arXiv subjects

Sahraoui Dhelim

Publications and source records attributed to Sahraoui Dhelim.

At least 19 recordsLinked to original sources

arXiv2026

AraMS-28k: The Largest Publicly Released Line-Level Dataset of Historical Arabic Manuscripts with Margin and Insertion-Anchor Annotations

We introduce AraMS-28k, the largest publicly released line-level dataset of genuine historical Arabic manuscripts, comprising 14 books, 3,043 pages, and 28,600 annotated text lines (27,971 main-text, 629 margin). Thirteen books are hand-copied manuscripts spanning three script traditions -- Naskh, Ruq'ah, and Maghrebi -- and one is a lithographed printed edition included to broaden format diversity. Each line is labelled as main-text or margin, and margin lines that have an unambiguous attachment point in the main text are further annotated with an insertion anchor, recovering the manuscript's true non-linear reading order at line-level granularity -- to our knowledge the first such annotation released for a historical Arabic manuscript corpus. Because reference transcriptions are fully vocalised while manuscript hands are typically undiacritised, we release both the raw diacritised transcription and a diacritic-normalised counterpart for every line. The dataset was constructed with RefLAM, a reference-grounded annotation pipeline that aligns multimodal-LLM OCR against independently sourced clean transcriptions and routes every line through human review, combining automatic verification with expert oversight. We describe the construction and quality-control process, present the annotation schema, report dataset statistics at both the corpus and per-book level, and provide baseline HTR results using Kraken and HATFormer, including a cross-script generalisation gradient from in-distribution pages to fully unseen books. AraMS-28k is released with page images, line-level annotations, and fixed train/val/test splits under CC BY-NC-SA 4.0 to support reproducible research on Arabic manuscript recognition, layout analysis, and reading-order recovery.

cs.CV↗

arXiv2026

RefLAM: A Reference-Grounded Line Annotation Pipeline for Historical Arabic Manuscripts

Existing approaches to building line-level Arabic handwritten-text-recognition (HTR) training data either rely on fully manual annotation, which does not scale, or on automatic OCR-to-reference alignment methods not yet extended to multi-script, two-zone (main-plus-margin) manuscript layouts with a provable correctness guarantee. We present RefLAM (Reference-grounded Line Annotation for Manuscripts), a pipeline converting manuscript page images and clean transcriptions into validated, line-level ground truth without sacrificing human oversight. RefLAM couples a deep-learning page-segmentation model with a multimodal large language model (MLLM) for structured OCR and a diacritic-agnostic fuzzy alignment engine that grounds each OCR line in a contiguous span of the reference text, with a character-level confidence score in $[0,100]$. A perfect score is provably equivalent to character-for-character identity of the normalised strings (the Confidence-100 rule), verified with no counterexample across the released corpus. A reviewer can thus trust a perfect score, confirming most lines at a glance rather than retyping them, so annotation becomes triaged, with attention concentrated on uncertain alignments. Across 7 fully page-validated books we measured a 75$\times$ throughput gain over manual annotation (3,000 vs. 40 lines/hr); applying the same guarantee to 7 further books, we retained 16,533 confidence-100 main-text lines within one week, excluding sub-100 lines rather than manually correcting them. Using RefLAM, we release AraMS-28k: 14 historical Arabic manuscript books, 3,043 pages, and 27,971 main-text and 629 margin-line annotations with bounding boxes, layout labels, and insertion anchors for 191 margin entries (30.4%). We also finetune Muharaf-pretrained baselines (including HATFormer) on AraMS-28k and report CER results confirming its practical utility for downstream HTR training.

cs.CV↗

arXiv2026

Auditing Medical Vision-Language Models on Chest Radiographs: Estimating Reference Agreement Across Institutions

