This paper aims at showing the didactic and theoretical-based perspectives in the experimental development of the geogebraTUTOR system (GGBT) in interaction with the students. As a research and technological realization developed in a... more
This paper aims at showing the didactic and theoretical-based perspectives in the experimental development of the geogebraTUTOR system (GGBT) in interaction with the students. As a research and technological realization developed in a convergent way between mathematical education and computer science, GGBT is an intelligent tutorial system, which supports the student in the solving of complex problems at a high school level by assuring the management of discursive messages as well as the management of problem situations. By situating the learning model upstream and the diagnostic model downstream, GGBT proposes to act on the development of mathematical competencies by controlling the acquisition of knowledge in the interaction between the student and the milieu, which allows for the adaptation of the instructional design (learning opportunities) according to the instrumented actions of the student. The inferential and construction graphs, a structured bridge (interface) between the contextualized world of didactical contracts and the formal computer science models, structure GGBT. This way allows for the tutorial action to adjust itself to the competential habits conveyed by a certain classroom of students and to be enriched by the research results in mathematical education.
Research Interests:
ABSTRACT Automatic keyword extraction is an important subfield of information extraction process. It is a difficult task, where numerous different techniques and resources have been proposed. In this paper, we propose a generic approach... more
ABSTRACT Automatic keyword extraction is an important subfield of information extraction process. It is a difficult task, where numerous different techniques and resources have been proposed. In this paper, we propose a generic approach to extract keyword from documents using encyclopedic knowledge. Our two-step approach first relies on a classification step for identifying candidate keywords followed by a learning-to-rank method depending on a user-defined keyword profile to order the candidates. The novelty of our approach relies on i) the usage of the keyword profile ii) generic features derived from Wikipedia categories and not necessarily related to the document content. We evaluate our system on keyword datasets and corpora from standard evaluation campaign and show that our system improves the global process of keyword extraction.
Research Interests:
ABSTRACT Semantic annotation is the process of identifying expressions in texts and linking them to some semantic structure. In particular, Linked data-based Semantic Annotators are now becoming the new Holy Grail for meaning extraction... more
ABSTRACT Semantic annotation is the process of identifying expressions in texts and linking them to some semantic structure. In particular, Linked data-based Semantic Annotators are now becoming the new Holy Grail for meaning extraction from unstructured documents. This paper presents an evaluation of the main linked data-based annotators available with a focus on domain topics and named entities. In particular, we compare the ability of each tool to annotate relevant domain expressions in text. The paper also proposes a combination of annotators through voting methods and machine learning. Our results show that some linked-data annotators, especially Alchemy, can be considered as a useful resource for topic extraction. They also show that a substantial increase in recall can be achieved by combining the annotators with a weighted voting scheme. Finally, an interesting result is that by removing Alchemy from the combination, or by combining only the more precise annotators, we get a significant increase in precision, at the cost of a lower recall.
Research Interests:
Research Interests:
The Cortex system is constructed of several different sentence selection metrics and a decision module. Our experiments have shown that the Cortex decision on the metrics always scores better than each system alone. In the INEX@ QA 2010... more
The Cortex system is constructed of several different sentence selection metrics and a decision module. Our experiments have shown that the Cortex decision on the metrics always scores better than each system alone. In the INEX@ QA 2010 task of Long ...
ABSTRACT Wikimeta Lab participation in DeFT 2013 - Machine Learning for Information Extraction and Classification of Cooking Recipes. This paper presents Wikimeta Lab participation in the Défi Fouille de Texte (DeFT) 2013. In 2013, this... more
ABSTRACT Wikimeta Lab participation in DeFT 2013 - Machine Learning for Information Extraction and Classification of Cooking Recipes. This paper presents Wikimeta Lab participation in the Défi Fouille de Texte (DeFT) 2013. In 2013, this evaluation campaign is focused on mining cooking recipes in French. The campaign consists of three classification tasks and an information extraction task. The corpus is composed of recipes from a collaborative web site, thus they are written by users with almost no constraints on labels, ingredients, and comments. This very context makes some of the corpus specificities difficult to model for a machine learning system. In this paper, we explain our approach for buiding relevant systems dealing with such a corpus.
Research Interests:
Semantics technologies and text-mining methods are now mature enough to be deployed in wide, scalable, robust applications. We present Wikimeta, a good example of integration of research technology in a commercial application.
