"1. INTRODUCTION. Organizations, when thinking of teacher professional development, often rely on refreshment courses given by experts, in-service training, or personalised learning trajectories. These formal training opportunities provided for professionals represent just the tip of the iceberg when imagining all learning that takes place triggered by the challenges professionals face in their daily practice. The more spontaneous and informal ways of learning are largely overlooked in organisations and the effects of it remain therefore implicit. At the same time there is a large body of literature that convincingly shows that these forms of spontaneous work related learning are important drivers for ongoing professional development [2], [3], [4]. However powerful informal learning may be, there is a problem when it comes to making it a real asset within organizations: Informal learning activities are mostly invisible to others and sometimes the learners themselves might not even be aware of the learning that occurs. As a consequence informal learning in organizations goes undetected, remains off the radar of HR departments and is therefore hard to asses, manage and value [1]. This problem poses an interesting challenge for the field of Learning Analytics, namely finding ways to capture and analyze traces of (social) informal learning in every day life and work networks. Learning Analytics is about harvesting and analyzing information about learners, with a main focus on learners’ behavior in virtual environments, focusing on the collection of information about learners behavior in formal learning activities, like completing assignments and taking exams and in more informal settings like posts on discussion forums and online social interactions [5]. Extracting and evaluating patterns of participation and interaction through examination of online conversations is a major component in this area of work. To get a full insight of the informal learning activities of workplace learners, Learning Analytics needs to go much further then this [5], but we feel that this field holds a lot of promise in doing so. In this paper we will contribute to this development by discussing methodologies that Learning Analytics can draw upon to make informal learning more explicit and accessible to analyse and to share amongst professionals and other organizations. Our current research is aimed at the design of an awareness tool that will make explicit how professionals within and between organizations (and beyond) create network connections to populate the social space in which learning occurs. To do so we created a holistic methodology to capture and analyse informal learning activities in the workplace. In our research practice our target group are teaching professionals working in school organizations. The informal learning activities mostly take place face-to-face and in the schools themselves [6]. In the future we want to translate this methodology again to use it in the virtual world and replace the data provided by our target group with data mining technologies to automatically build up our dataset. Here we represent our methodology of building our Tool, and we would like to invite experts in the field to debate how we can transfer this methodology to the virtual world again and complement it with possible data mining solutions. 2. THEORETICAL BACKGROUND. The Network Awareness Tool is founded on theories Networked Learning, Social Capital, Communities of Practice and Social Network Theory and Social Network Analysis. On the one hand these theoretical orientations provide a rationale for why such a tool is important when raising awareness about informal networked learning, on the other hand it provides crucial design elements such a tool should include to make it work. 2.1 Networked Learning Theory. In schools innovative teacher professional development involves opportunities for teachers to share their expertise, learn from peers, and collaborate on real-world projects [7]. This approach to learning embraces the participation metaphor [8], [9] where learning is seen as situated, embedded and maintained in the daily culture of (shared) practices and professional standards. A network in this sense can be regarded as a web of social relationships among teachers that reflects the flow of resources among them. People use their networks as a social infrastructure to gain access to what it is they are looking for. Key aspect in this paper is to use these network structures to study a relational approach to learning [10], where the emphasis is on the interaction between people. To know how people learn and create knowledge through these networks is very important, especially in the increasingly connected world of today. Understanding these networked processes helps to indicate areas of knowledge creation and problem solving and how this (new) expertise is being shared, but it also helps to identify networks of creativity and innovation within organizations. To study informal learning activities from this social network perspective we can build on Networked Learning Theory. Networked Learning Theory is an emerging perspective that tries to understand learning by asking the question how people develop and maintain a ‘web’ of social relations used for their own and reciprocal learning and professional development. Networked Learning is a form of informal learning situated in practice, where people rely strongly on their social contacts for assistance and development [11],[12]. Recent research has provided evidence linking Networked Learning to an array of positive outcomes like student performance and school improvement [13], [14], [15], [16]. Networked learning theory is useful for our analysis, moreover because it is closely linked to and uses methodologies of Social Network Theory. 