As shown by several studies, successful integration of technology in language learning requires a holistic approach in order to scientifically understand what learners do when working with web-based technology (cf. Raby, 2007). Additionally, a growing body of research in computer assisted language learning (CALL) evaluation, design and development, has indicated that analysis of learners’ behaviours is an essential element to implementing high-quality technology (e.g., Chapelle, 2001; Levy & Stockwell, 2006). Hence, carefully evaluating the effectiveness of CALL by collecting empirical data on user interactions while focusing on the process of learning is integral to a holistic understanding of students’ behaviours (e.g. Felix, 2005; Hémard, 2006).

This article examines a design-based research that seeks to analyse and understand the dynamics of user interactions with a specific web-based CALL tool in the context of a French as a second language (FSL) course. To this end, we present a sample of results based on an analysis of specific tasks carried out with this CALL tool that is designed in part to encourage students’ integration of critical and electronic literacies. By way of conclusion, we identify the steps that are necessary to enhance this particular CALL system and help users better achieve their learning goals. In particular, we explain the process of recycling our results in the next design phase of the CALL tool in a continuous improvement effort.

Organizing and implementing telecollaboration projects in foreign language curricula is not an easy endeavour (Belz & Thorne, 2006; Guth & Helm, 2010), as pedagogical, organizational and technical issues have to be addressed before cross-cultural interaction sessions can be carried out (O'Dowd & Ritter, 2006; O'Dowd, 2011). These issues make many teaching practitioners reluctant to try to integrate telecollaboration in their teaching, as they are more aware of the burden such initiatives might impose than of the benefits they might have for language learners.

Within the European project NIFLAR1 we have tried to study the added value that integrating synchronous collaboration projects through video-web communication or Second Life might have in language learning. The study presented in this paper measures the oral communicative growth of language students, who were allocated at random to one of three research conditions: (1) the VC experimental group carried out interactions with native peers through video-web communication; (2) the SL experimental group carried out the same tasks with native peers in Second Life and (3) the control group performed the tasks face to face with classroom peers and had no opportunity to interact with native experts. Communicative growth was measured by comparing oral pre- and post-tests across conditions. Results show significant differences, the experimental groups outperforming the control group.

Desktop videoconferencing (DVC) offers many opportunities for language learning through its multimodal features. However, it also brings some challenges such as gaze and mutual gaze, that is, eye-contact. This paper reports some of the findings of a PhD study investigating social presence in DVC interactions of English as a Foreign Language (EFL) teacher trainees. The case study approach involved the exploration of online interactions of five cases (pairs) within an interpretivist paradigm. Data collection included interviews, questionnaires and analysis of DVC recordings. The study emphasizes the importance of eye-contact in online multimodal communication to facilitate the establishment of social presence. Five types of gaze that were observed in learner interactions and participants’ perspectives on eye-contact are reported. The conclusions include technical suggestions for the use of a webcam as well as pedagogical implications of online video interaction.

The combined motion of the human thumb, index and middle fingers while rotating a small object across the extended, intermediate and flexed planes with respect to the fingers was analyzed. Auto reflective markers were attached on the fingers to track their motion across three postures and planes via a 3D motion capture system. Central, right and left rotation postures were considered in each plane for investigation and the rotation experiments were performed with 30 healthy subjects. The obtained data were used to compute the finger joint angles. Based on the three criteria of (i) manipulability measure, (ii) major axis direction angle of the manipulability ellipsoid and (iii) ratio of the minor over major axis lengths, the collective behavior of the fingers was studied. It has been found after analysis that the thumb and middle finger were active, while the index finger operated passively when manipulating small objects in cooperative rotational motion across the three planes. Activeness refers to the independence of a digit in controlling the motion of an object whereas passiveness denotes its dependence on other digits. An active finger governs the motion of an object whereas a passive finger simply supports it. The results of this investigation are of great importance in planning treatment for rehabilitation and for designing controllers for robotic therapists, finger exoskeletons and prostheses.

In this paper, the force control of a constrained one-link flexible arm is investigated using a feedback parallel compensation algorithm based on a linear distributed parameter model with internal damping of Kelvin–Voigt type. Generally, the non-collocation of the joint torque input and the tip contact force output comes along with the non-minimum phase in nature. To overcome this inherent limitation, a new input induced by the measurement of root-bending moment and its derivative, and a virtual contact force output generated by a parallel compensator are defined. Therefore, the transfer function from the new input to the virtual contact force output is proved not only strictly minimum phase but also in a stable condition. A PD controller then improves the performance of the overall closed-loop system. Furthermore, the perfect asymptotic tracking of a desired contact force trajectory with internal stability can be achieved accurately. The exact solutions of the infinite-dimensional system are obtained using the infinite product formulation. The proposed system promises stability robustness to parameter uncertainties, also free of spillover problems. Numerical simulations are provided to verify the effectiveness of the proposed approach.

