Search

National audience

International audience; Natural Language Processing (NLP) became a trending domain within recent years for many researches and companies due to its wide applicability and the new advances in technology. The aim of this paper is to introduce an updated version or our open-source NLP framework, ReaderBench (http://readerbench.com/), designed to support both students and tutors in multiple learning scenarios that encompass one or more of the following dimensions: Cohesion Network Analysis of discourse, textual complexity assessment, keywords’ extraction using Latent Semantic Analysis (LSA), Latent Dirichlet Allocation (LDA) and word2vec semantic models, as well as the analysis of online communities and discussions. The latest version of our ReaderBench framework (v4.1) includes: a) new features, Application Programing Interfaces (APIs) and visualizations (e.g., sociograms, analysis of interaction between participants inside a community), b) a new web interface written in Angular 6, and c) the integration of new technologies to increase performance (i.e., spaCy and AKKA), as well as modularity and ease of deployment (i.e., Artifactory and Maven modules). ReaderBench is a fully functional framework capable to enhance the quality of learning processes conducted in multiple languages (English, French, Romanian, Dutch, Spanish, and Italian), and covering both individual and collaborative assessments.

International audience; Natural Language Processing has massively evolved during the last years and many up-to-date applications integrate different speech tools in order to create an enhanced user experiences. For obtaining a seamless integration of existing speech recognition systems, there is a trending interest for developing and improving existing speech-to-text algorithms. The aim of this paper is to improve user interaction with the ReaderBench platform, by developing and integrating a speech recognition module designed so that young pupils can dictate their self-explanations to a given text. Afterwards, the ReaderBench framework is used to automatically evaluate the employed reading strategies based on the resulted speech transcriptions. A dataset containing 160 self-explanations from students ranging from 9 to 11 years old was analysed using both original transcripts, and the ones automatically generated by our custom speech recognition system. Multiple methods designed to perform speech recognition are also compared, while a new dedicated model was trained in order to improve the quality of the existing French model for speech recognition from CMUSphinx speech recognition system. Our revised model includes a pronunciation dictionary obtained after training a Long Short-Term Memory (LSTM) Grapheme-to-Phoneme neural network. The accuracy of our system is benchmarked in relation to the automated process of identifying reading strategies implemented in our ReaderBench framework, which is applied on both manual transcriptions and automated speech-to-text inputs. The obtained results argue for the adequacy of our method as the slight decrease in terms of identification accuracy is justifiable in contrast to the effort of manually transcribing each self-explanation.

International audience; A critical task for tutors is to provide learners with suitable reading materials in terms of difficulty. The challenge of this endeavor is increased by students' individual variability and the multiple levels in which complexity can vary, thus arguing for the necessity of automated systems to support teachers. This chapter describes ReaderBench, an open-source multi-dimensional and multilingual system that uses advanced Natural Language Processing techniques to assess textual complexity at multiple levels including surface-based, syntax, semantics and discourse structure. In contrast to other existing approaches, ReaderBench is centered on cohesion and makes extensive usage of two complementary models, i.e., Cohesion Network Analysis and the polyphonic model inspired from dialogism. The first model provides an in-depth view of discourse in terms of cohesive links, whereas the second one highlights interactions between points of view spanning throughout the discourse. In order to argue for its wide applicability and extensibility, two studies are introduced. The first study investigates the degree to which ReaderBench textual complexity indices differentiate between high and low cohesion texts. The ReaderBench indices led to a higher classification accuracy than those included in prior studies using Coh-Metrix and TAACO. In the second study, ReaderBench indices are used to predict the difficulty of a set of various texts. Although the high number of predictive indices (50 plus) accounted for less variance than previous studies, they make valuable contributions to our understanding of text due to their wide coverage.