2.2 Social Network Theory and Social Network Analysis. According to Moreno [17] Sociometric tests, the forerunner of Social Network Theory, show “in a dramatic and precise fashion that every group has beneath its superficial, tangible, visible, readable structure an underlying intangible, invisible, unofficial structure, but one which is more alive, real and dynamic than the other.” (p. 268). To investigate the dynamics of informal learning it is exactly this invisible and informal structure we want to bring to light. Social Network Theory asserts that the constitution of a network may influence the accessibility of information and resources and that the social structure may offer potential for the exchange of resources [18], [19], [20], [21], [22], [23], [24], [25]. Understanding the network structure can reveal important evidence on the information flow and shared knowledge within an organisation [26]. Teams with the same skill composition can act differently depending on the structure of relations within the team and similarly an individual can act differently depending on their position in a network. Therefore the teachers’ social environment will be looked at for explanations, rather than individual characteristics like in traditional social research [27]. Social Network Theory tries to explain both the antecedents and the consequences of social networks. Following Social Network Theory we can investigate for example if teachers with a central role in a network learn more from their colleagues (consequence) or investigate if teachers who learn a lot, get a central position in a network (antecedent). The structural dimension of a network can be investigated by using Social Network Analysis. According Social Network Analysis a network consists of nodes and ties. Nodes are the individual actors within a network and ties are the relationships between the actors. The impact of the structure of social networks can be studied on three levels: first the positions people have in a network (individual dimension), the relational level (ties dimension) and finally the overall network structure (network dimension). Frequently used network concepts in social science that Social Network Analysis draws upon to are: network brokers, network isolates, gatekeepers, actor centrality and structural holes [21]. 2.3 Social Capital. While Social Network Theory helps us to highlight the structural dimension informal networks, we use Social Capital Theory to frame Social Network studies from a ‘content’ perspective. Networks are always about something. Social Capital Theory provides a lens to look more closely at the relational resources embedded in social ties and how actors interact to gain access to these resources [24]. The first systematic analysis of social capital was produced by Bourdieu [28], who defined the concept as the aggregate of the actual or potential resources existing within the relationships of a durable network. According to Lin [29] the common denominator of all major Social Capital Theories is summarized as ‘’The resources embedded in social relations and social structure which can be mobilised when an actor wishes to increase the likelihood of success in purposive action’ (p. 24). This differentiates Social Capital from Human Capital which refers to the stock of competences, knowledge and personality attributes of the individual actors. Social Capital Theory asserts that a node’s position in a network determines in part the opportunities and constraints of the node, and in this way affects the nodes outcomes. 2.4 Communities of Practice: The social learning dimension from an individual and collective perspective While the learning network perspective deals mostly with the personal aspirations, attitudes and strategies used for of learning in networks the collective advancement of knowledge and the development of shared identities comes together in the community aspect of social learning, which we base on the well known concept of communities of practice [30], referring to the development of a shared identity within a network of people and the collective development of a particular domain. A shared identity represents a collective intention, mostly related to a certain practice. Theories argue that networks, to be fruitful and active, need a shared framework of values and norms. Learning in communities is a process where both individual and collective learning goals and agenda’s are carefully and constantly being negotiated, around a topic or domain that is of interest to each participant. In this way communities enable the learners to develop a space for a shared activity in which their learning is situated. Here they connect ideas, share problems and insights in a constructive way, and connect with concepts with which they are already familiar, using new knowledge that is collaboratively constructed through their dialogues and social interactions. If professionals engage in networked learning from personal frameworks that are too different, they will be unable to understand the norms of each other. As a result, trust and reciprocity in the network are too low, and learning does not take place.’’ 2.5 Individual (demographic) characteristics influencing professionals’ informal learning Although we mainly focus on network characteristics, the individual demographics are also important indicators and need to be taken into account. According to a study conducted by Moolenaar [31] findings indicate that differences relationships were associated with differences in gender, grade level, working hours, formal position, and experience. Age and years of experience can also have an impact on teachers’ professional development. Senior employees tend to take less initiatives in their professional development [32]. Hence to investigate the informal social activities within an organisation we need to investigate several dimensions: the actors and their attributes, the relations or ties between the actors, the content and resources within these ties and the overall network structure. Highlighting these dimensions of dynamic relationships helps us to understand how professionals engage in informal learning relationships and the value it produces[1]. 