AFEL · Project

Smart Analytics That Track How Employees Actually Learn Outside Formal Training

digitalTestedTRL 5Thin data (2/5)

web.archive.org

Most of what people learn at work doesn't happen in classrooms or e-learning courses — it happens by reading articles, watching videos, asking questions on forums, and following experts on social media. The problem is, companies have zero visibility into this "invisible learning." AFEL built tools that collect and analyze data from social platforms to reveal what people are actually learning, how they're learning it, and where the gaps are. Think of it like Google Analytics, but for informal learning happening across the web.

By the numbers

EUR 2,581,940

EU funding for development

consortium partners across 5 countries

total project deliverables produced

demo/prototype deliverables

releases of large-scale learning analytics dataset

The business problem

What needed solving

Companies invest heavily in formal training programs but have no way to measure or improve the informal learning that accounts for most actual skill development. Employees learn constantly through social media, professional communities, and online resources — but this learning is invisible to L&D teams, making it impossible to identify skill gaps, track development, or optimize learning support.

The solution

What was built

AFEL built a complete data extraction and management infrastructure that pulls learning activity data from social platforms (including GNOSS, Facebook, LinkedIn), plus ontological models to classify learning activities, integrated feature extraction using NLP and clustering, large-scale learning analytics datasets (3 releases), and reusable analytics components tested across multiple platforms. Total output: 34 deliverables including 8 working demos.

Audience

Who needs this

Corporate L&D departments at large enterprises wanting to measure informal learning ROIEdTech platforms looking to extend analytics beyond their own course contentHR technology companies building skills mapping and talent intelligence productsProfessional community platform operators wanting to demonstrate learning value to membersConsulting firms specializing in workforce development and digital transformation

Business applications

Who can put this to work

Corporate Learning & Development

enterprise

Target: Enterprise L&D departments or corporate training providers

If you are a corporate training provider struggling to measure learning that happens outside your LMS — this project developed data extraction tools and analytics models that capture informal learning activities from social platforms like LinkedIn and professional communities. The system was tested on a commercially available social platform (GNOSS) with 34 deliverables covering everything from data collection to visual analytics dashboards.

EdTech & Online Education

any

Target: Online learning platform operators

If you are an EdTech company that wants to understand how learners engage beyond your course catalog — AFEL created ontological models of learning activities and large-scale datasets for social learning analytics. The tools extract learning signals from social environments using natural language processing and clustering techniques, tested across multiple online social platforms by a 6-partner consortium across 5 countries.

HR Technology & Talent Management

mid-size

Target: HR tech companies building skills assessment tools

If you are an HR tech company trying to map employee skills development from real-world activity rather than self-reported data — AFEL built integrated feature extraction tools that identify learning activities from social media profiles, community participation, and online collaboration. The data management infrastructure uses linked data standards, making it interoperable with existing HR systems.

Frequently asked

Quick answers

What would it cost to implement these analytics tools?

The project received EUR 2,581,940 in EU funding across 6 partners over 3 years. The tools were built as reusable components tested on the GNOSS platform. Implementation costs would depend on your platform size and data sources, but the open-source nature of EU-funded research typically lowers licensing barriers.

Can this scale to thousands of users across a large organization?

AFEL specifically built a 'large-scale dataset for social learning analytics' delivered in three releases, each expanding and enriching the data. The data management infrastructure includes a triple store and data endpoint designed for volume. The system was validated on a live commercial platform (GNOSS), suggesting it handles real-world scale.

What about IP and licensing for these tools?

As an EU-funded Research and Innovation Action (RIA), results are typically available under open or favorable licensing terms. The consortium included 2 SMEs and 1 industry partner who had commercial interests. Contact the coordinator at University of Galway for specific licensing terms on individual components.

How does it integrate with our existing learning management system?

The project built its core data model using linked data standards and ontological models specifically designed for interoperability. The tools were extracted from one platform (GNOSS) and tested on other social platforms as 'reusable components,' which demonstrates portability across different systems.

What data sources can it actually pull learning signals from?

Based on the deliverables, the system built extractors for GNOSS platform logs, Facebook activity streams, and LinkedIn skills and education APIs. The architecture is designed to add new data sources through its modular data extraction infrastructure.

Is this proven technology or still experimental?

The project ran from 2015 to 2018 and delivered 34 deliverables including working prototypes tested on live platforms. The integrated feature extraction system uses established techniques like NLP and clustering. However, the project ended in 2018, so some API integrations may need updating for current platform versions.

Consortium

Who built it

The AFEL consortium brought together 6 partners from 5 countries (Austria, Germany, Spain, Ireland, UK), with 3 universities providing research depth and 2 research organizations adding analytical capability. The industry side is lean at just 1 industry partner (17% ratio), with 2 SMEs in the mix. The coordinator is University of Galway in Ireland. For a business buyer, the relatively low industry participation means these tools were primarily validated in academic and research settings rather than corporate environments — useful technology, but expect some adaptation work to fit enterprise workflows.

UNIVERSITY OF GALWAYCoordinator · IE
THE OPEN UNIVERSITYparticipant · UK
KNOW CENTER RESEARCH GMBHparticipant · AT
GOTTFRIED WILHELM LEIBNIZ UNIVERSITAET HANNOVERparticipant · DE

How to reach the team

University of Galway, Ireland — reach out to the computer science or data science department that led this project

Order a report on this consortium See UNIVERSITY OF GALWAY profile →

Next steps

Talk to the team behind this work.

Want to explore how AFEL's informal learning analytics could work for your organization? SciTransfer can connect you directly with the research team and help assess fit for your use case.

Order a report on this topic More in Digital & ICT