National audience; Il est de plus en plus question, et la diffusion des MOOCs (Massive Online Open Courses) le montre, de promouvoir des ressources éducatives ouvertes pour aider l’apprentissage. Pour autant, dans beaucoup de cas, « libre » signifie seulement « gratuit », et peu importe si cette gratuité s’accompagne en réalité d’une privatisation des données privées (Morozov, 2014), rendue encore plus performante par l’utilisation de techniques de Learning Analytics (LA). Ainsi, le courant des LA est souvent perçu comme gênant l’ouverture plutôt que la promouvant.Notre argument est que la puissance de traitement des LA, plutôt que d’être utilisées à assujettir un peu plus les utilisateurs en les espionnant ou les contrôlant, devrait plutôt être utilisée pour promouvoir des ressources libres au sens de Hilton et al. (2010), c’est-à-dire, des ressources non seulement gratuites, mais aussi accompagnées des permissions suivantes : retenir, réutiliser, réviser, remixer et redistribuer. Nous reprendrons ces droits en les exemplifiant avec ReaderBench, un système multilingue libre et gratuit que nous développons.

International audience; Dialog plays a central role in learning processes, particularly in shared understanding and social knowledge construction. Based on the socio-cultural paradigm, the polyphonic model of dialog considers social knowledge-building as inter-animation of voices in an analogy with polyphonic music, where novelty, creativity and coherence are assured by an interplay of dissonances and consonances, evolving towards a harmonious whole. In this analogy, voices are participants' newly introduced concepts and points of view. Automated tools implementing the polyphonic model cope with the high volume and complexity of data by relying on Cohesion Network Analysis, and applying advanced Natural Language Processing techniques. Corresponding tools were successfully used for automated dialog analysis in a wide variety of educational settings, ranging from formal philosophical discussions in text-based discussion forums to informal online blog discussions. Although the presented methods are proven to reliably predict comprehension of contents and further participation in discussions, they are limited by the nature of text-based communication. Further development of the polyphonic model will be extended to gestures and body language, both in individual and group settings.

National audience; Vocabulary acquisition is considered one of the most important aspects of a language because it defines and constraints communication. In order to overcome this, students and teachers use various techniques meant to improve the learning process and facilitate word acquisition. Using conceptual networks that establish links to prior knowledge is an approach with proven good results. The game presented in this paper, Semantic Taboo, is a serious game that helps users enlarge their conceptual network. The game has two modes, one for teachers and one for students. In the teacher module, users have the option of adding playing cards. The theme of the card (the seed) is taken as input, while the list of taboo words is selected from the list of semantically related words provided by the ReaderBench framework. The student model has the same game principle as the traditional game, but the list of taboo words and the description of the seed help the learner acquire new words. The preliminary validation consisted of 15 users who were impressed by the game concept and its educational value.

International audience; Computer Supported Collaborative Learning (CSCL) environments have become a viable learning alternative from which valuable data can be extracted and used for advanced analyses centered on evaluating participants' involvement and their interactions. Such automated assessments are implemented within our ReaderBench framework, a Natural Language Processing platform that contains multiple advanced text analysis functionalities. The ReaderBench framework is based on Cohesion Network Analysis from which different sociograms, relying on semantic similarity, are generated in order to reflect interactions between participants. In this paper, we briefly describe the enforced mechanisms used to analyze three scenarios: individual chat conversations, virtual communities of practice and MOOCs.

International audience; In this paper we introduce the online version of our ReaderBench framework, which includes multi-lingual comprehension-centered web services designed to address a wide range of individual and collaborative learning scenarios, as follows. First, students can be engaged in reading a course material, then eliciting their understanding of it; the reading strategies component provides an in-depth perspective of comprehension processes. Second, students can write an essay or a summary; the automated essay grading component provides them access to more than 200 textual complexity indices covering lexical, syntax, semantics and discourse structure measurements. Third, students can start discussing in a chat or a forum; the Computer Supported Collaborative Learning (CSCL) component provides in- depth conversation analysis in terms of evaluating each member’s involvement in the CSCL environments. Eventually, the sentiment analysis, as well as the semantic models and topic mining components enable a clearer perspective in terms of learner’s points of view and of underlying interests.

National audience; Nous décrivons un prototype de Manuel ouvert, online et massif (MOOT) auquel on peut intégrer diverses fonctionnalités utilisant des tech- niques de Traitement automatique de la langue, afin de promouvoir l’apprentissage auto-régulé. Nous discutons également de l’intérêt de disposer de ressources largement ouvertes, accessibles et flexibles dans l’enseignement supérieur.