This paper describes the multimodal pedagogical communication of two groups of online teachers; trainee tutors (second year students of the Master of Arts in Teaching French as a Foreign Language at the University Lumière-Lyon 2) and experienced teachers based in different locations (France, Spain and Finland). They all taught French as a Foreign Language to a group of students from UC Berkeley in 2010. They participated in a project using a desktop videoconferencing platform (VISU1) designed for delivering online courses. The study focuses on the webcam's effects on teaching and learning and tries to answer the following question: how does multimodal interaction affect interactive learning? Our hypothesis is that experienced teachers channel information through the webcam more efficiently and effectively in order to engage learners in knowledge construction. This paper presents the results of research based on an empirical method of collecting ecological data.

This article discusses a two-year telecollaborative project in teacher education that took an integrated approach to teaching about and through technological resources in order to introduce student-teachers to innovative methods for communicative-based language learning through computer-mediated communication (CMC). Via ‘technological immersion’, student-teachers in two groups in Spain and the US were required to work together online to give peer feedback and evaluation of several activities, including teaching sequences. They also co-created podcasts, along with accompanying educational activities. Some of the tools used were Moodle, Skype, emails, wikis, Second Life and podcasting. The article analyzes and discusses multimodal data collected during the collaboration. Results indicate that the online collaboration enhanced teacher development through opportunities unavailable in more traditional teacher education classrooms and enabled student-teachers to better make connections between theory and practice.

Our study aims to optimize a multimedia application for vocabulary learning for English as a Foreign Language (EFL). Our study is based on the concept that difficulty in reading a text in a second language is due to the need for more working memory for word decoding skills, although the working memory must also be used for text comprehension skills. This implies that the automatization of word decoding to free the working memory is indispensable. Therefore, developing a multimedia application for vocabulary learning with a time-control function, our study hypothesizes that vocabulary learning with this function brings about the automatization of word decoding or faster recall of the meanings, and, as a result of freeing more working memory, better comprehension of text. According to our research, multimedia application users not only retained more words and recalled their meanings more quickly but also obtained higher scores on reading comprehension tests that contained the words learned through the application. From this result, we conclude that multimedia vocabulary learning with a time-control function can automatize word decoding skills and therefore free more working memory, thereby leading to better comprehension of the text.

This paper proposes a new visual control approach based on sinusoidal inputs to be used on a nonholonomic robot. We present several contributions: In our method, developed considering a unicycle kinematic model, sinusoids are used in such a way that the generated vehicle trajectories are feasible, smooth and versatile. Our technique improves previous sinusoidal-based control works in terms of efficiency and flexibility. As further contributions, we present analytical expressions for the evolution of the robot's state, and propose a new state-feedback control law based on these expressions. All the information used in the control scheme is obtained from omnidirectional vision by means of the one-dimensional trifocal tensor. Stability analysis of the proposed approach is presented, and its performance is illustrated through experiments.

Research into the online informal learning of English has already shown it to be a widespread phenomenon involving a range of comprehension and production activities such as viewing original version television series, listening to music on demand and social networking with other English users.

Dynamic systems theory provides a suitable framework within which to study informal learning because it emphasises the unique range of resources, strategies and relationships which contribute to non-linear language development for each learner.

While research into the impact of these activities for language development has yielded some data regarding vocabulary gains and improvements in fluency and comprehension skills, the mechanisms at work in such language development have proved difficult to study because of the private nature of most online informal learning. In this study, an emic approach is adopted, involving a group of students experienced in second language acquisition research, who used blogs to report on their own online informal learning of English over a three-month period. Extensive examples from these blogs are used to build up a picture of the learning processes at work, within a framework suggested by the literature of complex dynamic systems. These results allow conclusions to be drawn regarding the way in which these processes are used by informal learners in different phases of real communicative tasks.

This paper addresses the task of automatic extraction of definitions by thoroughly exploring an approach that solely relies on machine learning techniques, and by focusing on the issue of the imbalance of relevant datasets. We obtained a breakthrough in terms of the automatic extraction of definitions, by extensively and systematically experimenting with different sampling techniques and their combination, as well as a range of different types of classifiers. Performance consistently scored in the range of 0.95–0.99 of area under the receiver operating characteristics, with a notorious improvement between 17 and 22 percentage points regarding the baseline of 0.73–0.77, for datasets with different rates of imbalance. Thus, the present paper also represents a contribution to the seminal work in natural language processing that points toward the importance of exploring the research path of applying sampling techniques to mitigate the bias induced by highly imbalanced datasets, and thus greatly improving the performance of a large range of tools that rely on them.

