This paper presents the basis of the GENIE Learn project, a coordinated three-year research project funded by the Spanish Research Agency. The main goal of GENIE Learn is to improve Smart Learning Environments (SLEs) for Hybrid Learning (HL) support by integrating Generative Artificial Intelligence (GenAI) tools in a way that is aligned with the preferences and values of human stakeholders. This article focuses on analyzing the problems of this research context, as well as the affordances that GenAI can bring to solve these problems, but considering also the risks and challenges associated with the use of GenAI in education. The paper also details the objectives, methodology, and work plan, and expected contributions of the project in this context.

Generative artificial intelligence (GenAI) tools, such as large language models (LLMs), generate natural language and other types of content to perform a wide range of tasks. This represents a significant technological advancement that poses opportunities and challenges to educational research and practice. This commentary brings together contributions from nine experts working in the intersection of learning and technology and presents critical reflections on the opportunities, challenges, and implications related to GenAI technologies in the context of education. In the commentary, it is acknowledged that GenAI’s capabilities can enhance some teaching and learning practices, such as learning design, regulation of learning, automated content, feedback, and assessment. Nevertheless, we also highlight its limitations, potential disruptions, ethical consequences, and potential misuses. The identified avenues for further research include the development of new insights into the roles human experts can play, strong and continuous evidence, human-centric design of technology, necessary policy, and support and competence mechanisms. Overall, we concur with the general skeptical optimism about the use of GenAI tools such as LLMs in education. Moreover, we highlight the danger of hastily adopting GenAI tools in education without deep consideration of the efficacy, ecosystem-level implications, ethics, and pedagogical soundness of such practices.

Personalised feedback in MOOCs is often prevented by the non-scalability of traditional approaches and the lack of pedagogical grounding of the current learning analytics (LA) solutions. One way to tackle such limitations is the adoption of a participatory approach through the active positioning of MOOC instructors in the design and development of LA solutions. Nonetheless, there is a scarcity of empirical proposals supporting these approaches in MOOCs. To that end, the current paper presents e-FeeD4Mi, a web-based tool, incorporating a set of catalogues, recommendations and a process, that aim to support instructors in the design of human-centred LA-informed feedback. We conducted an evaluative study with 6 MOOC instructors who employed the tool into their course designs to assess e-FeeD4Mi usefulness, usability and associated workload. The evidence gathered permitted us to understand how e-FeeD4Mi support the design of human-centred LA-based feedback in MOOCs. Altogether, the results showed a good usability and participants' satisfaction regarding the use of the tool for shaping personalised feedback. At the same time, participants offered ideas for further tool enrichment. This study expands upon the current body of empirical research on the human-centred approaches in the design of LA-driven interventions in MOOCs.

A correction to this paper has been published: https://doi.org/10.1007/s00521-021-05904-z

This paper presents the H2O Learn (Hybrid and Human-Oriented Learning) project, a coordinated research project funded by the Spanish Research Agency, which just started in 2021 and will last for three years. The main goal of the H2O Learn project is to build Trustworthy and Human-Centered Learning Analytics (TaHCLA) solutions to support human stakeholders when designing, orchestrating and (self-,socially- or co-) regulating learning in Hybrid Learning (HL). The contributions of H2O Learn consider key requirements for trustworthy Artificial Intelligence (AI), as defined by the European Commission: 1) fostering human (i.e., teachers, learners…) agency; 2) enabling transparency of the Learning Analytics (LA) systems; 3) seeking socio-emotional and inclusive wellbeing; and 4) promoting accountability by enabling the assessment of algorithms and design processes.

The vision of a Personalised Learning (PL) experience in education is a long-standing notion advocated by educators. Research literature accounts for several pedagogical and technological approaches to tailor learning to learners’ variability. This paper aims to contribute to PL and Learning Design (LD) research by conceptualising CPELDS (Coherent PErsonalised Learning DeSign) as a framework outlining consistent PL design principles. CPELDS blends learner profiles addressing the dimensions of interests, strengths and needs with personalisation principles. Considering evidence about the effectiveness of Universal Design for Learning in empirical studies in special education, we transform and augment its guidelines into personalisation principles of CPELDS to address broader learners’ variability. The personalisation principles employed involve (i) fostering learner’s engagement, (ii) providing alternative information perception and comprehension modes, including differentiation, and (iii) supporting alternative means for learners’ action and expression to demonstrate acquired knowledge and skills. We anticipate the conceptualisation of CPELDS to stimulate the representational form of an LD tool towards supporting educators as designers of PL.