International audience; ReaderBench is a fully functional automated software framework, designed to be an educational helper for students, teachers, and tutors. The system makes use of text-mining techniques based on advanced natural language processing and machine learning algorithms to design and deliver summative and formative assessments using multiple data sets (e.g., textual materials, behavior tracks, self-explanations).

National audience; Les recherches en sociolinguistique ont montré que le type de langage produit par les individus dépend, entre autres facteurs, du contexte de l’échange. Cet effet est constaté à un niveau macroscopique lorsqu’on distingue de grandes catégories de contextes comme les situations formelles (consultation médicale, entretien d’embauche) et les situations familières (repas en famille, discussion entre amis). Cet effet persiste également au niveau de micro-contextes caractérisant les moments successifs d’une même conversation en fonction de paramètres tels que le thème de l’échange en cours (Patterson, 1992).Peut-on penser qu’on s’adapte également, lorsqu’on évoque un texte précédemment lu, à sa thématique ? Y-a-t-il un lien entre les productions orales formelles d’élèves de cycle 3 et leur compréhension en lecture ? En effet, la compréhension en lecture passant par la compréhension du langage écrit (qui est un langage formel, ou en tout cas différent du langage oral dans son usage quotidien), nous avons souhaité observer si le niveau de compréhension en lecture et l’habileté à utiliser un langage formel lors de la restitution de textes lus étaient liés.80 élèves de cycle 3 (25 CE2, 28 CM1 et 27 CM2) âgés de 8 à 11 ans ont passé un test de compréhension obtenu à partir de la lecture silencieuse de deux textes suivis de questions présentées par écrit. De plus, ils devaient rappeler oralement 2 autres textes narratifs : une histoire fantastique et une histoire de la vie quotidienne. Après avoir lu les textes à voix haute, il était demandé aux élèves de les rappeler oralement. L’histoire fantastique racontait l’histoire d’une jeune Dieu qui, pour empêcher la sécheresse sur la Terre, a dû aller combattre dans le ciel un serpent qui buvait toute l’eau des nuages. L’histoire de la vie quotidienne racontait l’histoire d’une famille qui, pendant le repas, entendait des bruits bizarres dans la maison. Ces épreuves ont été réalisées dans le cadre du projet ANR DEVCOMP.Ces deux textes ont été analysés de manière automatique avec le logiciel Readerbench (voir l'article "ReaderBench : un outil pour évaluer la complexité de textes et identifier les stratégies de lecture") qui permet notamment une analyse multidimensionnelle de la complexité textuelle à partir de plus d’une centaine d’indicateurs de différents niveaux : analyse des facteurs de surface, des aspects lexicaux, morphologiques syntaxiques et sémantiques. Pour déterminer la complexité des textes, ReaderBench s’appuie sur un corpus d’entraînement composé d’ouvrages de littérature jeunesse et de manuels scolaires des différents niveaux de l’école élémentaire. Les deux textes sont de complexité équivalente et correspondent à un niveau CE2 selon le logiciel.Afin de déterminer la formalité des rappels oraux, nous avons analysé deux indicateurs :- une variable sociolinguistique : la réalisation optionnelle du /l/ dans le pronom clitique "il(s)" puisque des travaux ont montré que les adultes et les enfants le réalisent davantage en situation formelle (Armstrong, 1996 ; Chabanal, 2003). À partir du comptage du nombre d’occurrences de /l/ réalisées et non réalisées, nous avons calculé un pourcentage d’usage de la variante formelle, c’est-à-dire lorsque le segment consonantique est réalisé.- un indicateur linguistique : le nombre moyen de types, indicateur de complexité discursive, qui évalue le nombre moyen de mots différents produits. Cette analyse a été effectuée automatiquement avec ReaderBench.

We present a system for entity identification in free text. Each entity extracted will be identified as an entity from the system’s knowledge base – an ontology. We propose an algorithm that detects in a single pass the most probable related entity assignations instead of individually checking every possible entity combination. This unsupervised system employs graph algorithms applied on a graph extracted from the ontology as well as implementing a entity path scoring function to provide the most probable related entity assignations. We present the system’s implementation and results, as well as its strong and weak-points.