We consider a dependency-parsed text corpus as an instance of a labeled directed graph, where nodes represent words and weighted directed edges represent the syntactic relations between them. We show that graph walks, combined with existing techniques of supervised learning that model local and global information about the graph walk process, can be used to derive a task-specific word similarity measure in this graph. We also propose and evaluate a new learning method in this framework, a path-constrained graph walk variant, in which the walk process is guided by high-level knowledge about meaningful edge sequences (paths) in the graph. Empirical evaluation on the tasks of named entity coordinate term extraction and general word synonym extraction show that this framework is preferable to, or competitive with, vector-based models when learning is applied, and using small to moderate size text corpora.

The majority of current robotic joints are primarily actuated by rotational mechanisms. These electrical drives have substantially different features than the features found in human muscular systems. This paper presents a cost-effective solution to the backlash of a phenomenon known to cause positioning errors and other undesirable dynamic effects in drives. These errors are particularly pronounced when relatively major changes appear in the pre-load conditions of the motor such as in the case of a robotic leg or arm with a high degree of freedom. Current solutions require an accurate time-varying model of drives that is not available in the majority of practical cases. Therefore, in this paper a digitally controlled mechanical solution is proposed which is inspired by the human flexor–extensor mechanism. The idea is to construct an antagonistic actuator pair analogous to the flexor and extensor muscles. In order to obtain good control performance even in the low-speed range, permanent magnet stepper motors were chosen as actuators that are commutated in a digitally closed-loop fashion. The operation of the controlled structure has been verified in a real experimental environment where measurements showed good results and match with previous simulations.

This paper presents a general approach to automatically compile e-learning models to planning, allowing us to easily generate plans, in the form of learning designs, by using existing domain-independent planners. The idea is to compile, first, a course defined in a standard e-learning language into a planning domain, and, second, a file containing students learning information into a planning problem. We provide a common compilation and extend it to three particular approaches that cover a full spectrum of planning paradigms, which increases the possibilities of using current planners: (i) hierarchical, (ii) including PDDL (Planning Domain Definition Language) actions with conditional effects and (iii) including PDDL durative actions. The learning designs are automatically generated from the plans and can be uploaded, and subsequently executed, by learning management platforms. We also provide an extensive analysis of the e-learning metadata specification required for planning, and the pros and cons on the knowledge engineering procedures used in each of the three compilations. Finally, we include some qualitative and quantitative experimentation of the compilations in several domain-independent planners to measure its scalability and applicability.

In most of our daily motion tasks, the coordination between limbs is very crucial for successful execution of the tasks. In this paper, coordination among oscillators controlling in Cartesian space is studied to control bipedal walking. In our method, phase adjustment among oscillators is considered as one of the key issues to achieve coordination. A new phase adjustment method is proposed. With this method, an oscillator is able to coordinate other oscillators and maintain a desired phase relationship. This property is important for the walking control especially when external perturbations are given. To simplify the relationship between oscillators in a central pattern generator (CPG), a hierarchical CPG structure is adopted, where a main oscillator will be used to adjust other oscillators. In the simulation, the walking motion controlled by the CPG controller converges to a stable pattern even with external perturbations. We have implemented the controller in both the simulation model and real hardware robot.

We consider the connectivity of autonomous mobile robots. The robots navigate using simple local steering rules without requiring explicit communication among themselves. We show that using only position information of neighbors, the group connectivity can be sustained even in the case of bounded position measurement errors and the occlusion of robots by other robots in the group. In implementing the proposed scheme, sub-optimal solutions are invoked to avoid an excessive computational burden. We also discuss the possibility of deadlock which may bring the group to a standstill and show that the proposed methodology avoids such a scenario in real-life settings.

This paper presents a distributed architecture for automating data mining (DM) processes using standard languages. DM is a difficult task that relies on an exploratory and analytic process of processing large quantities of data in order to discover meaningful patterns. The increasing heterogeneity and complexity of available data requires some expert knowledge on how to combine the multiple and alternative DM tasks to process the data. Here, we describe DM tasks in terms of Automated Planning, which allows us to automate the DM knowledge flow construction. The work is based on the use of standards that have been defined in both DM and automated-planning communities. Thus, we use PMML (Predictive Model Markup Language) to describe DM tasks. From the PMML, a problem description in PDDL (Planning Domain Definition Language) can be generated, so any current planning system can be used to generate a plan. This plan is, again, translated to a DM workflow description, Knowledge Flow for Machine Learning format (Knowledge Flow file for the WEKA (Waikato Environment for Knowledge Analysis) tool), so the plan or DM workflow can be executed in WEKA.