Faiçal Azouaou

Using an Ontological and Rule-Based Approach for Contextual Semantic Annotations in Online Communities

Recently, a lot of research focuses on knowledge management and reuse. Pertinent reuse can facilitate learning, knowledge creation and sharing. In this research, we focus on the knowledge capitalization and reuse within Communities of Practice of E-learning (CoPEs). These communities are a virtual framework for exchanging and sharing techno-pedagogical knowledge and know-how between e-learning actors. In this chapter, we propose and discuss a knowledge capitalization approach for knowledge reuse within a CoPE. Our approach is based on contextual semantic annotations to model CoPEs members’ tacit and explicit knowledge. The context dimension represents the situation in which the members create or reuse annotations. To illustrate our approach, we have developed a prototype of knowledge capitalization system based on contextual semantic annotations, called CoPEAnnot. Ontological and rule-based context reasoning have been used to improve knowledge reuse by adapting CoPEAnnot features according to the current activity context of members. Preliminary tests and experimentation of CoPEAnnot conducted within a CoPE made up of members from the Algerian Higher National School of Computer Science show advantages and benefits.

Selection and Combination of Heterogeneous Mappings to Enhance Biomedical Ontology Matching

This paper presents a novel background knowledge approach which selects and combines existing mappings from a given biomedical ontology repository to improve ontology alignment. Current background knowledge approaches usually select either manually or automatically a limited number of different ontologies and use them as a whole for background knowledge. Whereas in our approach, we propose to pick up only relevant concepts and relevant existing mappings linking these concepts all together in a specific and customized background knowledge graph. Paths within this graph will help to discover new mappings. We have implemented and evaluated our approach using the content of the NCBO BioPortal repository and the Anatomy benchmark from the Ontology Alignment Evaluation Initiative. We used the mapping gain measure to assess how much our final background knowledge graph improves results of state-of-the-art alignment systems. Furthermore, the evaluation shows that our approach produces a high quality alignment and discovers mappings that have not been found by state-of-the-art systems.

WebAnnot: a learner's dedicated web-based annotation tool

In this paper, we propose a model of semantic annotation suitable for learners. It is used to implement a web annotation tool WebAnnot which enables the learners to annotate their pedagogical documents with graphical and semantic annotations. The annotation semantics is built using both generic and learning domain ontologies. WebAnnot provides two annotation functions: manual and pattern-based ones. WebAnnot is implemented as a Firefox add-on, using semantic web technologies and languages. We present the results of an evaluation study of the proposed model conducted with computer science undergraduate students, regarding their annotation experience using WebAnnot.

Multilingual Mapping Reconciliation between English-French Biomedical Ontologies

Even if multilingual ontologies are now more common, for historical reasons, in the biomedical domain, many ontologies or terminologies have been translated from one natural language to another resulting in two potentially aligned ontologies but with their own specificity (e.g., format, developers, and versions). Most often, there is no formal representation of the translation links between translated ontologies and original ones and those mappings are not formally available as linked data. However, these mappings are very important for the interoperability and the integration of multilingual biomedical data. In this paper, we propose an approach to represent translation mappings between ontologies based on the NCBO BioPortal format. We have reconciled more than 228K mappings between ten English ontologies hosted on NCBO BioPortal and their French translations. Then, we have stored both the translated ontologies and mappings on a French customized version of the platform, called the SIFR BioPortal, making the whole thing available in RDF. Reconciling the mappings turned more complex than expected because the translations are rarely exactly the same than the original ontologies as discussed in this paper.

Semantic Annotations and Context Reasoning to Enhance Knowledge Reuse in e-Learning

We address in this paper the need of improving knowledge reusability within online Communities of Practice of E-learning (CoPEs). Our approach is based on contextual semantic annotations. An ontological-based contextual semantic annotation model is presented. The model serves as the basis for implementing a context aware annotation system called “CoPEAnnot”. Ontological and rule-based context reasoning contribute to improving knowledge reuse by adapting CoPEAnnot’s search results, navigation and recommendation.The proposal has been experimented within a community of learners.

SQAR: An Annotation-Based Study Process to Enhance the Learner’s Personal Learning

Annotating pedagogical documents is a common habit among learners, thus, considering learners’ annotation activity in education system and especially in technology enhanced learning can have many benefits for learners. However, the annotation practice by oneself is not sufficient in helping a learner to work and better understand his pedagogic documents. So we propose, in this paper, an annotation-based pedagogical process called SQAR (Survey, Question, Annotation and Review) which aims to help the learner to enhance his learning activity and to make sure his learning evolution in both knowledge and ability. We use the annotation activity, within the context of our proposed SQAR process, to enhance the learner’s learning efficiency. The SQAR process is implemented into WebAnnot, a web–based annotation tool. This is because learners use increasingly web-based pedagogical documents that they annotate with their personal annotations. The evaluation experiment with undergraduate students shows that the annotation-based learning process SQAR, compared to other learning methods, provides an important guidance for the improvement of learners understanding and memorization of concepts and ideas, when they study their pedagogical documents.

An annotation-based pedagogical memory model for learner

This article proposes a new approach called PAML (Personal Annotation Memory for Learner) which aims to help learners in their learning process using the annotation activity. With our approach, the learner can build his own external memory, which is the set of whole annotations he creates on his pedagogical documents. We use also the annotation activity, within the context of our SQAR (Survey, Question, Annotation and Review) process, to improve the learning efficiency at learners. We start with defining the concept of learners external memory, and the learners personal annotation. Then we present the learners personal annotation formalism, and we show PAML architecture. Finally, we describe WebAnnot annotation tool that is the first prototype of our model; afterwards, we compare it with other related tools.

SentiALG: Automated Corpus Annotation for Algerian Sentiment Analysis

Data annotation is an important but time-consuming and costly procedure. To sort a text into two classes, the very first thing we need is a good annotation guideline, establishing what is required to qualify for each class. In the literature, the difficulties associated with an appropriate data annotation has been underestimated. In this paper, we present a novel approach to automatically construct an annotated sentiment corpus for Algerian dialect (A Maghrebi Arabic dialect). The construction of this corpus is based on an Algerian sentiment lexicon that is also constructed automatically. The presented work deals with the two widely used scripts on Arabic social media: Arabic and Arabizi. The proposed approach automatically constructs a sentiment corpus containing 8000 messages (where 4000 are dedicated to Arabic and 4000 to Arabizi). The achieved F1-score is up to 72% and 78% for an Arabic and Arabizi test sets, respectively. Ongoing work is aimed at integrating transliteration process for Arabizi messages to further improve the obtained results.

OLCPM: An Online Framework for Detecting Overlapping Communities in Dynamic Social Networks

Community structure is one of the most prominent features of complex networks. Community structure detection is of great importance to provide insights into the network structure and functionalities. Most proposals focus on static networks. However, finding communities in a dynamic network is even more challenging, especially when communities overlap with each other. In this article , we present an online algorithm, called OLCPM, based on clique percolation and label propagation methods. OLCPM can detect overlapping communities and works on temporal networks with a fine granularity. By locally updating the community structure, OLCPM delivers significant improvement in running time compared with previous clique percolation techniques. The experimental results on both synthetic and real-world networks illustrate the effectiveness of the method.