Learning Analytics in Formal and Informal Learning

The table below is a synopsis of my recent paper "Community Analytics - Challenges and Oportunties". In the paper, I try to motivate that current learning analytics in formal learning approaches are limited in applications of recorded data because they are already biased by the learning analytics environment. In informal learning, in particular in community-regulated learning, because of the eco-system like community-regulated learning environments and the dynamic nature of learning activities, learning goals and learning communities more data mining and recommender technologies can be applied. So, informal learning is more interesting for computer scientists than formal learning :-)
So, what are highlights of new opportunities for learning analytics in informal learning. First, since we have no formal teacher or instructor role and expertise in a community of learners is a highly dynamic concept, we need to identify experts in a learning community. Experts may be good candidates for mentors, advisers or teachers in an informal learning community. Newcomers (learners) in a community may appreciate an indication about the expertise of persons in the community. For sure, these algorithms must be robust enough to survive attacks of trolls and other misbehaving persons. Second, if we have the concept of a learning community, it may be an important questions for learners which community one should join. For example, if I am an intermediate learner of Spanish, I would not like a learner community where the majority are beginners (unless I want to become an immediate expert :-) But, as a learner I would not learn enough, since I know too much already. I would also not like to join a very advanced learner group, because I do not know enough to learn effectively. In formal learning this is often addressed as a competence level of a course. In informal learning we need to extent current expert recommender systems to community recommender systems. If we have communities,  we may want to join forces with other learners to exchange knowledge about learning tools, goals etc. But how to find such communities in the Web? Here, (overlapping) community detection algorithms may help to find other communities with similar profiles.With methods of social network analysis, communities can be found and knowledge can be co-created and thus scaled in communities of practice.

	Formal Learning	Learning Analytics	Community-regulated Learning	Community Learning Analytics
Environment	LMS	EDM/VA	CIS/ROLE	DM/VA/SNA/ Role Mining
Tools	Fixed	LMS Specific	Ecosystem	Tool Recommender
Activities	Fixed	Content Recommender	Dynamic	Content Recommender/ Expert Recommender
Goals	Fixed	Progress	Dynamic	Progress / Goal Mining / Refinement
Communities	Fixed	Not applicable	Dynamic	(Overlapping) Community Detection
Use Cases	Courses	Learning Paths	Peer Production / Scaffolding	Semantic Networks of Learners / Annotations