(PDF) Learn-B: a social analytics-enabled tool for self-regulated workplace learning

Learn-B: A Social Analytics-enabled Tool for Self-regulated Workplace Learning Melody Siadaty1, 2 , Dragan Gašević1, 2, Jelena Jovanović1,2, 3 , Nikola Milikić3, Zoran Jeremić3, Liaqat Ali1, Aleksandar Giljanović2 and Marek Hatala2 1 School of Computing and Information Systems, Athabasca University, Canada 2 School of Interactive Arts and Technology, Simon Fraser University, Canada 3 FON-School of Business Administration, University of Belgrade, Serbia

,

,

,

,

,

,

,

ABSTRACT Such a perspective to learning is especially important in the con- text of workplace [5], where learning is social, affects and is af- In this design briefing, we introduce the Lear-B environment, fected by the social context and the available collective our attempt in designing and implementing a research prototype knowledge. to address some of the challenges inherent in workplace learn- ing: the informal aspect of workplace learning requires To keep up with and adapt to the contextual needs of workplace knowledge workers to be supported in their self-regulatory settings, learning at workplace mostly happens as a by-product learning (SRL) processes, whilst its social nature draws attention of work. This “on-demand” and informal approach to learning to the role of collective in those processes. Moreover, learning at [1] requires contemporary knowledge workers to have Self- workplace is contextual and on-demand, thus requiring organi- Regulatory Learning (SRL) skills in identifying their learning zations to recognize and motivate the learning and knowledge needs and conducting appropriate learning strategies to attain building activities of their employees, where individual learning them [8]. The majority of conventional interpretations of SRL goals are harmonized with those of the organization. In particu- are based on an individualistic perspective, where the impact of lar, we focus on the analytics-based features of Lear-B, illustrate the collective is often assumed less significant than individual- their design and current implementation, and discuss how each based factors [6]. Such perspectives contradict the nature of the of them is hypothesized to target the above challenges. workplace, where individuals’ work and learning activities are highly social and collective-centred. The recent research on Categories and Subject Descriptors workplace learning clearly stresses the role of the collective and J.1 [Administrative Data Processing] Education; K.3.1 other forms of social exchange in both individual learning and [Computer Uses in Education] Collaborative learning organizational development [4][1]; findings on patterns of de- fining learning goals in the workplace show that in the process Keywords of setting and managing their learning goals, individuals draw Learning Analytics, workplace learning, self-regulated learning, from and contribute to the collective knowledge in their organi- collaborative learning, semantic technologies zation [8]. To support users’ in their SRL processes in modern workplaces 1. INTRODUCTION as well as scaffolding organizational learning, there is a need for In the last few years, the growing emergence and acceptance of systems that collect learning–related contributions, re-aggregate social software tools, social media and Social Web (Web 2.0) and analyse them to create further new knowledge, and make paradigm have brought forth a new perspective to the concept of this new knowledge available to users. Such new knowledge can learning [4][14][7], demonstrating a transition from convention- be beneficial to users in every step of their learning process from al pedagogical approaches to a more social and collective identifying their learning needs and setting their goals (e.g. they knowledge paradigm of learning, in that creativity, social- can get aware how other employees with similar organizational embeddedness, and the capacity to gain knowledge from a sea of positions have defined their goals), to monitoring their learning collective are highly expected and valued [9][13]. progress and comparing it with that of their colleagues who hold the same position or work in the same project, and sharing and documenting their learning experiences (e.g. by observing how actively their colleagues are sharing their learning experiences and comparing it with their own sharing activities, or to see how Permission to make digital or hard copies of all or part of this work for their shared knowledge has been useful to other members of the personal or classroom use is granted without fee provided that copies organization). are not made or distributed for profit or commercial advantage and that Designing systems that unlock the collective knowledge, and the copies bear this notice and the full citation on the first page. To copy collective intelligence in higher levels of inference for the pur- otherwise, or republish, to post on servers or to redistribute to lists, re- pose of scaffolding learning, however, is not a straightforward