Explanations are central to improving transparency, trust, and user satisfaction in recommender systems (RS), yet it remains unclear how different explanation formats (visual vs. textual) are suited to users with different personal characteristics (PCs). To this end, we report a within-subject user study (n=54) comparing visual and textual explanations and examine how explanation format and PCs jointly influence perceived control, transparency, trust, and satisfaction in an educational recommender system (ERS). Using robust mixed-effects models, we analyze the moderating effects of a wide range of PCs, including Big Five traits, need for cognition, decision making style, visualization familiarity, and technical expertise. Our results show that a well-designed visual, simple, interactive, selective, easy to understand visualization that clearly and intuitively communicates how user preferences are linked to recommendations, fosters perceived control, transparency, appropriate trust, and satisfaction in the ERS for most users, independent of their PCs. Moreover, we derive a set of guidelines to support the effective design of explanations in ERSs.

Educational recommender systems (ERS) are playing a pivotal role in providing recommendations of personalized resources and activities to students, tailored to their individual learning needs. A fundamental part of generating recommendations is the learner modeling process that identifies students’ knowledge state. Current ERSs, however, have limitations mainly related to the lack of transparency and scrutability of the learner models as well as capturing the semantics of learner models and learning materials. To address these limitations, in this paper we empower students to control the construction of their personal knowledge graphs (PKGs) based on the knowledge concepts that they actively mark as ’did not understand (DNU)’ while interacting with learning materials. We then use these PKGs to build semantically-enriched learner models and provide personalized recommendations of external learning resources. We conducted offline experiments and an online user study (N=31), demonstrating the benefits of a PKG-based recommendation approach compared to a traditional content-based one, in terms of several important user-centric aspects including perceived accuracy, novelty, diversity, usefulness, user satisfaction, and use intentions. In particular, our results indicate that the degree of control students are able to exert over the learner modeling process, has positive consequences on their satisfaction with the ERS and their intention to accept its recommendations.

Explainable recommender systems (RS) have traditionally followed a one-size-fits-all approach, delivering the same explanation level of detail to each user, without considering their individual needs and goals. Further, explanations in RS have so far been presented mostly in a static and non-interactive manner. To fill these research gaps, we aim in this paper to adopt a user-centered, interactive explanation model that provides explanations with different levels of detail and empowers users to interact with, control, and personalize the explanations based on their needs and preferences. We followed a user-centered approach to design interactive explanations with three levels of detail (basic, intermediate, and advanced) and implemented them in the transparent Recommendation and Interest Modeling Application (RIMA). We conducted a qualitative user study (N = 14) to investigate the impact of providing interactive explanations with varying level of details on the users’ perception of the explainable RS. Our study showed qualitative evidence that fostering interaction and giving users control in deciding which explanation they would like to see can meet the demands of users with different needs, preferences, and goals, and consequently can have positive effects on different crucial aspects in explainable recommendation, including transparency, trust, satisfaction, and user experience.

Providing system-generated explanations for recommendations represents an important step toward transparent and trustworthy recommender systems. Explainable recommender systems provide a human-understandable rationale for their outputs. Over the past two decades, explainable recommendation has attracted much attention in the recommender systems research community. This paper aims to provide a comprehensive review of research efforts on visual explanation in recommender systems. More concretely, we systematically review the literature on explanations in recommender systems based on four dimensions, namely explanation aim, explanation scope, explanation method, and explanation format. Recognizing the importance of visualization, we approach the recommender system literature from the angle of explanatory visualizations, that is using visualizations as a display style of explanation. As a result, we derive a set of guidelines that might be constructive for designing explanatory visualizations in recommender systems and identify perspectives for future work in this field. The aim of this review is to help recommendation researchers and practitioners better understand the potential of visually explainable recommendation research and to support them in the systematic design of visual explanations in current and future recommender systems.

Despite the acknowledgment that the perception of explanations may vary considerably between end-users, explainable recommender systems (RS) have traditionally followed a one-size-fits-all model, whereby the same explanation level of detail is provided to each user, without taking into consideration individual user’s context, i.e., goals and personal characteristics. To fill this research gap, we aim in this paper at a shift from a one-size-fits-all to a personalized approach to explainable recommendation by giving users agency in deciding which explanation they would like to see. We developed a transparent Recommendation and Interest Modeling Application (RIMA) that provides on-demand personalized explanations of the recommendations, with three levels of detail (basic, intermediate, advanced) to meet the demands of different types of end-users. We conducted a within-subject study (N=31) to investigate the relationship between user’s personal characteristics and the explanation level of detail, and the effects of these two variables on the perception of the explainable RS with regard to different explanation goals. Our results show that the perception of explainable RS with different levels of detail is affected to different degrees by the explanation goal and user type. Consequently, we suggested some theoretical and design guidelines to support the systematic design of explanatory interfaces in RS tailored to the user’s context.