Each domain, along with its knowledge base, changes over time and every timeframe is centered on specific topics that emerge from different ongoing research projects. As searching for relevant resources is a time-consuming process, the automatic extraction of the most important and relevant articles from a domain becomes essential in supporting researchers in their day-today activities. The proposed analysis extends other previous researches focused on extracting co-citations between the papers, with the purpose of comparing their overall importance within the domain from a semantic perspective. Our method focuses on the semantic analysis of paper abstracts by using Natural Language Processing (NLP) techniques such as Latent Semantic Analysis, Latent Dirichlet Allocation or specific ontology distances, i.e., WordNet. Moreover, the defined mechanisms are enforced on two different subdomains from the corpora generated around the keywords " e-learning " and " computer ". Graph visual representations are used to highlight the keywords of each subdomain, links among concepts and between articles, as well as specific document similarity views, or scores reflecting the keyword-abstract overlaps. In the end, conclusions and future improvements are presented, emphasizing nevertheless the key elements of our research support framework.

International audience; As different learning methods and educational scenarios highly influence the corresponding outcomes, our aim is to highlight quantifiable discrepancies in terms of the complexity gap between presentations and handouts versus full documents (i.e. academic papers), expressed as concrete factors that directly influence the perceived difficulty. Although there are multiple dependant variables that affect the interpretation of a given topic (e.g., order of presented materials, difference in personal styles if materials originate from multiple authors), we limit the scope of our analysis to solely identifying textual traits that can be automatically extracted from conference papers and their corresponding slide presentations. Our approach represents the starting point for adapting MOOC (Massive Open Online Courses) materials to their target audience in terms of: textual complexity, learner comprehension and content reusability. Therefore, this study performs a detailed comparison using a wide variety of textual complexity metrics as background, ranging from surface, syntactic, morphological and semantic factors in order to grasp the specificities of each material. In other words, our goal consists of providing a set of required metrics for adapting learning materials in order to best suit the underlying educational activities. Preliminary results reflect a strong correlation between the two alternative presentation forms of the same material (papers and corresponding slides) and a similar degree of perceived textual complexity, emphasizing the strong and unitary writing characteristics of the author.

Session: Educational Data Mining; International audience; ReaderBench is a multi-purpose, multi-lingual and flexible environment that enables the assessment of a wide range of learners' productions and their manipulation by the teacher. ReaderBench allows the assessment of three main textual features: cohesion-based assessment, reading strategies identification and textual complexity evaluation, which have been subject to empirical validations. ReaderBench covers a complete cycle, from the initial complexity assessment of reading materials, the assignment of texts to learners, the capture of metacognitions reflected in one's textual verbalizations and comprehension evaluation, therefore fostering learner's self-regulation process.

International audience; With the increasing spread of the social web, identifying emotions in texts has proved to have various applications in fields like opinion mining or market analysis. Emotion recognition from written statements does not only reveal information about the person who wrote them, but can also be used in predicting how the emotional state of the readers can be affected. We propose a novel automatic method for analyzing texts that predicts how reading a news article can influence in turn the emotional state of the reader. This method integrates several word-count approaches and natural language processing techniques, such as Latent Semantic Analysis. Moreover, our implemented system contains a module designed to personalize the provided feedback according to the reader's current emotional state. A preliminary validation has been performed and results are promising.

National audience

National audience; Rezumat. Una din sarcinile dificile din prelucrarea limbajului natural ţine de evaluarea complexităţii sau a dificultăţii înţelegerii a textelor. Astfel, de un interes aparte sunt instrumentele de analiză automată a textelor care permit realizarea unei predicţii a complexităţii textelor pornind de la factori lexicali, sintactici, morfologici sau chiar semantici în funcţie de specificitatea abordării. Totodată, elementele de complexitate sunt dependente de context şi de domeniul de aplicare. Astfel, o analiză pur tehnică care se bazează doar pe metrici fără să ia în considerare constrângerile de psihologie, modele umane, vârstă şi motivaţia este insuficientă pentru a realiza o predicţie adecvată. În plus, alte aspecte ale analizei complexităţii sunt corelate cu etapele de achiziţie (împreună cu acurateţea şi fluenţa), corelate cu adaptarea mesajului comunicat audienţei din prisma corectitudinii, coerenţei şi adaptabilităţii la nivelul acesteia. Adiţional, metricile de complexitate textuală reprezintă indicatori importanţi de înţelegere şi coerenţă pentru texte comune întâlnite uzual în Internet, lucrări publicate şi cărţi.

