students by using text from a particular context (e.g., country or region, common interests).”
- 23. RESEARCH INTO MOOC LINGUISTIC SUPPORT
- 24. FLAX Academic English Collections http://flax.nzdl.org/greenstone3/flax?a=fp&sa=library
- 25. MOOC Research Participants • CopyrightX (Harvard University – formerly an edX MOOC, now a networked course) • ContractsX (Harvard University with edX) • English Common Law (University of London with Coursera)
- 26. Role and courses taken by respondents 2014- 2016
- 27. Age bands of respondents
- 28. Educational background of respondents
- 29. Languages spoken by MOOC learners • English (95.71%), followed by increasingly smaller numbers of participants who identified as being able to speak fluent: • Spanish (16.56%), French (12.88%), German (8.59%), Italian (7.98%), Catalan (3.0%), Chinese, Finnish, Gujarati, Swahili (1.84%), French Creole, Hindi, Japanese, Korean, Luo, Norwegian, Portuguese, Russian, Serbian (1.23%), Arabic, Georgian, Slovak, Thai, Turkish, Ukrainian, Urdu, Vietnamese (0.61%).
- 30. “When you want to find out how to express something in English what resource(s) do you use? You can select more than one.” Language Resources Informal learners (N=163) CopyrightX teachers (N=11) Paper-based dictionaries 18.40% 18.18% Online dictionaries 76.07% 100.00% Online reference resources (e.g. Wikipedia) 52.15% 81.82% Search engines (e.g. Google, using inverted commas "" and asterisks * to search for keywords/phrases for language use) 57.67% 100.00% Corpora / searchable web- based language collections (e.g. FLAX, WebCorp) 7.98% 0.00% Grammar books 11.66% 9.09% Language course books 1.84% 9.09% Ask someone 31.90% 27.27% Need nothing 2.45% 0.00%
- 31. Keyword search for creative in CopyrightX collection
- 32. Wikify function in the CopyrightX MOOC collection
- 33. Preview of some of the top 100 collocations in the CopyrightX collection displaying summary judgment
- 34. Learner motivations for using FLAX and other language support resources
- 35. FLAX user experience for learners
- 36. Learner feedback on the searchability of the FLAX ContractsX MOOC collection
- 37. Learner feedback on the collocations and Cherry Basket features in the English Common Law MOOC collection
- 38. Negative features of FLAX according to respondents
- 39. Positive features of FLAX according to respondents
- 40. Extra comments on FLAX from respondents
- 41. RESEARCH INTO THE REUSE OF MOOC LINGUISTIC CONTENT
- 42. Fitzgerald, A., Marin. M.J., Wu, S. & Witten, I.H. (2017). Evaluating the Efficacy of the Digital Commons for Scaling Data-Driven Learning. In M. Carrier, R. M. Damerow, & K. M. Bailey (Eds.), Digital Language Learning and Teaching: Research, Theory, and Practice (pp. 38 – 51). New York, NY: Routledge & TIRF.
- 43. The Digital Commons Typically, the digital commons involves the creation and distribution of informational resources and technologies that have been designed to stay in the digital commons using various open licenses, including the GNU Public License and the Creative Commons suite of licenses (Wikipedia, 2016). One of the most widely used informational resources developed by and for the digital commons is Wikipedia.
- 44. Data Collection Procedure • 52 students in the fourth year of the Translation Degree program at the University of Murcia (Spain) were selected as informants. • All the students’ linguistic competence level complied with the Common European Framework of Reference for Languages requirements for the B2 level.
- 45. Experimental & Control Groups • The experimental group (16 informants organized into four sub-groups) were requested to only consult the FLAX English Common Law MOOC collection as the single source of information to draft their essays. • The remaining 36 students (divided into nine different sub-groups) would act as the control group, following the traditional method for the design and drafting of essays before this experiment was carried out, that is, using any information source available.
- 46. Term Average in each corpus FLAX Corpus Non-FLAX Corpus Terms Identified by Themostat (A) (Drouin, 2003) 226 385 Corpus Size After Reduction 16,939 16,264 Number of Topics (B) 4 9 Term Average (A/B) 56.5 42.77 Standardized type/token ratio 35.3 38.63
- 47. Findings from Reuse Study • According to the data, the members of the experimental group appear to have acquired the specialized terminology of the area better than those in the control group, as attested by the higher term average obtained by the texts in the FLAX-based corpus (56.5) as opposed to the non-FLAX-based text collection, at 13.73 points below • However, the standardized type/token ratio assigned to each set of texts, which is often indicative of the richness of the vocabulary (the higher, the richer), is lower for the FLAX-based texts, standing at 3 points below the texts written by the control group
- 48. References • Biber, D., Conrad, S., & Cortes, V. (2004). If you look at . . .: lexical bundles in university teaching and textbooks. Applied Linguistics, 25, 371–405. Biber, D. (2006). University Language, A corpus-based study of spoken and written registers. John Benjamins, Amsterdam. • Biber, D., Barbieri F. (2007). Lexical bundles in university spoken and written registers. English for Specific Purpose, 26, 263–286. • Fitzgerald, A., Marin. M.J., Wu, S. & Witten, I.H. (2017). Evaluating the Efficacy of the Digital Commons for Scaling Data-Driven Learning. In M. Carrier, R. M. Damerow, & K. M. Bailey (Eds.), Digital Language Learning and Teaching: Research, Theory, and Practice (pp. 38 – 51). New York, NY: Routledge & TIRF. • Johns, T. (2002). Data-driven learning: the perpetual challenge. In B. Kettemann & G. Marko (Eds.), Teaching and Learning by Doing Corpus Analysis. Proceedings of the Fourth International Conference on Teaching and Language Corpora, Graz 19-24 July, 2000, (pp. 107-117). Amsterdam: Rodopi. • Milne, D. & Witten, I.H. (2013). An open-source toolkit for mining Wikipedia. Artificial Intelligence, 194, 222-239. • Wu, S., Li, L., Witten, I.H., Yu, A. (2016). Constructing a Collocation Learning System from the Wikipedia Corpus. International Journal of Computer-Assisted Language Learning and Teaching (IJCALLT), 6, issue 3, pp. 18-35
- 49. Thank You Special Thanks: Ruth Crymes TESOL Fellowship for Graduate Study The International Research Foundation (TIRF) for English Language Education FLAX Language Project & Software Downloads: http://flax.nzdl.org/ FLAX Language Project Research: https://www.researchgate.net/project/FLAX-Flexible- Language-Acquisition-flaxnzdlorg The How-to eBook of FLAX: http://flax- doc.nzdl.org/BOOK_OF_FLAX/BookofFLAX%20fullsize%20with%20links.pdf FLAX Game-based Apps for Android via Google Play Store (free): https://play.google.com/store/apps/developer?id=FLAX%20TEAM&hl=en Ian Witten (FLAX Project Lead): ihw@cs.waikato.ac.nz Shaoqun Wu (FLAX Research and Development): shaoqun@waikato.ac.nz Alannah Fitzgerald (FLAX Open Language Research): a_fitzg@education.concordia.ca TOETOE Technology for Open English Blog: www.alannahfitzgerald.org Slideshare: http://www.slideshare.net/AlannahOpenEd/ Twitter: @AlannahFitz
Editor's Notes
- #12: Teachers can construct collections of different types: for different purposes and for different types of students. The collections can be: item specific domain and/or topic specific graded for levels of difficulty representative of a particular source or of a particular genre subsets of a larger corpus e.g. BAWE. Potentially students can also construct collections (see Charles, 2012)