quires prior specific permission and/or a fee. LAK12: 2nd International Conference on Learning Analytics & task [4]. Semantic technologies and Linked Data paradigm Knowledge, April 29– May 2, 2012, Vancouver, BC, Canada could provide the required technical backbone for tackling this Copyright 2010 ACM 1-58113-000-0/00/0010…$10.00. challenge. Today’s knowledge workers often use diverse tools and services in their everyday working and learning practices; grated tools (i.e. MediaWiki, Elgg and the Tagging Tool), pro- therefore, the traces and outcomes of their activities are dis- cessing and analyzing the gathered data, and providing users persed among different tools/services that often lack the capabil- with the resulting feedback and analytics. In particular, Event ity of interchanging and/or integrating user’s data. If properly Dispatcher (Figure 1.K) is responsible for processing all of the applied, the Linked Data paradigm and the associated Semantic events occurring in the Lear-B environment, storing them into Web technologies would enable meaningful data integration and the RDF repository (Figure 1.A) and distributing them to other knowledge structuring. services. Analytics Service (Figure 1.L) is responsible for pro- Needless to say, to be successfully deployed and to lead to the cessing and analyzing the data about users’ learning activities expected results, these advanced technologies need to be sup- and their interaction with diverse kinds of learning resources. It ported by proper pedagogical and motivational approaches. We makes use of the interaction data stored in the RDF repository to base the foundations of our pursued pedagogical approach on a provide users with feedback, primarily through different kinds well-known organizational knowledge building model proposed of visualizations, to support them in planning, performing and by Nonaka and Takeuchi [10] (to address the challenge of har- monitoring their learning process. monizing individual and organizational learning), and extend it with SRL practices (to support users’ in initiating and conduct- ing their individual learning processes) [11], and motivational elements [12] (to address the challenge of motivating users to share their personal knowledge and learning experiences, and contributing to the collective knowledge in their organization). For this pedagogical framework to work effectively, we hypoth- esize that Learning Analytics (i.e. collecting users’ contribu- tions, aggregating them, analyzing them and reporting back to the users and the organization) play an important role: it allows for the organization to better align its learning objectives with those of its employees by knowing about their learning practic- Figure 1. The architecture of the Lear-B Environment es; it supports users’ SRL processes by providing them with the Usage Information is one type of the provided analytics which necessary input from the social context of the workplace; and it comes in the form of statistics, Social Waves or the collective enhances the motivation of individuals to take part in learning stand. Derived from the collected knowledge within the system, and knowledge building activities and sharing their experiences this functionality supports the recommender services (Figure by providing them with feedback from the collective. In this de- 1.H-I) and more importantly, provides users with analytics rep- sign briefing, we introduce the Lear-B environment, our attempt resenting the collective knowledge around a resource and assists in designing and implementing a research prototype to support them in planning their learning processes. Statistics and Social workplace learning that addresses the above challenges. Lear-B Wave analytics are implemented as a set of various visualization stands for Learning Biosis (“biosis” meaning a way of life), i.e. charts, each conveying the intended feedback/analytics data. The learning as a way of life. In particular, in this design briefing we feedback reflecting the collective stand about a learning resource report on the learning analytics aspects presently supported by comes in diverse forms such as annotations, reflections (e.g. Lear-B. comments and notes), ratings and tags of other users. 2. THE LEAR-B ENVIRONMENT For instance, illustrates how each organizational objective, de- The design of the Lear-B environment was driven by the re- fined in terms of competences, is accompanied by statistical ana- quirements for effective learning and knowledge building in or- lytics such as the number of users who have acquired that com- ganizational settings. It is designed to integrate different tools petence and their roles in the organization, and the Social Wave that employees often interact with during their everyday (work- stream of that competence showing the activities performed on ing and learning) practices. In particular, so far we have inte- or events happened around it over a certain period. Such analyt- grated a wiki (MediaWiki), a social networking and collabora- ics represent the “popularity” of a given