We introduce, TermSuite, a JAVA and UIMA-based toolkit to build terminologies from corpora. TermSuite follows the classic two steps of terminology extraction tools, the identification of term candidates and their ranking, but implements new features. It is multilingually designed, scalable, and handles term variants. We focus on the main components: UIMA Tokens Regex for defining term and variant patterns over word annotations, and the grouping component for clustering terms and variants that works both at morphological and syntactic levels.
Absrtact: Discovering temporal patterns hidden in a sequence of events has applications in numerous areas like network failure analysis, customer behaviour analysis, web navigation pattern discovery, etc. In this article, we present an approach to the discovery of chronicles hidden in the interaction traces of a human activity with the intention of characterizing some interesting tasks. Chronicles are a special type of temporal patterns, where temporal orders of events are quantified with numerical bounds. The algorithm we present is the first existing chronicle discovery algorithm that is complete. It is a chronicle discovery framework that can be configured to behave exactly as non-complete algorithms existing in litterature with no reduction of performance, but it can also be extended to other useful chronicle discovery problems like hybrid episode discovery. We show that the complete chronicle discovery problem has a very high complexity but we argue and illustrate that this high complexity is acceptable when the knowledge discovery process in which our algorithm takes part is real time and interactive. The platform Scheme Emerger, also presented in this paper, has been developed in order to implement the algorithm and to support graphically the real time and interactive chronicle discovery process.
Cet article présente un outil de visualisation interactive de traces d'interactions dans le cadre d'une activité d'apprentissage collaboratif synchrone. Cet outil a été développé en collaboration entre l'entreprise eLycée S.A.S., et une équipe de recherche travaillant sur l'ingénierie de l'expérience tracée et les EIAH. L'hypothèse de facilitation de la tâche d'apprentissage par les processus métacognitifs liés à une activité réflexive est à la base de la contribution. L'article est l'occasion de situer précisément les enjeux du travail engagé, de décrire l'environnement et les outils développés, et de présenter les propriétés des modèles sous-jacents. Bien que cet outil de visualisation de traces n'ait pas encore fait l'objet d'expérimentations, les tests techniques auprès d'un public varié ont rencontré une forte adhésion. L'article pointe les aspects génériques des mécanismes de traçage développés, en particulier les possibilités de faire évoluer dynamiquement l'environnement par l'utilisateur mais aussi par les concepteurs d'activités et les enseignants. MOTS CLÉS : Traces d'utilisation ou d'interactions, plate-forme collaborative synchrone, apprentissage à distance, visualisation interactive de traces, réflexivité. ABSTRACT : This paper presents an interactive visualization tool for interaction traces, in the context of a synchronous collaborative e-learning activity. This tool has been developed by an e-learning company, eLycée, in collaboration with a research team working on e-learning, interaction traces, and experience reuse. This work adopts the grounding hypothesis that metacognitive processes and reflexive activities can facilitate learning. The paper presents the specific issues of such usage trace construction and visualization, the design of the hosting collaborative synchronous platform, and the associated tools and underlying models properties. This tool has so far not been experimented with students, but the technical tests with various researchers have been very promising and encountered a wide acceptance. The paper also points out the generic aspects of the tracing mechanisms, and the possibility for the student, teacher and designer to configure, update and extend it dynamically.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.