Vision-language models return structured chest-radiograph findings through interfaces exposing no confidence score, so a receiving institution cannot read off how far to trust an individual judgment. Whether agreement with an institution's reference standard transfers across sites, findings, prediction directions and question formats is largely unmeasured. We evaluated three generative vision-language models on three institutional chest-radiograph corpora and six findings under two elicitation protocols, comprising more than 345,000 finding-level predictions, and estimated finding-by-direction reference agreement at a receiving institution from a small budget of local labels. Estimation strategies were then stress-tested under repeated strict institution-held-out evaluation. Under evaluation excluding the receiving institution from development entirely, adaptive selection among the seven estimators that design admits did not improve on simple fixed alternatives: it achieved a mean Brier score of 0.1083, against 0.0853 for always using a Beta-Binomial empirical-Bayes estimator and 0.0855 for a target-only logistic model. Those two differ by 0.0003, less than this family's own sensitivity to a change of solver version, and each leads in about half the settings, so no default can be recommended. Their advantage over estimators pooling across institutions was concentrated at one site and not confirmatory once clustered by institution, and a plug-in empirical-Bayes posterior-predictive count interval at a nominal 95% level covered 87.0%, less at the hardest institution. Reference agreement therefore has to be re-evaluated per site and per interface; these results concern agreement with institutional labels, not clinical correctness.

cs.CV↗

arXiv2026

IMOVNO+: A Regional Partitioning and Meta-Heuristic Ensemble Framework for Imbalanced Multi-Class Learning

Class imbalance, overlap, and noise degrade data quality, reduce model reliability, and limit generalization. Although widely studied in binary classification, these issues remain underexplored in multi-class settings, where complex inter-class relationships make minority-majority structures unclear and traditional clustering fails to capture distribution shape. Approaches that rely only on geometric distances risk removing informative samples and generating low-quality synthetic data, while binarization approaches treat imbalance locally and ignore global inter-class dependencies. At the algorithmic level, ensembles struggle to integrate weak classifiers, leading to limited robustness. This paper proposes IMOVNO+ (IMbalance-OVerlap-NOise+ Algorithm-Level Optimization), a two-level framework designed to jointly enhance data quality and algorithmic robustness for binary and multi-class tasks. At the data level, first, conditional probability is used to quantify the informativeness of each sample. Second, the dataset is partitioned into core, overlapping, and noisy regions. Third, an overlapping-cleaning algorithm is introduced that combines Z-score metrics with a big-jump gap distance. Fourth, a smart oversampling algorithm based on multi-regularization controls synthetic sample proximity, preventing new overlaps. At the algorithmic level, a meta-heuristic prunes ensemble classifiers to reduce weak-learner influence. IMOVNO+ was evaluated on 35 datasets (13 multi-class, 22 binary). Results show consistent superiority over state-of-the-art methods, approaching 100% in several cases. For multi-class data, IMOVNO+ achieves gains of 37-57% in G-mean, 25-44% in F1-score, 25-39% in precision, and 26-43% in recall. In binary tasks, it attains near-perfect performance with improvements of 14-39%. The framework handles data scarcity and imbalance from collection and privacy limits.

cs.LG↗

arXiv2025

MedLiteNet: Lightweight Hybrid Medical Image Segmentation Model

Accurate skin-lesion segmentation remains a key technical challenge for computer-aided diagnosis of skin cancer. Convolutional neural networks, while effective, are constrained by limited receptive fields and thus struggle to model long-range dependencies. Vision Transformers capture global context, yet their quadratic complexity and large parameter budgets hinder use on the small-sample medical datasets common in dermatology. We introduce the MedLiteNet, a lightweight CNN Transformer hybrid tailored for dermoscopic segmentation that achieves high precision through hierarchical feature extraction and multi-scale context aggregation. The encoder stacks depth-wise Mobile Inverted Bottleneck blocks to curb computation, inserts a bottleneck-level cross-scale token-mixing unit to exchange information between resolutions, and embeds a boundary-aware self-attention module to sharpen lesion contours.

