arXiv subjects

Tarek Al Mustafa

Publications and source records attributed to Tarek Al Mustafa.

3 recordsLinked to original sources

arXiv2025

From Metadata to Meaning: A Semantic Units Knowledge Graph for the Biodiversity Exploratories

Knowledge Graphs (KGs) bear great potential for ecology and biodiversity researchers in their ability to support synthesis and integration efforts, meta-analyses, reasoning tasks, and overall machine interoperability of research data. However, this potential is yet to be realized as KGs are notoriously difficult to interact with via their query language SPARQL for many user groups alike. Additionally, a further hindrance for user-KG interaction is the fundamental disconnect between user requirements and requirements KGs have to fulfill regarding machine-interoperability, reasoning tasks, querying, and further technical requirements. Thus, many statements in a KG are of no semantic significance for end users. In this work, we investigate a potential remedy for this challenge: Semantic Units (SUs) are semantically significant, named subgraphs in a KG with the goal to enhance cognitive interoperability for users, and to provide responses to common KG modelling challenges. We model and construct a KG from publication and dataset metadata of the Biodiversity Exploratories (BE), a research platform for functional biodiversity research across research plots in Germany to contribute to biodiversity research from the perspective of computer science. We contribute further by delivering the first implementation of semantic units on a knowledge graph and investigate how SUs impact KG querying. Finally, we present two implementations of tasks that show how large language models (LLMs) can be used to extract structured metadata categories from publication and dataset titles and abstracts, and how embedding models can be used to enrich metadata with latent information, in an effort to support the creation of structured and FAIR (findable, accessible, interoperable, and reusable) metadata.

cs.DB↗

arXiv2025

Enhancing Domain-Specific Encoder Models with LLM-Generated Data: How to Leverage Ontologies, and How to Do Without Them

We investigate the use of LLM-generated data for continual pretraining of encoder models in specialized domains with limited training data, using the scientific domain of invasion biology as a case study. To this end, we leverage domain-specific ontologies by enriching them with LLM-generated data and pretraining the encoder model as an ontology-informed embedding model for concept definitions. To evaluate the effectiveness of this method, we compile a benchmark specifically designed for assessing model performance in invasion biology. After demonstrating substantial improvements over standard LLM pretraining, we investigate the feasibility of applying the proposed approach to domains without comprehensive ontologies by substituting ontological concepts with concepts automatically extracted from a small corpus of scientific abstracts and establishing relationships between concepts through distributional statistics. Our results demonstrate that this automated approach achieves comparable performance using only a small set of scientific abstracts, resulting in a fully automated pipeline for enhancing domain-specific understanding of small encoder models that is especially suited for application in low-resource settings and achieves performance comparable to masked language modeling pretraining on much larger datasets.

cs.CL↗

arXiv2025

Mining for Species, Locations, Habitats, and Ecosystems from Scientific Papers in Invasion Biology: A Large-Scale Exploratory Study with Large Language Models

This paper presents an exploratory study that harnesses the capabilities of large language models (LLMs) to mine key ecological entities from invasion biology literature. Specifically, we focus on extracting species names, their locations, associated habitats, and ecosystems, information that is critical for understanding species spread, predicting future invasions, and informing conservation efforts. Traditional text mining approaches often struggle with the complexity of ecological terminology and the subtle linguistic patterns found in these texts. By applying general-purpose LLMs without domain-specific fine-tuning, we uncover both the promise and limitations of using these models for ecological entity extraction. In doing so, this study lays the groundwork for more advanced, automated knowledge extraction tools that can aid researchers and practitioners in understanding and managing biological invasions.

cs.CL↗