Self-explanations elicited during reading capture the reading strategies used on-line by readers. In this paper, we first describe the results obtained with 79 children from grade 3 to 5 that self-explained during reading a narrative text. Afterwards, we present ReaderBench, a system that automatically assesses the reading strategies. Correlations between human coders and ReaderBench show the system behaves satisfactorily., La technique des auto-explications permet de saisir les stratégies de compréhension utilisées pendant la lecture. Nous détaillons ici les résultats obtenus avec cette technique auprès de 79 élèves de Cycle 3. Nous présentons ensuite ReaderBench, un système qui diagnostique le type de stratégies mises en œuvre. Une validation à partir des corrélations entre le codage des stratégies par des juges humains et ReaderBench montre que le système évalue les auto- explications de manière satisfaisante.

National audience; Pensum et ReaderBench, deux environnements personnels d'appren- tissage dont le but est de favoriser l'apprentissage auto-régulé, sont tour à tour présentés. Le premier évalue automatiquement, dans les résumés ou synthèses de cours produites, les sauts de cohérence, les hors-sujets, et la pertinence des phrases. Le deuxième évalue la complexité textuelle des textes soumis à la lec- ture et les stratégies de lecture à partir des auto-explications des élèves. Leur ar- chitecture et les méthodes de traitement automatique de la langue utilisées sont décrites et discutées.

International audience; We present a method devised to automatically analyze pupils' verbalizations during reading and to reveal some of the strategies they use. An experiment with 44 primary school pupils (3rd and 5th grade) reading a narrative text and verbalizing what they understood at predefined breaks showed, firstly, that machine results are correlated with experts'; secondly, that the recall of recent sentences is not uniform across verbalizations and subject to a grade effect; thirdly, that there is a grade effect in the recall of distal causal sentences.

no abstract

This report presents Version 1.5 of the Learning support and feedback services (delivering recommendations based on interaction analysis and on students’ textual production) that can be integrated within an e-learning environment.

no abstract

This deliverable describes the objectives, approach, planning and results of the second pilot round, and updates the validation design for round 3. The objectives of the second pilot round were directed towards enhancing the quality of the software services in the next round and beginning to measure the success of the individual services. Validation activities took place at eight pilot sites with stakeholders (learners, tutors, teaching managers), preceded by technical verification. A small-scale study on transferability of the software took place within the LTfLL team. Results demonstrated that the LTfLL software provides effective solutions to pedagogic problems of feedback and knowledge acquisition, subject to enhancements identified during validation. There is strong evidence for integrating the feedback services with a knowledge acquisition service. Software elements for replacement or further development were identified. The transferability study showed that language technologies appear to be most appropriate for problem-based and self-regulated learning. A number of barriers to adoption were identified. In conclusion, the pilots have met their objectives for the second pilot round.The goals of the final pilot round are to support exploitation of the services through providing evidence of the utility of the services and factors affecting their adoption, and to inform the end of project road map, through extended full pilots and validation associated with dissemination activities.

no abstract

no abstract

This deliverable describes the objectives, approach, planning and results of the third pilot round, in which both individual and threaded services underwent validation. The two goals of this round were to provide input to the LTfLL exploitation plan and roadmap (deliverable 2.5). 531 participants (316 learners) took part in the pilots, which used LTfLL services based on five different languages. The average timespan of the pilots was three weeks and involved learners, tutors, teaching managers, the LTfLL team and Technology Enhanced Learning experts. The validation approach was based on Prototypical Validation Topics derived from the Round 2 validation topics, which refocused the validation topics on exploitation and allowed conclusions to be drawn across all services. Results demonstrated the areas of strength and weakness of each service, informing the selling points and barriers to adoption within the exploitation strategy, as well as suggesting possible further contexts of use. All services were noted to have high relevance in addressing burning issues for organizations, but further improvements to accuracy from a user viewpoint are required. Results on future enhancements to improve likelihood of adoption contribute to the roadmap. Results also provide an indication of each service's current readiness for adoption and provided insights into transferability issues. The overall conclusion is that some LTfLL services are more ready than others for adoption now, with some being currently more suited to sustainability in research settings.