cs.CV↗

arXiv2025

AGI Enabled Solutions For IoX Layers Bottlenecks In Cyber-Physical-Social-Thinking Space

The integration of the Internet of Everything (IoX) and Artificial General Intelligence (AGI) has given rise to a transformative paradigm aimed at addressing critical bottlenecks across sensing, network, and application layers in Cyber-Physical-Social Thinking (CPST) ecosystems. In this survey, we provide a systematic and comprehensive review of AGI-enhanced IoX research, focusing on three key components: sensing-layer data management, network-layer protocol optimization, and application-layer decision-making frameworks. Specifically, this survey explores how AGI can mitigate IoX bottlenecks challenges by leveraging adaptive sensor fusion, edge preprocessing, and selective attention mechanisms at the sensing layer, while resolving network-layer issues such as protocol heterogeneity and dynamic spectrum management, neuro-symbolic reasoning, active inference, and causal reasoning, Furthermore, the survey examines AGI-enabled frameworks for managing identity and relationship explosion. Key findings suggest that AGI-driven strategies, such as adaptive sensor fusion, edge preprocessing, and semantic modeling, offer novel solutions to sensing-layer data overload, network-layer protocol heterogeneity, and application-layer identity explosion. The survey underscores the importance of cross-layer integration, quantum-enabled communication, and ethical governance frameworks for future AGI-enabled IoX systems. Finally, the survey identifies unresolved challenges, such as computational requirements, scalability, and real-world validation, calling for further research to fully realize AGI's potential in addressing IoX bottlenecks. we believe AGI-enhanced IoX is emerging as a critical research field at the intersection of interconnected systems and advanced AI.

cs.NI↗

arXiv2025

A Novel Double Pruning method for Imbalanced Data using Information Entropy and Roulette Wheel Selection for Breast Cancer Diagnosis

Accurate illness diagnosis is vital for effective treatment and patient safety. Machine learning models are widely used for cancer diagnosis based on historical medical data. However, data imbalance remains a major challenge, leading to hindering classifier performance and reliability. The SMOTEBoost method addresses this issue by generating synthetic data to balance the dataset, but it may overlook crucial overlapping regions near the decision boundary and can produce noisy samples. This paper proposes RE-SMOTEBoost, an enhanced version of SMOTEBoost, designed to overcome these limitations. Firstly, RE-SMOTEBoost focuses on generating synthetic samples in overlapping regions to better capture the decision boundary using roulette wheel selection. Secondly, it incorporates a filtering mechanism based on information entropy to reduce noise, and borderline cases and improve the quality of generated data. Thirdly, we introduce a double regularization penalty to control the synthetic samples proximity to the decision boundary and avoid class overlap. These enhancements enable higher-quality oversampling of the minority class, resulting in a more balanced and effective training dataset. The proposed method outperforms existing state-of-the-art techniques when evaluated on imbalanced datasets. Compared to the top-performing sampling algorithms, RE-SMOTEBoost demonstrates a notable improvement of 3.22\% in accuracy and a variance reduction of 88.8\%. These results indicate that the proposed model offers a solid solution for medical settings, effectively overcoming data scarcity and severe imbalance caused by limited samples, data collection difficulties, and privacy constraints.

cs.LG↗

arXiv2024

Machine Learning for Dynamic Management Zone in Smart Farming

Digital agriculture is growing in popularity among professionals and brings together new opportunities along with pervasive use of modern data-driven technologies. Digital agriculture approaches can be used to replace all traditional agricultural system at very reasonable costs. It is very effective in optimising large-scale management of resources, while traditional techniques cannot even tackle the problem. In this paper, we proposed a dynamic management zone delineation approach based on Machine Learning clustering algorithms using crop yield data, elevation and soil texture maps and available NDVI data. Our proposed dynamic management zone delineation approach is useful for analysing the spatial variation of yield zones. Delineation of yield regions based on historical yield data augmented with topography and soil physical properties helps farmers to economically and sustainably deploy site-specific management practices identifying persistent issues in a field. The use of frequency maps is capable of capturing dynamically changing incidental issues within a growing season. The proposed zone management approach can help farmers/agronomists to apply variable-rate N fertilisation more effectively by analysing yield potential and stability zones with satellite-based NDVI monitoring.