The Semantic Annotation (SA) task consists in establishing the relation between a textual entity (word or group of words designating a named entity of the real world or a concept) and its corresponding entity in an ontology. The main... more
The Semantic Annotation (SA) task consists in establishing the relation between a textual entity (word or group of words designating a named entity of the real world or a concept) and its corresponding entity in an ontology. The main difficulty of this task is that a textual entity might be highly polysemic and potentially related to many different ontological representations. To solve this specific problem, various Information Retrieval techniques can be used. Most of those involves contextual words to estimate wich exact textual entity have to be recognized. In this paper, we present a resource of contextual words that can be used by IR algorithms to establish a link between a named entity (NE) in a text and an entry point to its semantic description in the LinkedData Network.
This chapter presents the elaboration of an ontology-based application called Combine. This applica-tion aims to optimize and enhance e-Recruitment processes in the domain of Information Technologies' staffing services, and... more
This chapter presents the elaboration of an ontology-based application called Combine. This applica-tion aims to optimize and enhance e-Recruitment processes in the domain of Information Technologies' staffing services, and especially e-Recruitment processes that use Social ...
Research Interests:
FORMALIZANDO AS AMBIGUIDADES DO 1 ADJUNTO ADVERBIAL TEMPORAL: ATE SNA Gloria da Silva. Elena Godoy, Michel Gagnon ' ' Universldade Federal do Parané _ Introdugfio Existcrn algumas abordagcns propostas para interpretar, em um... more
FORMALIZANDO AS AMBIGUIDADES DO 1 ADJUNTO ADVERBIAL TEMPORAL: ATE SNA Gloria da Silva. Elena Godoy, Michel Gagnon ' ' Universldade Federal do Parané _ Introdugfio Existcrn algumas abordagcns propostas para interpretar, em um processo ...
ABSTRACT Wikimeta Lab participation in DeFT 2013 - Machine Learning for Information Extraction and Classification of Cooking Recipes. This paper presents Wikimeta Lab participation in the Défi Fouille de Texte (DeFT) 2013. In 2013, this... more
ABSTRACT Wikimeta Lab participation in DeFT 2013 - Machine Learning for Information Extraction and Classification of Cooking Recipes. This paper presents Wikimeta Lab participation in the Défi Fouille de Texte (DeFT) 2013. In 2013, this evaluation campaign is focused on mining cooking recipes in French. The campaign consists of three classification tasks and an information extraction task. The corpus is composed of recipes from a collaborative web site, thus they are written by users with almost no constraints on labels, ingredients, and comments. This very context makes some of the corpus specificities difficult to model for a machine learning system. In this paper, we explain our approach for buiding relevant systems dealing with such a corpus.
Research Interests:
This paper presents a high level view of a project which aims at improving the communication between people who do not share the same language by using a 3D animation upper layer to lift the inherent limitations of written text. We will... more
This paper presents a high level view of a project which aims at improving the communication between people who do not share the same language by using a 3D animation upper layer to lift the inherent limitations of written text. We will produce animation using the Collada file format, a standard based on the XML format. This project (GITAN ) has started in January 2010, and we expect to have the first results by the end of 2010. A limited set of sentences will validate the global process and help us refine the tools developed in the pipeline.
Research Interests:
The Cortex system is constructed of several different sentence selection metrics and a decision module. Our experiments have shown that the Cortex decision on the metrics always scores better than each system alone. In the INEX@ QA 2010... more
The Cortex system is constructed of several different sentence selection metrics and a decision module. Our experiments have shown that the Cortex decision on the metrics always scores better than each system alone. In the INEX@ QA 2010 task of Long ...
Research Interests:
Automatic keyword extraction is an important subfield of information extraction process. It is a difficult task, where numerous different techniques and resources have been proposed. In this paper, we propose a generic approach to extract... more
Automatic keyword extraction is an important subfield of information extraction process. It is a difficult task, where numerous different techniques and resources have been proposed. In this paper, we propose a generic approach to extract keyword from documents using encyclopedic knowledge. Our two-step approach first relies on a classification step for identifying candidate keywords followed by a learning-to-rank method depending on a user-defined keyword profile to order the candidates. The novelty of our approach relies on i) the usage of the keyword profile ii) generic features derived from Wikipedia categories and not necessarily related to the document content. We evaluate our system on keyword datasets and corpora from standard evaluation campaign and show that our system improves the global process of keyword extraction.