In this paper we attempt to measure the semantic distance between forum dialogues in five MOOCs of different subject matters (humanities, mathematics and computing). We examine if such distance can provide evidence of transferable linguistic features, suggesting transferability of forum classification models. For this purpose, we employed Latent Dirichlet Allocation in order to extract the main discussion topics for each MOOC and proceeded to a pairwise semantic comparison between them. We focused on three categories of forum topics, i.e., Content-Related, Logistics-Related and Social Building. In the Content-Related discussions, the study confirmed semantic distance between the humanities course and the other four, while the distances among the mathematics and computing courses were shorter than expected. On the other hand, Logistics-Related and Social Building topics contained a more transferable set of features. The approach that we followed provides insights that could help MOOC researchers interpret the transferable features of their available data to employ in their classification tasks.

This paper presents SLEek, an ontology for the context-aware recommendation of learning activities in Smart Learning Environments (SLEs). SLEek creates an actor-artifact network that is especially suitable for the context-aware recommendation of activities across spaces in formal and informal contexts. SLEek implementation reuses vocabularies from the Web of Data and is currently used in a dataset of 17K learning activities.

Detecting learners who face problems in MOOCs usually poses difficulties due to the high instructor-learners ratio, the diversity of the population, and the asynchronous participation mode. Existing solutions mainly draw on self-reported problems in discussion forums and on dashboards displaying learners’ activity traces. However, these approaches cannot scale up easily or do not consider the course learning design. This paper presents a conceptual framework aimed at guiding MOOC instructors in the identification of potential learners' problems and indicators of such problems, considering the learning design of the course (e.g., types of activities, difficulty, etc.). An instrumental qualitative case study served for the evaluation and refinement of the framework. The results showed that the framework positively helped instructors to reflect on potential learners’ problems they had not considered beforehand, and to associate such problems with a set of indicators related to their learning designs.

Peer reviews offer many learning benefits. Understanding students’ engagement in them can help design effective practices. Although learning analytics can be effective in generating such insights, its application in peer reviews is scarce. Theory can provide the necessary foundations to inform the design of learning analytics research and the interpretation of its results. In this paper, we followed a theory-based learning analytics approach to identifying students’ engagement patterns in a peer review activity facilitated via a web-based tool called Synergy. Process mining was applied on temporal learning data, traced by Synergy. The theory about peer review helped determine relevant data points and guided the top-down approach employed for their analysis: moving from the global phases to regulation of learning, and then to micro-level actions. The results suggest that theory and learning analytics should mutually relate with each other. Mainly, theory played a critical role in identifying a priori engagement patterns, which provided an informed perspective when interpreting the results. In return, the results of the learning analytics offered critical insights about student behavior that was not expected by the theory (i.e., low levels of co-regulation). The findings provided important implications for refining the grounding theory and its operationalization in Synergy.

Supervised classification models are commonly used for classifying discussions in a MOOC forum. In most cases these models require a tedious process for manual labeling the forum messages as training data. So, new methods are needed to reduce the human effort necessary for the preparation of such training datasets. In this study we follow an incremental approach in order to examine how soon after the beginning of a new course, we have collected enough data for training a supervised classification model. We show that by employing features that derive from a seeded topic modeling method, we achieve classifiers with reliable performance early enough in the course life, thus reducing significantly the human effort. The content of the MOOC platform is used to bias the topic extraction towards discussions related to (a) course content, (b) logistics, or (c) social interactions. Then, we develop a supervised model at the start of each week based on the topic features of all previous weeks and evaluate its performance in classifying the discussions for the rest of the course. Our approach was implemented in three different MOOCs of different subjects and different sizes. The findings reveal that supervised models are able to perform reliably quite early in a MOOC's life and retain a steady overall accuracy across the remaining weeks, without requiring to be trained with the entire forum dataset.