This deliverable reports on a vision of the future for the LTLL tools and language technology for e-learning generally. The LTfLL tools and infrastructure are presented in terms of three layers (subsections), which are ultimately based on the final validation results. Layer one describes achievements and their importance. Layer two describes requiremets for adoption and sustain ability. And layer three describes logical extensions which could be made for any future versions of the tools. The report ends with a discussion of the future of language technology-based e-learning tools, with the guiding theme being personalisable tools for social learning.

This deliverable describes the objectives, approach, planning and results of the first pilot round, and updates the validation design for round 2. The objectives of the first pilot round were directed towards enhancing the quality of the scenarios and services in the next round, through validation activities with stakeholders and technical verification. Results demonstrated that showcase scenarios described realistic solutions to real stakeholder problems, subject to enhancements identified during validation. Software elements for replacement or further development were identified. In conclusion, the pilots indicate that the utility of advanced learning technologies can be generalized across different stakeholder groups and pedagogic problems, in the areas of positioning, feedback and knowledge sharing.In the second round, the validation objectives are extended to provide an early evaluation of the project using primarily quantitative methods, while continuing formative stakeholder evaluation and requirements gathering activities.

The purpose of this workshop was to focus on using language technologies in support of learning and teaching - by trying to identify what questions and problems are solved, but also to raise and discuss how well the metrics and algorithms developed assist in the provision of support and the construction of feedback for learning. What are the most efficient ways? To what extent do they match distances inferred by a teacher's assessments? The workshop addresses the problem of how support can be provided and feedback be generated in order to help students learn and teachers to assess their progress.

International audience; The purpose of this paper is to present and discuss a Natural Language Processing-based operationalization of Bakhtin's dialogism ideas, and to insert it in the context of distance learning. Three core concepts from Bakhin are introduced and their application in the domain of distance learning discussed. Then, two computer-based systems based on these ideas, PolyCAFe and Pensum are presented.

Topic modeling is a growing research field and novel ways of interpreting and evaluating results are necessary. We propose a method for evaluating and improving the performance of topic models generating algorithms relying on WordNet data. We first propose a measure for determining a topic model's fitness factoring in its broadness and redundancy. Then, for each individual topic, the amount of relevant information it provides, along with its most important words and related concepts are determined by defining a cohesion function based on the topic's projection on WordNet concepts. The model as a whole is improved by eliminating each topic's outliers with respect to the ontology projection. We define a inter topic ontology based distance and we further use it to investigate the impact of removing redundant topics from a model with regard to the overlap between topics' ontological projections. Clustering similar topics into conceptually cohesive groups is tried as an alternative to pruning less relevant topics. Results show that evaluating and improving statistical models with WordNet is a promising research track that leads to more coherent topic models.

This report presents Version 1 of the support and feedback services (delivering recommendations based on interaction analysis and on students’ textual production) that can be integrated within an e-learning environment. Further steps toward the implementation of Version 2 of these services and their future integration with all the LTfLL services are also suggested.

International audience; With the continuous evolution of collaborative environments, the needs of automatic analyses and assessment of participants in instant messenger conferences (chat) have become essential. For these aims, on one hand, a series of factors based on natural language processing (including lexical analysis and Latent Semantic Analysis) and data-mining have been taken into consideration. On the other hand, in order to thoroughly assess participants, measures as Page's essay grading, readability and social networks analysis metrics were computed. The weights of each factor in the overall grading system are optimized using a genetic algorithm whose entries are provided by a perceptron in order to ensure numerical stability. A gold standard has been used for evaluating the system's performance.

