Axel-Cyrille Ngonga Ngomo

Axel-Cyrille Ngonga Ngomo

Auto-generated profile, based on publicly available information

Axel-Cyrille Ngonga Ngomo

Paderborn University

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 November 2021 in The VLDB Journal
Verified
Journal Article
Closed access
PDF file
WAWaqas Ali
Waqas Ali
MSMuhammad Saleem
Muhammad Saleem
BYBin Yao
Bin Yao
AHAidan Hogan
Aidan Hogan
ANAxel-Cyrille Ngonga Ngomo
Axel-Cyrille Ngonga Ngomo

11 September 2024
Verified
Book Chapter
Open access
PDF file
HZHamada M. Zahera
Hamada M. Zahera
MAManzoor Ali
Manzoor Ali
MSMohamed Ahmed Sherif
Mohamed Ahmed Sherif
DMDiego Moussallem
Diego Moussallem
ANAxel-Cyrille Ngonga Ngomo
Axel-Cyrille Ngonga Ngomo
Purpose: SPARQL is a highly expressive query language for knowledge graphs; yet, formulating precise SPARQL queries can be challenging for non-expert users. A potential solution is translating natural questions into SPARQL queries, known as SPARQL generation. This paper addresses the challenges of translating natural language questions into SPARQL queries for different knowledge graphs. Methodology: We propose COT-SPARQL, our approach to generate SPARQL queries from input questions. Our approach employs Chain-of-thoughts prompting that guides large language models through intermediate reasoning steps and facilitates generating precise SPARQL queries. Furthermore, our approach incorporates entities and relations from the input question, and one-shot example in the prompt to provide additional context during the query generation process. Findings: We conducted several experiments on benchmark datasets and showed that our approach outperforms the state-of-the-art methods by a large margin. Our approach achieves a significant improvement in F1 score of 4.4% and 3.0% for the QALD-10 and QALD-9 datasets, respectively. Value: Our COT-SPARQL approach contributes to the semantic web community by simplifying access to knowledge graphs for non-expert users. In particular, COTSPARQL enables non-expert end-users to query knowledge graphs in natural languages, where COT-SPARQL converts user natural languages queries into SPARQL queries, which can be executed via the knowledge graph’s SPARQL endpoint.

25 August 2021 in IEEE Access
Verified
Journal Article
Open access
HZHamada M. Zahera
Hamada M. Zahera
RJRricha Jalota
Rricha Jalota
MSMohamed Ahmed Sherif
Mohamed Ahmed Sherif
ANAxel-Cyrille Ngonga Ngomo
Axel-Cyrille Ngonga Ngomo
Social media plays a significant role in disaster management by providing valuable data about affected people, donations, and help requests. Recent studies highlight the need to filter information on social media into fine-grained content labels. However, identifying useful information from massive amounts of social media posts during a crisis is a challenging task. In this paper, we propose I-AID, a multimodel approach to automatically categorize tweets into multi-label information types and filter critical information from the enormous volume of social media data. I-AID incorporates three main components: i) a BERT-based encoder to capture the semantics of a tweet and represent as a low-dimensional vector, ii) a graph attention network (GAT) to apprehend correlations between tweets’ words/entities and the corresponding information types, and iii) a Relation Network as a learnable distance metric to compute the similarity between tweets and their corresponding information types in a supervised way. We conducted several experiments on two real publicly-available datasets. Our results indicate that I-AID outperforms state-of-the-art approaches in terms of weighted average F1 score by +6% and +4% on the TREC-IS dataset and COVID-19 Tweets, respectively.

13 September 2019 in Data Science
Verified
Journal Article
Open access
PDF file
MRMichael Röder
Michael Röder
DKDenis Kuchelev
Denis Kuchelev
ANAxel-Cyrille Ngonga Ngomo
Axel-Cyrille Ngonga Ngomo
An increasing number of solutions aim to support the steady increase of the number of requirements and requests for Linked Data at scale. This plethora of solutions leads to a growing need for objective means that facilitate the selection of adequate solutions for particular use cases. We hence present Hobbit, a distributed benchmarking platform designed for the unified execution of benchmarks for Linked Data solutions. The Hobbit benchmarking platform is based on the FAIR principles and is the first benchmarking platform able to scale up to benchmarking real-world scenarios for Big Linked Data solutions. Our online instance of the platform has more than 300 registered users and offers more than 40 benchmarks. It has been used in eleven benchmarking challenges and for more than 13000 experiments. We give an overview of the results achieved during 2 of these challenges and point to some of the novel insights that were gained from the results of the platform. Hobbit is open-source and available at http://github.com/hobbit-project .

1 August 2023
Verified
Conference Paper
Open access
PDF file
CDCaglar Demir
Caglar Demir
ANAxel-Cyrille Ngonga Ngomo
Axel-Cyrille Ngonga Ngomo
Models computed using deep learning have been effectively applied to tackle various problems in many disciplines. Yet, the predictions of these models are often at most post-hoc and locally explainable. In contrast, class expressions in description logics are ante-hoc and globally explainable. Although state-of-the-art symbolic machine learning approaches are being successfully applied to learn class expressions, their application at large scale has been hindered by their impractical runtimes. Arguably, the reliance on myopic heuristic functions contributes to this limitation. We propose a novel neuro-symbolic class expression learning model, DRILL, to mitigate this limitation. By learning non-myopic heuristic functions with deep Q-learning, DRILL efficiently steers the standard search procedure in a quasi-ordered search space towards goal states. Our extensive experiments on 4 benchmark datasets and 390 learning problems suggest that DRILL converges to goal states at least 2.7 times faster than state-of-the-art models on all learning problems. The results of our statistical significance test confirms that DRILL converges to goal states significantly faster (p-value <1%) than state-of-the-art models on all benchmark datasets. We provide an open-source implementation of DRILL, including pre-trained models, training and evaluation scripts.

16 October 2022
Verified
Book Chapter
Closed access
PDF file
UQUMAIR QUDUS
UMAIR QUDUS
MRMichael Röder
Michael Röder
MSMuhammad Saleem
Muhammad Saleem
ANAxel-Cyrille Ngonga Ngomo
Axel-Cyrille Ngonga Ngomo