Small language models (SLMs) have shown promises for relation extraction (RE) when extracting RDF triples guided by SHACL shapes focused on common Datatype Properties. This paper investigates how SLMs handle both Datatype and Object Properties for a complete RDF graph extraction. We show that the key bottleneck is related to long-tail distribution of rare properties. To solve this issue, we evaluate several strategies: stratified sampling, weighted loss, dataset scaling, and template-based synthetic data augmentation. We show that the best strategy to perform equally well over unbalanced target properties is to build a training set where the number of occurrences of each property exceeds a given threshold. To enable reproducibility, we publicly released our datasets, experimental results and code. Our findings offer practical guidance for training shape-aware SLMs and highlight promising directions for future work in semantic RE.

Electronic medical records (EMR) contain key information about the different symptomatic episodes that a patient went through. They carry a great potential in order to improve the well-being of patients and therefore represent a very valuable input for artificial intelligence approaches. However, the explicit knowledge directly available through these records remains limited, the extracted features to be used by machine learning algorithms do not contain all the implicit knowledge of medical expert. In order to evaluate the impact of domain knowledge when processing EMRs, we augment the features extracted from EMRs with ontological resources before turning them into vectors used by machine learning algorithms. We evaluate these augmentations with several machine learning algorithms to predict hospitalization. Our approach was experimented on data from the PRIMEGE PACA database that contains more than 350,000 consultations carried out by 16 general practitioners (GPs).

The increase in the amount of structured data published on the Web using the principles of Linked Data means that now it is more likely to find resources on the Web of Data that represent real life concepts. Discovering and recommending resources on the Web of Data related to a given resource is still an open research area. This work presents a framework to deploy and execute Linked Data based recommendation algorithms to measure their accuracy and performance in different contexts. Moreover, application developers can use this framework as the main component for recommendation in various domains. Finally, this paper describes a new recommendation algorithm that adapts its behavior dynamically based on the features of the Linked Data dataset used. The results of a user study show that the algorithm proposed in this paper has better accuracy and novelty than other state-of-the-art algorithms for Linked Data.

Users in question-answer sites generate huge amounts of high quality and highly reusable information. This information can be categorized by topics but since users' interests change with time, uncovering the temporal patterns and trends in their activity is of prime interest to detect their current expertize. These temporal variations have long remained unexplored in question-answer sites while detecting them enables us to improve tasks such as: question routing, expert recommending and community life-cycle management. In this paper, we propose a generative model of such a community and its dynamics, and we perform experiments with real-world data extracted from the StackOverflow website to confirm the effectiveness of our model to study the users' behaviors and topics dynamics.

Axiom scoring is a critical task both for the automatic enrichment/learning and for the automatic validation of knowledge bases and ontologies. We designed and developed an axiom scoring heuristic based on possibility theory, which aims at overcoming some limitations of scoring heuristics based on statistical inference and taking into account the open-world assumption of the linked data on the Web. Since computing the possibilistic score can be computationally quite heavy for some candidate axioms, we propose a method based on time capping to alleviate the computation of the heuristic without giving up the precision of the scores. We evaluate our proposal by applying it to the problem of testing SubClassOf axioms against the DBpedia RDF dataset.

In many social networks, people interact based on their interests. Community detection algorithms are then useful to reveal the sub-structures of a network and help us find interest groups. Identifying these social communities can bring benefit to understanding and predicting users behaviors. However, for some kind of online community sites such as question-and-answer (Q&A) sites or forums, there is no friendship based social network structure, which means people are not aware who they are in contact with. Therefore, many traditional community detection techniques do not apply directly. In this paper, we propose an empirical approach for extracting data from Q&A sites suitable to apply community detection methods. Then we compare three kinds of community detection methods we applied on a dataset extracted from the popular Q&A site StackOverflow. We analyze and comment the results of each method.

Querying and linking distributed and heterogeneous databases is increasingly needed, as plentiful data resources are published over the Web. This work describes the design of a versatile query system named KGRAM that supports (i) multiple query languages among which the SPARQL 1.1 standard, (ii) federation of multiple heterogeneous and distributed data sources, and (iii) adaptability to various data manipulation use cases. KGRAM provides abstractions for both the query language and the data model, thus delivering unifying reasoning mechanisms. It is implemented as a modular software suite to ease architecting and deploying dedicated data manipulation platforms. Its design integrates optimization concerns to deliver high query performance. Both KGRAM's software versatility and performance are evaluated.

In this paper we present the KGRAM abstract machine dedicated to querying knowledge graphs. It is the result of an abstraction process we performed to reach a generic solution to the problem of querying graphs in various models. We identified high level abstract primitives that constitute the expressions of the query language and the interfaces of KGRAM for both its data structures and its operations.

The basic principle of the Semantic Web carried by the RDF data model is that a collection of RDF statements coexist all together and are universally true. However some case study imply contextual relevancy and truth. The SPARQL query language is provided with patterns enabling to choose the RDF dataset against which a query is executed. This is a first step to handle contextual metadata. Based on it, we present in this paper a design pattern to handle contextual metadata hierarchically organized. This is done by means of a subStateOf property reifying RDF entailment between contextual graphs. The subStateOf relation is modelled within RDF and therefore context hierarchies can be described and queried by means of SPARQL queries. We propose a slight syntactic extension to SPARQL to facilitate the query of context hierarchies, together with rewriting rules to return to standard SPARQL.