International audience; In AIED research, providing feedback for learning entails measuring differences among learners; between learners and their desired characteristics (e.g., knowledge, competences, motivation, self-regulation processes); or between learners and their looked-for resources (e.g., web-links, articles, courses) has often been performed by computing and analysing ‘distances’ using several techniques like factorial analysis, instance-based learning, clustering, and so on. Corpora on which these measures are made are all writing-based, that is, are multiple forms of pieces of evidence such as texts read (written by teachers), spoken utterances, essays, summaries, forum or chat messages. Some of these metrics are based on shallow syntactical and morphological aspects of the interaction and production artefacts (e.g., text length). Others are focused more on semantic and pragmatic aspects. These measures are used for providing various kinds of feedback for supporting learning and connections between learners. For instance, relations between learners’ utterances, knowledge, concept acquisition, emotional states, essay scores, and even learners themselves have all been investigated with the help of computing semantic distances.The purpose of this workshop is to focus on the latter two – semantics and pragmatics – by trying to identify what questions and problems are solved, but also to raise and discuss how well the metrics developed assist in the provision of support and the construction of feedback for learning. What are the most efficient ones? To what extent do they match distances inferred by teachers’ assessments?Presentations on topics like the following ones will fuel the research on NLP in support of learning: automated essay scoring and grading, summarization and writing assistance, methodological issues of distance-based semantic processing techniques, cognitive modelling using distance-based semantic processing techniques, analysis, assessment, and feedback generation of content and inter-animation in CSCL through chats or forums.

International audience; In AIED research, providing feedback for learning entails measuring differences among learners; between learners and their desired characteristics (e.g., knowledge, competences, motivation, self-regulation processes); or between learners and their looked-for resources (e.g., web-links, articles, courses) has often been performed by computing and analysing ‘distances’ using several techniques like factorial analysis, instance-based learning, clustering, and so on. Corpora on which these measures are made are all writing-based, that is, are multiple forms of pieces of evidence such as texts read (written by teachers), spoken utterances, essays, summaries, forum or chat messages. Some of these metrics are based on shallow syntactical and morphological aspects of the interaction and production artefacts (e.g., text length). Others are focused more on semantic and pragmatic aspects. These measures are used for providing various kinds of feedback for supporting learning and connections between learners. For instance, relations between learners’ utterances, knowledge, concept acquisition, emotional states, essay scores, and even learners themselves have all been investigated with the help of computing semantic distances.The purpose of this workshop is to focus on the latter two – semantics and pragmatics – by trying to identify what questions and problems are solved, but also to raise and discuss how well the metrics developed assist in the provision of support and the construction of feedback for learning. What are the most efficient ones? To what extent do they match distances inferred by teachers’ assessments?Presentations on topics like the following ones will fuel the research on NLP in support of learning: automated essay scoring and grading, summarization and writing assistance, methodological issues of distance-based semantic processing techniques, cognitive modelling using distance-based semantic processing techniques, analysis, assessment, and feedback generation of content and inter-animation in CSCL through chats or forums.

Cet article présente un environnement-tuteur d'apprentissage pouvant s'insérer dans une plate-forme de FOAD permettant à des étudiants de prendre connaissance et de réviser des contenus de cours. Cet environnement puise son originalité dans le fait qu'il allie une approche cognitive et des références à l'apprentissage dialogique pour permettre : — de fournir aux apprenants les cours les plus pertinents à étudier étant donné leur compréhension du cours ; — de délivrer des feedback sur la compréhension d'un cours et le contenu de résumés produits sur ce même cours ; — de renvoyer des feedback sur les fils de discussions menées dans des chats.

This deliverable describes the framework, approach and early planning for validation of the software services and evaluation of the LTfLL methodology, project and work approach.

This report is about the way to deliver relevant feedback to students’ written production, either for free texts (e.g., essays, syntheses, notes) or chat conversation, in order for the students to build knowledge. This report presents an overview and a selection of existing models, methods and resources for: 1) the automatic analysis of learner interactions using language technologies or social network analysis (Task 5.1) and 2) the automatic analysis of learner text (Task 5.2). Secondly , a proposition of the tools to be developed for the project is presented, with their possible scenarios of use.