The effectiveness of using learning analytics for learning design primarily depends upon two concepts: grounding and alignment. This is the primary conjecture for the study described in this paper. In our design-based research study, we design, test, and evaluate teacher-facing learning analytics for an online inquiry science unit on global climate change. We design our learning analytics in accordance with a socioconstructivism-based pedagogical framework, called Knowledge Integration, and the principles of learning analytics Implementation Design. Our methodology for the design process draws upon the principle of the Orchestrating for Learning Analytics framework to engage stakeholders (i.e. teachers, researchers, and developers). The resulting learning analytics were aligned to unit activities that engaged students in key aspects of the knowledge integration process. They provided teachers with actionable insight into their students' understanding at critical junctures in the learning process. We demonstrate the efficacy of the learning analytics in supporting the optimization of the unit's learning design. We conclude by synthesizing the principles that guided our design process into a framework for developing and evaluating learning analytics for learning design.

While project-based software engineering courses aim to provide learning opportunities grounded in professional processes, it is not always possible to replicate every process in classrooms due to course constraints. Previous studies observed how students react to various processes and gave retroactive recommendations. In this study, we instead combine a field study on professional Agile (eXtreme Programming, XP) teams and an established team process taxonomy to proactively select team processes to incorporate in a project-based software engineering course. With collected knowledge from the field study, we choose three XP processes to augment the design of a mature software engineering project course. We choose processes that are 1) considered important by professionals, and 2) complete with respect to coverage of the taxonomy's main categories. We then compare the augmented course design with the original design in a case study. Our results suggest that 1) even without extra resources, adding these new processes does not interfere with learning opportunities for XP processes previously existing in the course design; 2) student teams experience similar benefits from these new processes as professional teams do, and students appreciate the usefulness and value of the processes. In other words, our approach allows instructors to make conscious choices of XP processes that improve student learning outcomes while exposing students to a more complete set of processes and thus preparing them better for professional careers. Course designers with limited resources are encouraged to use our methodology to evaluate and improve the designs of their own project-based courses.

Peer review is one approach to facilitate formative feedback exchange in MOOCs; however, it is often undermined by low participation. To support effective implementation of peer reviews in MOOCs, this research work proposes several predictive models to accurately classify learners according to their expected engagement levels in an upcoming peer-review activity, which offers various pedagogical utilities (e.g. improving peer reviews and collaborative learning activities). Two approaches were used for training the models: in situ learning (in which an engagement indicator available at the time of the predictions is used as a proxy label to train a model within the same course) and transfer across courses (in which a model is trained using labels obtained from past course data). These techniques allowed producing predictions that are actionable by the instructor while the course still continues, which is not possible with post-hoc approaches requiring the use of true labels. According to the results, both transfer across courses and in situ learning approaches have produced predictions that were actionable yet as accurate as those obtained with cross validation, suggesting that they deserve further attention to create impact in MOOCs with real-world interventions. Potential pedagogical uses of the predictions were illustrated with several examples.

With the aim of supporting instructional designers in setting up collaborative learning activities in MOOCs, this paper derives prediction models for student participation in group discussions. The salient feature of these models is that they are built using only data prior to the learning activity, and can thus provide actionable predictions, as opposed to post-hoc approaches common in the MOOC literature. Some learning design scenarios that make use of this actionable information are illustrated.

Collaborative learning can improve the pedagogical effectiveness of MOOCs. Group formation, an essential step in the design of collaborative learning activities, can be challenging in MOOCs given the scale and the wide variety in such contexts. We discuss the need for considering the behaviours of the students in the course to form groups in MOOC contexts, and propose a grouping approach that employs homogeneity in terms of students’ engagement in the course. Two grouping strategies with different degrees of homogeneity are derived from this approach, and their impact to form successful groups is examined in a real MOOC context. The grouping criteria were established using student activity logs (e.g. page-views). The role of the timing of grouping was also examined by carrying out the intervention once in the first and once in the second half of the course. The results indicate that in both interventions, the groups formed with a greater degree of homogeneity had higher rates of task-completion and peer interactions, Additionally, students from these groups reported higher levels of satisfaction with their group experiences. On the other hand, a consistent improvement of all indicators was observed in the second intervention, since student engagement becomes more stable later in the course.