Learning analytics (LA) and Educational data mining (EDM) have emerged as promising technology-enhanced learning (TEL) research areas in recent years. Both areas deal with the development of methods that harness educational data sets to support the learning process. A key area of application for LA and EDM is learner modelling. Learner modelling enables to achieve adaptive and personalized learning environments, which are able to take into account the heterogeneous needs of learners and provide them with tailored learning experience suited for their unique needs. As learning is increasingly happening in open and distributed environments beyond the classroom and access to information in these environments is mostly interest-driven, learner interests need to constitute an important learner feature to be modeled. In this paper, we focus on the interest dimension of a learner model and present Wiki-LDA as a novel method to effectively mine user’s interests in Twitter. We apply a mixed-method approach that combines Latent Dirichlet Allocation (LDA), text mining APIs, and wikipedia categories. Wiki-LDA has proven effective at the task of interest mining and classification on Twitter data, outperforming standard LDA.

In the last few years, there has been a growing interest in learning analytics (LA) in technology-enhanced learning (TEL). Generally, LA deals with the development of methods that harness educational data sets to support the learning process. Recently, the concept of open learning analytics (OLA) has received a great deal of attention from LA community, due to the growing demand for self-organized, networked, and lifelong learning opportunities. A key challenge in OLA is to follow a personalized and goal-oriented LA model that tailors the LA task to the needs and goals of multiple stakeholders. Current implementations of LA rely on a predefined set of questions and indicators. There is, however, a need to adopt a personalized LA approach that engages end users in the indicator definition process by supporting them in setting goals, posing questions, and self-defining the indicators that help them achieve their goals. In this paper, we address the challenge of personalized LA and present the conceptual, design, and implementation details of a rule-based indicator definition tool to support flexible definition and dynamic generation of indicators to meet the needs of different stakeholders with diverse goals and questions in the LA exercise.

Knowledge Management (KM) and Technology Enhanced Learning (TEL) became a very important issue in modern organizational professional learning and work process integration. Former learning and KM theories which characterize knowledge as a thing or process no longer fit today's digital world where the amount of required information is no more manageable and the half-time of knowledge in general is rapidly decreasing. Younger approaches such as the Learning as a Network (LaaN) theory describe knowledge as complex and emergent and put a heavier focus on knowledge networking. The LaaN theory further stresses the convergence of the learning and work processes in professional learning settings and views KM and TEL as two sides of the same coin. Driven by the LaaN theory, the Professional Reflective Mobile Personal Learning Environments (PRiME) project describes an integrated KM and TEL framework which connects learning and work processes. It enables the professional learner to harness implicit knowledge and offers knowledge networking at three different layers: the Personal Learning Environment (PLE), the Personal Knowledge Network (PKN) and the Network of Practice (NoP). Continuous knowledge networking results in constant evolution of knowledge leading to personal as well as organizational learning.

Massive Open Online Courses (MOOCs) have a remarkable ability to expand access to a large scale of participants worldwide, beyond the formality of the higher education systems. MOOCs support participants to be actively involved in collaborative learning and construct their own learning experience in a variety of domains. However, one of the biggest challenges facing MOOCs is how to assess the learners’ performance in a massive learning environment beyond traditional automated assessment methods. To address this challenge, peer assessment has been proposed as an effective assessment method in MOOCs. The problem is, however, how to ensure the quality of the peer assessment in terms of validity and reliability. Moreover, assessment in blended MOOCs (bMOOCs) introduces unique challenges regarding the best peer assessment model in a learning environment that brings together face-to-face interactions and online activities. This paper presents the details of a study conducted to investigate peer assessment in bMOOCs. The study results show that flexible rubrics have the potential to make the feedback process more accurate, credible, transparent, valid, and reliable, thus ensuring the quality of the peer assessment task.

A key area of application for Learning Analytics (LA) and Educational Data Mining (EDM) is lifelong learner modeling. The aim is that data gathered from different learning environments would be fed into a personal lifelong learner model that can be used to foster personalized learning experiences. As learning is increasingly happening in open and networked environments beyond the classroom and access to knowledge in these environments is mostly context-sensitive and interest-driven, learner’s contexts and interests need to constitute important features to be modeled. The context data of a learner as it is already represented by the Learning Context Data Model (LCDM) specification, describes the learner’s activities, her biological conditions, as well as the characteristics of the learning environment. Towards a lifelong learner model, a model consisting of context data can further be refined with an evolving set of interests. This paper describes an approach to extend the existing LCDM specification with interests, taking into account the importance of the interests as well as their evolution over time.

Learning analytics (LA) deals with the development of methods that harness educational data sets to support the learning process. To achieve particular learner entered LA objectives such as intelligent feedback, adaptation, personalization, or recommendation, learner modeling is a crucial task. Learner modeling enables to achieve adaptive and personalized learning environments, which are able to take into account the heterogeneous needs of learners and provide them with tailored learning experience suited for their unique needs. In this paper, we focus on learner modeling in academic networks. We present theoretical, design, implementation, and evaluation details of PALM, a service for personal academic learner modeling. The primary aim of PALM is to harness the distributed publication information to build an academic learner model.