3. BUILDING A TOOL TO ANALYSE INFORMAL LEARNING. To gather and analyse the required data we needed to built a tool: The Network Awareness Tool [33]; This Tool serves different goals. First of all we built it to collect data from our target group, focusing on informal learning happening in the workplace. This differentiates the tool from a general learning analytics approach, because we do not use already existing data in an existing database. We would like to refer to our article submitted for Computers in Human Behavior, that focuses on the strategies describing how to collect the data. Here we will focus on the second goal of the tool: How will we analyse the data: The tool will be used to gain insight into the dimensions of informal networks in organizations (see figure 1). Third, the tool generates instant feedback to our target group to get insight into their own informal learning activities. The second goal (analysis) is be relevant for this conference, because we think this approach could help learning analytics to translate data mining technologies to a framed and theoretically based research methodology. 3.1 How the Network Awareness Tool works. In short the tool works as follows: Users can register and create a profile page, with a focus on their work and field of expertise. Within the learning networks page users can create a learning network based on a theme or subject users are exploring. Then users can add persons they interact with on this theme to the learning network. Users can also explore existing learning networks within their organisation and link themselves to a learning network they are active in within their organisation (see figure 2). The Network Awareness Tool can be used on different mobile devices like smart phones, android tabs and i-pads. Next to the mobile devices, the Network Awareness Tool can also be used on a desktop computer using a standard web browser. The data will be gathered through the REST protocol and gathered in a central MySQL database. 3.2 What data we collect. First of all we have the actors within the informal learning network so we want to collect the individual (demographic) characteristics. To gain in depth information about the individual participants we included a personal profile page for all the actors to fill in, including gender, age, the organisation(s) they work, their profession and role in the organisation and their specific areas of expertise. This profile can be updated at all times. Extra features can be added, if specific research data is required. After creating a profile, participants can create a network on a certain topic or join an existing network (by existing we mean here, already set-up by another user on our Network Awareness Tool). Within the network they indicate the people they interact with. We also gather data on the quality or nature of the ties. For each person that the participant indicates as a learning relation in other words a ‘’learning resource’’ there is the possibility to indicate the frequency of their interactions, and the quality of the learning interaction with that person. We add this feature in the tool by means of a rating system. Actors can rate the frequency of their contacts (1 = daily, 2 = weekly, 3 = monthly, 4 = less than monthly) using stars and rate the quality for learning: What is the value of this tie in relation to the topic of this learning network, using stars as well (ranging from a low quality (1 star) to high quality 5 stars). To gather data about the content of the ties we added a box where the participants can fill in meta-tags about the issues they are talking about and share information on. People can describe the content of their interactions, by filling in keywords in the tool. 3.3 How we analyse the data: Social Network Analysis and Semantic Analysis To analyse the data we use the methodology based on the theoretical theories described in the theoretical background. We gather data on the components visualised in fig. 1. Through Social Network Analysis we can visualise the structural dimension of the learning networks. We used UCINET [27] to conduct Social Network Analysis. But we are building a feature to automatically visualize the data gathered in our database. Network analysis aims at finding out who is talking to whom with respect to a particular theme. This step visualizes existing informal networks in which people talk about the theme in question and shows the extent to which they are (or are not) connected throughout the entire organization. We analyze data on the three dimensions of the network: Individual level, Tie level and Network level. First we look at the ego-perspective. We start with calculating the out-degree centrality (number of ties that an actor directs to others) which indicates the extent to which an individual interacts with other members in the network [21]. The user profile with different attributes about the participant (like area’s of expertise) gives the opportunity to create informed network visualizations allowing the organization to see how professionals with a specific expertise are connected within the network, or how they are accessible by other colleagues f.e. Within organisations it is often interesting to see the gatekeepers between different locations or between different departments. This could be interesting to track the information flow within an organisation. To get more insight into the Tie level, we combine the data about frequency and quality to the social network analysis. In this respect we can investigate the role of strong and weak ties in a learning network. Combining data on the frequency and the quality can of course be very valuable. This data is also needed to draw the right conclusions about the learning outcomes of a learning network. If the network structure is very dense for example, but the frequency of all relations is very low, we can see this network is not so active. Therefore learning outputs can be low, although the network density is high f.e. The network data is used to measure the density of a network, the centrality of persons within a network, detect key persons and investigate the structure of the network [27] [20]. By calculating the density of the network, this is the proportion of ties within the network, we know how well the networked learning activities is distributed within the team. The centrality of a learning network is interesting to see if knowledge and learning is distributed over the whole network, or if it is centralized around a few people. The Network Awareness Tool will