Collaborative learning flow patterns (CLFPs) encode solutions to recurrent pedagogical problems, which have been successfully applied to the design of learning experiences. However, the pedagogical knowledge encoded in these patterns has seldom been exploited in learning analytics (LA). This paper analyzes four of the most common CLFPs to extract the intrinsic constraints that lead to a successful collaborative learning activity, and use them to enhance existing LA solutions. To understand the added value of applying such codified knowledge in LA, we present evidence from five authentic case studies in which such constraints aided university teachers in monitoring complex collaborative scripts. The results not only illustrate quantitatively such added value but also unearth qualitative benefits, such as raising practitioners' awareness about how the current state of activities may affect future phases of the script.

In blended learning scenarios, evidence needs to be gathered from digital and physical spaces to obtain a more complete view of the teaching and learning processes. However, these scenarios are highly heterogeneous, and the varying data sources available in each particular context can condition the accuracy, relevance, interpretability and actionability of the Learning Analytics (LA) solutions, affecting also the user's sense of agency and trust in such solutions. To aid stakeholders in making use of learning analytics, we propose a process to involve teachers in customizing multimodal LA (MMLA) solutions, adapting them to their particular blended learning situation (e.g., identifying relevant data sources and metrics). Since measuring the added value of adopting an LA solution is not straightforward, we also propose a concrete method for doing so. The results obtained from two case studies in authentic, blended computer-supported collaborative learning settings show an improvement in the sensitivity and F1 scores of the customized MMLA solution. Aside from these quantitative improvements, participant teachers reported both an increment in the effort involved, but also increased relevance, understanding and actionability of the results.

Recent technological advances in mobile devices enable the connection of classrooms with other virtual and physical spaces. Some approaches aim at helping teachers carry out learning situations across such spaces. However, these proposals tend to be isolated from other activities in teachers’ current common practices, and do not allow teacher-controlled flexibility of what students do during the enactment. Aiming to overcome such limitations, the Bucket-Server is a system that enables teachers to include learning buckets in their learning situations: containers of learning artifacts generated and/or consumed across-spaces by students during the enactment. Teachers create learning buckets at design time, configuring them with constraints to regulate the degree of freedom offered to the students. These learning buckets can be integrated into multiple existing technologies used in different educational spaces (e.g., web, physical and 3D virtual world spaces), thus helping embed buckets in the teachers’ current common practices.

Modeling artifact flow in learning design of CSCL processes can be helpful, since it aims to handle effectively dependencies among group or individual activities. However, the cognitive overload problem for the teacher-designer may impede efficient modeling. Therefore, it is convenient to assess such an overload issue through objective metrics, adopted from the process modeling field, and to detect the underlying causes. Such an analysis may allow the proposal of alternative ways of handling the modeling process without incurring in excessive work and cognitive load. Our findings through an analysis of a set of significant CSCL scenarios, suggest that artifact assignment, access mode and size of collaborative scenarios are important influencing factors.

Adaptive Collaborative Scripting systems provide learning benefits by adapting leaner scaffolding to the students and their current context. However, their development is still in its infancy and they are not widespread in the Technology Enhanced Learning TEL practice, which often uses VLEs like Moodle and other Web 2.0 tools. In order to assess the feasibility of applying the ACS approach on a larger scale, this paper presents the initial results of a short-term evaluation of the GLUE! suite of tools. The main goal of this specific evaluation process was to identify possible opportunities and ideas on how to design and deploy adaptive Computer- Supported Collaborative Learning (CSCL) activities using widespread VLEs and Web 2.0 tools in order to maximise community acceptance and lower development efforts. The main findings of the evaluation provide incentive to further explore both the impact and the complexity of the design and the deployment of adaptive collaboration scripts.