competence, indicating tion platform (Elgg), and a bookmarking tool (Tagging tool – whether and to what extent it is (socially) alive. The comments implemented within this research as a Firefox plugin). Lear-B of other users can be viewed under the Comments tab in . The serves as the central hub for this integrated environment, and re- recommendation of a learning path, via the Learning Path Rec- lies on an interlinked set of ontologies as its underlying (linked) ommender service (Figure 1.H), to achieve a competence (in this data model. These ontologies are available at: research, each learning path is comprised of one or more learn- http://goo.gl/Saui4. A current demo of the main functionalities ing activities that lead to the attainment of a specific competence of the Lear-B environment is available at: http://goo.gl/RaiIm. at a specific level) is further augmented with analytics such as the number of users who have successfully finished this learning Figure 1 illustrates the multi-layer architecture of Lear-B which path or a revision of it, or are still working on it, or have aban- can be adapted to and applied in a wide range of organizations. doned it. Users can also see the organizational positions of users There is no strong boundary between the layers and components in each of the above categories (i.e. active, finished, abandoned). defined within each layer. In this design briefing, we only focus Similar to the competences, each learning activity is also ac- on the Analytics-enabled functionalities provided within the companied by Social stream and collective stand analytics. Processing Service Group. This service group is responsible for tracking all the events that happen in Lear-B, and other inte- Figure 2. Analytics - Usage Information provided for each organizational objective a) Statistics b) Social Waves Progress-o-meters represent another type of the provided analyt- ics; they aid users to monitor their learning progress in the or- ganizational context, by showing them their progress flow in achieving their defined learning goals and the competences in- cluded within those goals, and are implemented as a set of line charts (Figure 3). Moreover, Progress-o-meters provide users with a comparison of their progress flow with that of their col- leagues who have the same learning goal (e.g., a goal shared by the members of a project), or are working on the same compe- tence. We hypothesize that observing oneself within the social context of the organization helps users to monitor their progress toward achieving their goals, thus also assisting them in further regulating their learning strategies. Figure 4. Analytics – Knowledge Sharing Profiles Motivational Messages are another type of provided analytics which aim to support users’ stronger engagement with the sys- tem. Generally, a user (learner) model represents user knowledge, goals, interests, and other features that allow for bet- ter recommendations or provided adaptivity by the sys- tem. Opening the learner model may bring additional benefits to users, allowing them to take charge of their own learning expe- rience. However, collecting explicit data from users is often challenging and a strong motivation is needed on learners’ part to provide explicit feedback about their learning [1]. Motiva- tional Messages aim to tackle this challenge by providing users with personalized messages indicating to what extent the collec- tive has opened their models in terms of sharing their personal preferences and learning experiences. For instance, Figure 5 Figure 3. Analytics – Progress-o-meters shows a set of motivational messages related to the degree of Knowledge Sharing Profiles inform users of their reflections, in completeness of a user’s ‘preferences’ compared to other users, terms of sharing their learning resources, within an organization. where these preferences are used to adjust the recommendations Via this type of provided analytics, users can see how actively generated for the user. they are sharing each of their learning resources, and also com- pare their sharing activities with the average within their organi- zation (Figure 4). As a factor targeting individuals’ extrinsic motivation [12], we hypothesize that such feedback can help us- ers to regulate their knowledge sharing activities. Figure 5. Analytics – Motivational Messages Last but not least, the Analytics Service supports the harmoniza- 5. REFERENCES tion of individual and organizational learning objectives. Brows- [1] Bull, S., Kay, J. 2010. Open Learner Models. In R. Nkam- ing the different forms of Analytics available for a certain com- bou, J. Bordeau & R. Miziguchi (eds), Advances in Intelli- petence, updates the managers of an organization on, for in- gent Tutoring Systems, Springer. 301-322. stance, how frequently this specific competence has been used within the organization, in the context of which learning goals it [2] Fenwick, T. 2008. Understanding relations of individual- has been used, by users of what organizational positions, and collective learning in work: A review of research. Man- what the main issues regarding this competence are. This allows agement Learning. 