then automatically visualize the overall structure of the network to the users. Actors can not only see the list of people they indicated as being connected to, but they can also see a list of people within the network who mention him or her as a learning connection. A graphical representation of the ego- and overall network structure will be visualized reflecting the current state of the network. The visualisations and network data will be used to automatically compute SNA results about the density of a network, the centrality of persons within a network, the structure, cliques, etc, in real time or a specified period (to be able to go back in time to be able to study the network dynamics). With this methodology we can detect multiple (isolated) networks in the organization, connect ideas and foster collaboration beyond existing boundaries. Using the Network Awareness Tool organizations can link in with existing informal networks of practice and unlock their potential for organizational learning by giving them a voice and make their results more explicit within the organization. To incorporate the resources and the quality of the relations we will look at solutions in the field of semantic analysis. If we combine the tags of all the ties within a network we can investigate the overall content of the network, represented by a tag cloud of keywords, or another graphical presentation indicating the main topics the participants are learning about within their organizations. The use of tags can also make it possible to investigate if there is a shared language and possible a shared identity within a network and if the members are working towards a shared goal, tackle shared problems and address similar, different or compatible solutions. To gain more insight into this aspect we also included a quality rating on the level of the overall network. By asking all members of a particular network to rate the impact this network has on individual growth, developing their practice and the organizations as a whole, we can make explicit the informal learning potential of a certain network and see how this changes over time using repeated measurements. When presenting a visualization of the whole network, participants can see not only ‘who talks to whom’, but also ‘what they are talking about’ [34]. 4. CONCLUSION. As more and more people participate in multiple networks – learning by observation in some, and participating strongly in others – extending the scope of learning to include lean and rich engagement in social networks is becoming more important for understanding individual experience in a multi-dimensional, multi-membership, and multi-identity world. Informal Learning thus engages with a wider view of the influences and impacts on individual’s ideas and knowledge acquisition, a view that is synergistic with the greater availability of information and social contacts accompanying developments on the Internet in an increasingly networked society [35]. Building the Network Awareness Tool we not only assist learners in making their connected world more visible but will also assist strategic networked learning by providing insight in what possible networks to join. This tool emphasises the relational approach to learning through which we can gain further insight about who learns from whom, what they learn from each other, what kind of interaction happens between people who learn together, how frequently learning interactions happen over time. In our research on social professional development networks among teachers in and between schools, we find that working with these visualizations stimulates a networking attitude amongst teachers in the school towards learning. They become aware that they are not alone in their classroom and that professional development is also a social activity; one that is spontaneous and deeply connected to day-to-day challenges in the workplace. Another advantage of these visualizations is that they serve as very concrete artefacts for the teachers to help them reflect on how they act as networkers building a social space for informal learning. This research shows that the presented methodology is a useful research driven intervention tool to detect, connect and facilitate informal networked learning. With this methodology we can detect multiple (isolated) networks in the organization and connect ideas and stimulate participants to think of solutions to support their own professional development in certain domains. Using this approach, organizations can link in with existing informal networks of practice and unlock their potential for organizational learning by giving them a voice and make their results more explicit within the organization. 5. DISCUSSION. In the future we would like to apply this methodology to the virtual world and investigate online teacher professional environments like Open Education Resource groups, Online discussion for a and other initiatives. We believe that if we apply this methodology to the virtual world, we can collect even more data, including data form communication via email, online discussion, blogs commentaries, or twitter streams, as well as hyperlink analyses of connectivity across sites. Semantic analysis can be conducted for creating tag clouds dealing with the content of the networks. Social networks analysis can be applied to datasets who interacts with whom and who downloads resources, etc. If we use this set-up a more holistic and full story can be created about the online informal learning activities of people and organisations can therefore analyse their users and see how to support and encourage online communities to share and learn from each other. It is tools like these we believe that can extend the discussion on the application of Learning Analytics and this paper is an attempt to stimulate a discussion amongst researchers coming from Technology Enhanced Learning, Networked Learning, Data Mining, Artificial Intelligence and Learning Analytics about technological solutions and methodologies to gather and analyse relational data on learning to create a holistic view of peoples off- and online informal life long learning activities in education, work and society."