This paper proposes a novel method to detect anomalies in network traffic, based on a nonrestricted α-stable first-order model and statistical hypothesis testing. To this end, we give statistical evidence that the marginal distribution of real traffic is adequately modeled with α-stable functions and classify traffic patterns by means of a Generalized Likelihood Ratio Test (GLRT). The method automatically chooses traffic windows used as a reference, which the traffic window under test is compared with, with no expert intervention needed to that end. We focus on detecting two anomaly types, namely floods and flash-crowds, which have been frequently studied in the literature. Performance of our detection method has been measured through Receiver Operating Characteristic (ROC) curves and results indicate that our method outperforms the closely-related state-of-the-art contribution described in. All experiments use traffic data collected from two routers at our university-a 25,000 students institution-which provide two different levels of traffic aggregation for our tests (traffic at a particular school and the whole university). In addition, the traffic model is tested with publicly available traffic traces. Due to the complexity of α-stable distributions, care has been taken in designing appropriate numerical algorithms to deal with the model.

This work presents a CSCL case study, for evaluating the Collage editor (an IMS-LD compliant authoring tool for collaboration scripts). 21 postgraduate students were guided to work in dyads and design their own CLFP-based collaboration scripts using Collage editor. The study objective was to provide additional evaluation data regarding the usability and efficiency of Collage editor as a tool for supporting beginner CSCL script designers. Students evaluated Collage as a user-friendly and intuitive tool (even for non-experienced users), while also suggesting a number of additions that could further improve its pedagogical usefulness and flexibility.

Many studies show the advantages of using a learning design approach to help teachers and instructional designers to create technology enhanced collaborative learning activities. However, some of these studies highlight lacks in such approach. This paper addresses one of them: the representation of the flow of data in the learning activity models. A generic specification is proposed that aims at pointing out the conceptual needs that should be taken into account in the languages to improve the approaches.

In line with the suggestions deriving from the new problem-solving approaches to ID, this paper points out the need of integrating formalized descriptions of good practices, such as Collaborative Learning Flow Patterns (CLFP) with other representations able to support unskilled designers in selecting and instantiating scripts, setting up effective collaborative groups and their “social structures”. Based on these premises, the paper analyzes a real course designed with the backing of a CLFP-based authoring tool for the creation of CSCL scripts and describes the support offered by the Social Structures Representations (SSRs) in selecting and reusing scripts.

Communities of Practice (CoP) are nowadays widely used for implementing collaborative learning approaches, under the scope of contemporary learning theories. Several communication and/or collaboration web modules are used in order to support the organizational structure and facilitate the operational purpose of such communities. Following the same approach, Scientific Networks (SN) are widely used as a common method of expanding research and sharing academic resources. The need for appropriate tools in order to support the members of a CoP or a SN and further facilitate their collaborative tasks, or even resulted learning is often highlighted by the literature. In this paper we propose the use of Interaction Analysis indicators for that matter. Our experience from the implementation of such indicators, supporting the Kaleidoscope Network of Excellence is presented.

Advances in educational technology have enabled new forms of organizing and managing collaborative learning environments. One of the currents trends in research on Technology Enhanced Learning is the application of computer-interpretable scenarios, such as those formalized with IMS Learning Design (IMSLD), for the description and deployment of Computer Supported Collaborative Learning (CSCL) scripts. The life-cycle of IMS-LD scripts involves several phases, from the conception to their enactment in actual learning situations. In the context of collaborative learning, group/roles management and composition have special relevance, and are in fact some of the tasks that the teacher must accomplish before students can follow the scripts. This paper presents a software tool (InstanceCollage) that implements a pattern-based approach, which has been previously validated for the task of script design. In this way, it aims at facilitating the population of roles and groups generally described in IMS-LD scripts.

With the aim of facilitating teachers the design of units of learning and assessment, this paper proposes the use of different educational pattern types. Depending on the type of patterns, the resulting didactic materials are specified using IMS Learning Design or IMS Question and Test Interoperability extended with integrated Web2.0-like services.

Interaction Analysis (IA) is an essential element for formative evaluation, self-regulation and other important functions, especially in Computer Supported Collaborative Learning (CSCL) scenarios. However, several problems have impeded its regular use in real environments, in spite of the significant advancement of the field. This paper analyzes the main problems and alternatives encountered in the process of effectively supporting IA in CSCL, based on lessons learned from the authors’ active involvement in this field during the last decade. Thus, it proposes a decoupled architecture together with the use of a standard common data format for interaction as main components of a new framework for a global design process that will support practitioners, researchers and technology providers in a sustainable and transparent way.