39 (3), 227-243. managers to apply any necessary modifications in the definition [3] Fischer, G. 2000. Lifelong Learning - More Than Training. of the competence itself or the learning paths associated with it, Journal of Interactive Learning Research, 11 (3/4), 265- to better harmonize organizational and individual learning 294. needs. Also discovery of emerging competences or other learn- [4] Gruber, T. 2008. Collective Knowledge Systems: Where ing resources can be learned via this service. On the other hand, the Social Web meets the Semantic Web. Journal of Web if some user-created competences are frequently being re-used Semantics, 6(1), pp. 4-13. by members of an organization; the managers might consider them as ‘emerging’ organizational goals. As can be seen, organ- [5] Hart, J. 2010. The State of Learning in the Workplace To- izational goals are also dynamic and can evolve via the contribu- day. Center for Learning and Performance Technologies tions of the community members. Accordingly, this targets indi- website. Available online at: http://goo.gl/Ipt2g viduals’ intrinsic motivation for knowledge sharing by giving [6] Jackson, T., MacKenzie, J., & Hobfoll, S. E. 2000. Com- them the feeling of being competent in contributing to the or- munal aspects of self-regulation. In M. Boekaerts, P. R. ganization’s goals and objectives. Pintrich, and M. Zeidner (Eds.), Handbook of self- regulation. San Diego, CA: Academic Press. 275-299. 3. CONCLUSIONS [7] Jovanovic, J., Gasevic, D., Torniai, C., Bateman, S., Hatala, In this design briefing, we demonstrated the analytics-based fea- M. 2009. The Social Semantic Web in Intelligent Learning tures of Lear-B, which are built on Semantic technologies and Environments - State of the Art and Future Challenges. In- Linked Data paradigm, and backed with an extended pedagogi- teractive Learning Environments, 17(4), 273-308. cal and motivational framework. In our empirical work with Lear-B, we aim to answer if and to what extend these features, [8] Littlejohn, A., Margaryan, A, and Milligan, C. 2009. Chart- along with the other functionalities provided within Lear-B, ad- ing collective knowledge: Supporting self-regulated learn- dress the existing challenges in supporting workplace learning. ing in the workplace. In. Proc. of the 9th IEEE Interna- In particular, to support users’ self-regulatory practices in the tional Conference on Advanced Learning Technologies, context of workplace learning, we hypothesize that the usage in- 208-212. formation analytics accompanying each learning resource in [9] McLoughlin, C. and Lee, M.J.W. 2010. Pedagogy 2.0: Lear-B (i.e. statistics, Social Waves and the collective stand) as- Critical challenges and responses to Web 2.0 and social sist users in planning their learning goals; Progress-o-meters, on software in tertiary teaching. In M.J.W. Lee & C. the other hand, provide users with feedback on their progress McLoughlin (Eds), Web 2.0-based e-learning: applying flow and thus help them with monitoring and evaluating their social informatics for tertiary teaching. Hershey, PA: In- learning progress. Knowledge Sharing Profiles inform users of formation Science Reference. 43–69. their reflections, in terms of sharing their learning resources and [10] Nonaka, I., & Takeuchi, H. 1995. The knowledge-creating experiences within an organization. Accordingly, we propose company. How Japanese companies create the dynamics of that these profiles support users to align their reflections and innovation. New York: Oxford University Press. sharing of their learning resources and experiences. Motivation- al Messages are another analytics-based means designed to fos- [11] Siadaty, M., Jovanović, J., Pata, K., Holocher-Ertl, T., ter users’ contributions and to motivate them to provide higher Gašević, D., Milikić, N. 2011. A Semantic Web-enabled quality inputs to the system. Tool for Self-Regulated Learning in the Workplace. In Proc. 11th IEEE Intl. Conf. on Advanced Learning Tech- nologies. (Athens, Athens, GA , USA, 2011). 66-70. 4. ACKNOWLEDGMENTS This demonstration was partially supported/co-funded by [12] Siadaty, M., Jovanović, J., Gašević, D., Jeremić, Z., Holo- NSERC, Athabasca University, and the European Community cher-Ertl, T. 2010. Leveraging Semantic Technologies for under the Information and Communication Technologies theme Harmonization of Individual and Organizational Learning. of the 7th Framework Program for R&D. This document does In Proc. 5th Euro. Conf. Technology-enhanced Learning. not represent the opinion of NSERC, Athabasca University, and (Barcelona, Spain, 2010). 340-356. the European Community, and NSERC, Athabasca University, [13] Tynjälä, P. 2008. Perspectives into learning at the work- and the European Community are not responsible for any use place, Educational Research Review, 3, 130-154. that might be made of its content. [14] Vassileva, J. 2008. Towards Social Learning Environ- ments, IEEE Trans. on Learning Technologies, 1 (4), 199- 214.