Massive Open Online Courses (MOOCs) have gained a lot of attention in the last years as a new technology enhanced learning (TEL) approach in higher education. MOOCs provide more educational opportunities to a massive number of learners to attend free online courses around the globe. Discussions around MOOCs have been focusing on the potential, social, institutional, technological, relevance, and marketing issues and less on the quality design of MOOC environments. Several studies have reported a high drop-out rate in average of 95% of course participants and other pedagogical problems concerning assessment and feedback. Thus, the quality of MOOCs design is worth additional investigation. Although several studies identified a large set of criteria to the successful design of TEL systems in general, not all of them can be used in the MOOC context, due to some unique features of MOOCs. This study is a first step towards identifying specific criteria that need to be considered when designing and implementing MOOCs. The results of this empirical study are based on a large survey targeting learners as well as professors, both with MOOC experience. As a result, we identified and rated 74 indicators classified into our two main dimensions of pedagogical and technological criteria distributed over six categories. From these, the learning analytics and assessment categories were found to be the key features for effective MOOCs.

Learning analytics tools should be useful, i.e., they should be usable and provide the functionality for reaching the goals attributed to learning analytics. This paper seeks to unite learning analytics and action research. Based on this, we investigate how the multitude of questions that arise during technology-enhanced teaching and learning systematically can be mapped to sets of indicators. We examine, which questions are not yet supported and propose concepts of indicators that have a high potential of positively influencing teachers' didactical considerations. Our investigation shows that many questions of teachers cannot be answered with currently available research tools. Furthermore, few learning analytics studies report about measuring impact. We describe which effects learning analytics should have on teaching and discuss how this could be evaluated.

The Personal Learning Environment (PLE) driven approach to learning suggests a shift in emphasis from a teacher driven knowledge-push to a learner driven knowledge pull learning model. One concern with knowledge-pull approaches is knowledge overload. Thus, there is a crucial need for knowledge filters to help learners cope with the problem of knowledge overload. In this paper, we present the details of PLEM as a Web 2.0 driven service for personal learning management, which acts as a knowledge filter for learning. The primary aim of PLEM is to harness the collective intelligence and leverage social filtering methods to help learners find quality knowledge nodes that can populate their PLEs.

Software development is today done in teams of software developers who may be distributed all over the world. Software development has also become to contain more social aspects and the need for collaboration has become more evident. The importance of teaching development methods used in collaborative development is of importance, as skills beyond traditional software development are needed in this modern setting. A novel, student centric approach was tried out at Tampere University of Technology where a new environment called KommGame was introduced. This environment includes a reputation system to support the social aspect of the environment and thus supporting the learners collaboration with each other. In this paper, we present the KommGame environment and how it was applied on a course for practical results.

Nonaka and his knowledge transformation model SECI revolutionized the thinking about organizations as social learning systems. He introduced technical concepts like hypertext into organizational theory. Now, after 15 years Web 2.0 concepts seem to be an ideal fit with Nonaka's SECI approach opening new doors for more personal, dynamic, and social learning on a global scale. In this paper, we present an extended view of blended learning which includes the combination of formal and informal learning, knowledge management, and Web 2.0 concepts into one integrated solution, by discussing what we call the Web 2.0 driven SECI model based learning process.

User-driven annotation of learning content allows the user to interact with it in a flexible and an intuitive manner and hence personalize the content. This interaction, in most cases, follows the "paper and pen" note-taking paradigm. Though not without its advantages, this paradigm does not scale when the courses are delivered online as web pages. These paper notes are not synchronous with the content, hence may seem out of context, they may be misplaced, and don’t make allowances for collaborative learning. In this paper we describe u-Annotate, a user-driven freeform digital ink annotation application for web e- Learning content which aims at facilitating the learner to annotate the online content with the aid of pen computing devices such as graphic tablets, etc. Learners can freely mark up the content, save the annotations for recall at a later date, as well as share these with other learners.

Social software is used widely in organizational knowledge management and professional learning. The PROLEARN network of excellence appreciates the trend of lowering the barriers between knowledge and learning management strategies for organizations and individuals. But, companies should not underestimate the needs for systematic support based on sound theories and technologies. We illustrate the requirements by examples and research issues for collaborative adaptive learning platforms for workplace learning in organizations

Our shared belief is that learning, like other human activities, cannot and is not confined within rigidly defined course systems or learning repositories, inclosing learning resources which cannot be tailored to the different learner's needs, skills, interests, preferences, goals, etc. Therefore, a learning environment, besides supporting communication between knowledge providers and consumers, has to be organized in a flexible manner based on different learner profiles. Learner modeling has become a highly challenging task to provide personalized, adaptive and context-based learning. The work presented in this paper addresses this issue by providing a meta-level solution for the description, transformation and matching of learner models, based on standards such as IMS LIP, IEEE PAPI, XML to foster the reuse and exchange of learner models between learning platforms, both by universities and corporations.