Collaborative learning models are widely used in educational institutions. These models require a high interaction level among students and are mainly oriented towards in-class scenarios. But when collaborative models are deployed in a distant scenario, user expressiveness is significantly reduced thus creating a gap that hinders the effectiveness of this collaboration. A computer-supported model provides a set of tools to compensate for the distant scenario and reduce this gap. This paper presents the issues and solutions derived from the design and deployment of a complex collaborative model in a distant scenario. The course structure was captured using the Learning Design specification, and an architecture based on virtual network computing was used to provide the required collaborative tools. The course was included as part of a regular undergraduate program in three higher educational institutions.

This paper presents the design of a new simulation environment based on the well-known ns-2 network simulator with two significant characteristics. First, it enables the possibility of running a large number of simulations even if the computational and storage resources locally available are scarce. Second, it provides a number of features in order to facilitate collaboration among students. In this way, the proposed simulation environment enables the realization of many collaborative learning scenarios with pedagogical interest for computer networks education.

Data flow between tools cannot be specified using the current IMS Learning Design specification (LD). Nevertheless by specifying this data flow between tools, several degrees of activity automation may augment the system intervention opportunities for data flow management. Service automation, data flow automation and data flow validation may enhance the continuity of the learning design realization, reduce the student's cognitive load and obtain system-support for error prone situations. In this paper a novel approach based on the composition of LD and a standard workflow technology is proposed. Unlike other current approaches, our approach maintains interoperability with both LD and workflow standards. Then an architectural solution based on the composition approach is presented.

The success of groupware software largely depends on its capability for being reused in different collaborative scenarios without requiring significant software development efforts or user technical involvement. During the last decade several toolkits, frameworks, and specialized middleware software layers have been proposed as potential solutions to the groupware reusability problem. Those solutions cover common groupware functionality such as group formation and management, group awareness, shared workspace management, etc. This paper is focused on the reusable support to one of these common functionalities: the so-called object-level coordination that deals with multiple participants' sequential or simultaneous access to the same set of shared objects in a collaborative setting. The paper compares existing object-level coordination proposals and analyzes their capabilities and level of reuse. Starting from that analysis, and using a bottom-up approach, the paper also proposes a high-level reference model that identifies the desirable set of required functional elements, as well as their relationships, that reusable object-level coordination support should contain. The purpose of this reference model is twofold: to be used as a comparison framework for analyzing existing or yet-to-come object-level coordination building blocks; and, to be used as a starting point for developing new object-level coordination solutions (or redesigning existing ones) satisfying the whole set of requirements implicitly collected in the model. In addition to detailing the proposed reference model, the paper also illustrates and discusses both potential ways of using it

This paper describes an approach for modeling and implementing a collaborative learning situation, which is part of a real lifelong learning scenario in astronomy. We adopt and slightly adjust IMS Learning Design for computationally representing the script so that collaborative tools can be satisfactorily employed. We use Collage editor, which enables the re-use of general structures proposed by Collaborative Learning Flow Patterns, for authoring the script and Gridcole system for executing it. The paper also shows the log-files produced by the employed tools. The teacher can use these log-files to monitor the evolution of the process for evaluation and regulation purposes.

This paper proposes collaborative learning flow patterns (CLFPs), which represent best practices in collaborative learning structuring, as a central element of a kind of bi-directional linkage that facilitates that teachers can play the role of designers influencing in the behavior of CSCL (computer-supported collaborative learning) technological solutions. Additionally, this paper describes a technological approach for achieving such a scenario. That approach is based on the Collage authoring tool that provides CLFPs as IMS LD templates and the Gridcole system, capable of interpreting the resulting CLFP-based LDs and integrating the service-oriented tools needed to support the (collaborative) learning activities as prescribed in those LDs

