International audience

It is a challenging task to identify eLearning courses parts that have to be revised to best suit learners’ requirements. Reading being one of the most salient learning activities, one way of doing so is to study how learners consume courses. We intend to support course authors (e.g. teachers) during courses revision by providing them with reading indicators. We use the concept of reading session to denote a learner’s active reading period, and we provide several associated reading indicators. In our server-side approach, reading sessions and indicators are calculated using web server logs. We evaluate the relevance of our proposals using logs from a major French eLearning platform. Results are promising: calculated reading sessions are theoretically more precise than other best applicable approaches, and course authors consider suggested indicators to be appropriate to courses revision. Using reading sessions and associated indicators could facilitate authors’ work of course reengineering.

Adaptive and personalized behavior is becoming essential and desirable in Human-Robot Interactive systems. We are interested in adaptive robots that learn from interaction traces (previous interactions with users). Our proposal is based on types of interactions where users express their level of satisfaction through feedback. Indeed, depending on the situation of interaction and the user himself, the robot behavior should adjust, and therefore can be judged, differently. From interaction traces (including robot actions and users' feedback), we aim to extract adaptation rules that give the dependencies between certain attributes of the interaction situation and/or the user profile, and the level of user satisfaction. We propose two learning algorithms to learn these adaptation rules. The first algorithm is direct, certain and optimal but slow to converge. The second is able to detect the importance of certain attributes in the adaptation process. It generalizes adaptation rules on unknown situations and to first time users, which makes it an approach with risk. We detail in this paper, our proposed model, both learning algorithms, and an evaluation of the learned rules from both algorithms by simulations and through a scenario with real users.

This ongoing work investigates usage-based document reengineering as a means to support authors in modifying their documents. Document usages (i.e. usage feedbacks) cover readers' explicit annotations and their reading traces. We first describe a conceptual framework with various levels of assistance for document reengineering: indications on reading, problem detection, reconception suggestions and automatic reconception propositions, taking our example in e-learning document management. We then present a technical framework for usage-based document reengineering and its associated models for documents, annotations and traces representation.

Service robots have become increasingly important subjects in our lives. However, they are still facing problems like adaptability to their users. While major work has focused on intelligent service robots, the proposed approaches were mostly user independent. Our work is part of the FUI-RoboPopuli project, which concentrates on endowing entertainment companion robots with adaptive and social behaviour. In particular, we are interested in robots that are able to learn and plan so that they adapt and personalize their behaviour according to their users. Markov Decision Processes (MDPs) are largely used for adaptive robots applications. However, one challenging point is reducing the sample complexity required to learn an MDP model, including the reward function. In this article, we present our contribution regarding the representation and the learning of the reward function through analysing interaction traces (i.e. the interaction history between the robot and their users, including users' feedback). Our approach permits to generalise the learned rewards so that when new users are introduced, the robot may quickly adapt using what it learned from previous experiences with other users. We propose, in this article, two algorithms to learn the reward function. The first is direct and certain, the robot applies with a user what it learned during interaction with same kind of users (i.e. users with similar profiles). The second algorithm generalises what it learns to be applied to all kinds of users. Through simulation, we show that the generalised algorithm converges to an optimal reward function with less than half the samples needed by the direct algorithm.

Our approach to address the question of online video accessibility for people with sensory disabilities is based on video annotations that are rendered as video enrichments during the playing of the video. We present an exploratory work that focuses on video accessibility for blind people with audio enrichments composed of speech synthesis and earcons (i.e. nonverbal audio messages). Our main results are that earcons can be used together with speech synthesis to enhance understanding of videos; that earcons should be accompanied with explanations; and that a potential side effect of earcons is related to video rhythm perception.

Videos are commonly being augmented with additional content such as captions, images, audio, hyperlinks, etc., which are rendered while the video is being played. We call the result of this rendering "enriched videos". This article details an annotation-based approach for producing enriched videos: enrichment is mainly composed of textual annotations associated to temporal parts of the video that are rendered while playing it. The key notion of enriched video and associated concepts is first introduced and we second expose the models we have developed for annotating videos and for presenting annotations during the playing of the videos. Finally, an overview of a general workflow for producing/viewing enriched videos is presented. This workflow particularly illustrates the usage of the proposed models in order to improve the accessibility of videos for sensory disabled people.