The use of patterns in e-learning is being recently proposed with different purposes and scopes. This paper provides a unifying view of several representative proposals in order to situate and introduce the types of patterns that can be used for generating collaboration scripts that are suitable of being computationally represented and interpreted by learning management systems (LMSs). The paper also presents a hierarchical structure that relates these patterns according to their focus and granularity. The structure shows how the different types of patterns are complementary and should be considered in the process of creating a script. In addition to authoring tools and LMSs, there are other types of tools in which the application of patterns may be useful. This paper points out these tools locating them along the "life cycle" of the scripts

Interaction analysis has become a basic function in the field of collaborative learning, as a means for supporting both regulation and evaluation of collaborative learning processes. In spite of the fact that these processes rely on the same basic functionalities, there is a lack of proposals or systems that integrate them. The most outstanding difference between the tools that support either of the approaches is that they are oriented to different types of users. Then, an interaction analysis tool able to adapt to different roles would be able to perform both functionalities. This paper presents an experience performed in an authentic learning scenario using interaction analysis for supporting regulation for students and evaluation for teachers adapting its functionality depending on analysis needs of teacher and learners that participate in the activity. A first evaluation carried out shows that the participants assess positively this experience and consider the information received easy to understand, reliable, useful and capable to modify their way of collaboration.

CSCL systems can benefit from using a grid, since it offers a common infrastructure allowing an extended pool of resources that can provide supercomputing capabilities as well as specific hardware resources. Adopting a service oriented architecture such as OGSA can further benefit CSCL systems, enabling increased flexibility to adapt and reuse learning software. However, proliferation of services enormously challenges the search of appropriate ones. Common service discovery mechanisms, such as the Index Service or UDDI, perform poor searches since they rely on keyword matching and do not properly describe service functionality. To address these difficulties, ontologies can be used to formally describe services allowing for semantic searches. This paper proposes an ontology that describes learning services using educators abstractions in order to support the search. This way, service providers can describe their services in terms of the ontology, while educators can search for them using domain concepts. The integration of this enhanced service search mechanism in Gridcole, a real grid-based CSCL system, is described to illustrate its performance.

In this paper, we describe the experience of designing a collaborative learning activity for a traditional historical/cultural museum. The activity, based on a Mystery in the Museum story, involves collaboration of small groups of students through mobile handheld devices. An application has been built that permits authoring of such activities, while a usability evaluation study was performed that revealed some of the limitations of the design. The reported findings can be of use to those interested in following similar approaches in cultural and educational settings, and draw conclusions of general interest relating to interaction and collaboration through mobile technology.

Research in the computer supported collaborative learning (CSCL) field is currently tackling three important problems which are closely related: difficult integration of CSCL tools, scarce software reuse and technification. This work presents a Grid Collaborative Learning Environment (GRIDCOLE) that combines IMS learning design (IMS-LD) and Open Grid Services Architecture (OGSA) technologies in order to address these issues. More specifically, GRIDCOLE allows users with no technical skills easy integration and use of applications that effectively support collaborative learning processes. IMS-LD is employed to formally describe the design of these collaborative learning processes while applications are built using tools provided by third-party suppliers as OGSA-compliant Grid services. With this aim, the GRIDCOLE architecture and its main functionalities are introduced. A simple prototype is also presented in order to show the feasibility of the most important concepts introduced in this paper. Furthermore, a real collaborative learning scenario in which we plan to validate GRIDCOLE is introduced.

This paper describes the integration of the Grid topic within a computer architecture engineering course. Students are engaged in a project of design and evaluation of computing systems for a specific customer/market sector. The contents of the course and the case studies have been adapted in order to motivate the study of the Grid. In particular two case studies have been analyzed: one concerns an Internet-based music distribution enterprise, while the other is related to the assembly of genome sequences. A preliminary analysis of this experience shows that most of the students chose to evaluate the suitability of the Grid for their case study, and were able to argue on the benefits and drawbacks of this and other alternatives. In the future, we plan to set up a Grid-based learning environment so the Grid is not only the object of study, but also a tool for learning.

Collaborative learning patterns (CLPs) are detailed descriptions of best practices in collaborative learning. These patterns provide a way for a representation of key aspects of CSCL (computer-supported collaborative learning) that is easy to understand by software developers. To formalize these CLPs we have focused our attention on IMS learning design (IMS-LD). IMS-LD provides a means of expressing many different pedagogical approaches (including collaborative learning), however we have found some limitations in reflecting learning experiences that are group-based. Although this specification supports multiple roles in a learning activity, it is not possible to specify how they are going to interact. This paper points out this deficiency and proposes an extension of IMS-LD. The process that can be followed in order to obtain a unit of learning based on a CLP is illustrated with an example.

A new architecture called muARTMAP is proposed to impact a category proliferation problem present in Fuzzy ARTMAP. Under a probabilistic setting, it seeks a partition of the input space that optimizes the mutual information with the output space, but allowing some training error, thus avoiding overfitting. It implements an inter-ART reset mechanism that permits handling exceptions correctly, thus using few categories, especially in high dimensionality problems. It compares favorably to Fuzzy ARTMAP and Boosted ARTMAP in several synthetic benchmarks, being more robust to noise than Fuzzy ARTMAP and degrading less as dimensionality increases. Evaluated on a real-world task, the recognition of handwritten characters, it performs comparably to Fuzzy ARTMAP, while generating a much more compact rule set.

A new architecture, called MicroARTMAP, is proposed to impact the category proliferation problem present in Fuzzy ARTMAP. It handles probabilistic information through the optimization of the mutual information between the input and output spaces, but allowing a small training error, thus avoiding overfitting. While reducing the number of categories used by Fuzzy ARTMAP, it holds several desirable properties, such as a correct treatment of exceptions and a fast algorithm, as opposed to other approaches like BARTMAP. In addition, it is shown that MicroARTMAP is less sensitive than Fuzzy ARTMAP with respect to the the pattern presentation order, and that it degrades less if the training set is noisy.

Proposes an approach to the problem of logical labeling in structured documents, employing a new recurrent neuro-fuzzy system called RFasArt (Recurrent Fuzzy Adaptive System, ART-based). RFasArt preserves the fine characteristics, such as modularity, stability and flexibility, of its predecessors Fuzzy ARTMAP and FasArt. In this paper, the documents are considered as pseudo-temporal sequences, and context information is exploited in an integrated form. Two working prototypes for a MIME-based mailing system and for a digital library were tested with over 90% of the recognition rate and less ambiguous decisions than in the previous systems. A manageable knowledge base was constructed using fuzzy rules that were easily interpretable by human users, and examples of rule creation and fusion are shown.

Logical labeling, i.e. matching of the physical and logical structure, is an essential part of any document processing system. Contrary to the classical rule-based approach, where inflexible, heuristic knowledge is involved, a new neuro-fuzzy system is proposed in this paper. The main module, called FasArt (fuzzy adaptive system ART-based), is based on the well-known supervised neural architecture fuzzy ARTMAP. This new proposed architecture maintains the positive characteristics of its predecessor, such as capacity for incremental learning, respect to the plasticity-stability dilemma and generation of a symbolic representation of the system performance. Additionally, FasArt has a consistent formulation as a fuzzy logic system, a proven ability for approximation of any continuous function and a parameter that permits us to regulate its fuzziness degree. This new architecture is combined with a module that implements the Viterbi algorithm with the statistical properties of a certain document class, in order to resolve ambiguities or inconsistencies of the classification results provided by FasArt. Experimental results are shown for a variety of business letters, within a system of automatic feeding of an institution mailing database FasArt shows better results than Fuzzy ARTMAP while the use of the Viterbi-based modules boosts the correct labeling results to a 95% from the initial 80%.

In modern document management systems it is difficult to include large vocabularies (more than 150,000 words long) to detect on-line errors. The main drawback lies in the manipulation of the great amounts of data. This difficulty becomes critical if the system incorporates character recognition modules. In this paper we propose a new technique that stems from a written text segmentation based on phonetical and etymological criteria. The procedure we use integrates dictionary n-gram techniques. It checks whether the given character sequence matches a sequence of pseudosyllables (using a dictionary) and simultaneously checks if the pairs of pseudosyllables is admissible (through n-gram techniques). The results obtained from the proposed method using lists of words of various sizes, as well as a corpus in Spanish, are better than the n-gram methods typically used in error detection. Furthermore, it requires less memory and processing time, as compared with dictionary look-up methods.