cs.CY↗

arXiv2024

Adapting Mental Health Prediction Tasks for Cross-lingual Learning via Meta-Training and In-context Learning with Large Language Model

Timely identification is essential for the efficient handling of mental health illnesses such as depression. However, the current research fails to adequately address the prediction of mental health conditions from social media data in low-resource African languages like Swahili. This study introduces two distinct approaches utilising model-agnostic meta-learning and leveraging large language models (LLMs) to address this gap. Experiments are conducted on three datasets translated to low-resource language and applied to four mental health tasks, which include stress, depression, depression severity and suicidal ideation prediction. we first apply a meta-learning model with self-supervision, which results in improved model initialisation for rapid adaptation and cross-lingual transfer. The results show that our meta-trained model performs significantly better than standard fine-tuning methods, outperforming the baseline fine-tuning in macro F1 score with 18\% and 0.8\% over XLM-R and mBERT. In parallel, we use LLMs' in-context learning capabilities to assess their performance accuracy across the Swahili mental health prediction tasks by analysing different cross-lingual prompting approaches. Our analysis showed that Swahili prompts performed better than cross-lingual prompts but less than English prompts. Our findings show that in-context learning can be achieved through cross-lingual transfer through carefully crafted prompt templates with examples and instructions.

cs.CL↗

arXiv2024

A Multi-Modal Latent-Features based Service Recommendation System for the Social Internet of Things

The Social Internet of Things (SIoT), is revolutionizing how we interact with our everyday lives. By adding the social dimension to connecting devices, the SIoT has the potential to drastically change the way we interact with smart devices. This connected infrastructure allows for unprecedented levels of convenience, automation, and access to information, allowing us to do more with less effort. However, this revolutionary new technology also brings an eager need for service recommendation systems. As the SIoT grows in scope and complexity, it becomes increasingly important for businesses and individuals, and SIoT objects alike to have reliable sources for products, services, and information that are tailored to their specific needs. Few works have been proposed to provide service recommendations for SIoT environments. However, these efforts have been confined to only focusing on modeling user-item interactions using contextual information, devices' SIoT relationships, and correlation social groups but these schemes do not account for latent semantic item-item structures underlying the sparse multi-modal contents in SIoT environment. In this paper, we propose a latent-based SIoT recommendation system that learns item-item structures and aggregates multiple modalities to obtain latent item graphs which are then used in graph convolutions to inject high-order affinities into item representations. Experiments showed that the proposed recommendation system outperformed state-of-the-art SIoT recommendation methods and validated its efficacy at mining latent relationships from multi-modal features.

cs.SI↗

arXiv2024

Maximizing UAV Fog Deployment Efficiency for Critical Rescue Operations

In disaster scenarios and high-stakes rescue operations, integrating Unmanned Aerial Vehicles (UAVs) as fog nodes has become crucial. This integration ensures a smooth connection between affected populations and essential health monitoring devices, supported by the Internet of Things (IoT). Integrating UAVs in such environments is inherently challenging, where the primary objectives involve maximizing network connectivity and coverage while extending the network's lifetime through energy-efficient strategies to serve the maximum number of affected individuals. In this paper, We propose a novel model centred around dynamic UAV-based fog deployment that optimizes the system's adaptability and operational efficacy within the afflicted areas. First, we decomposed the problem into two subproblems. Connectivity and coverage subproblem, and network lifespan optimization subproblem. We shape our UAV fog deployment problem as a uni-objective optimization and introduce a specialized UAV fog deployment algorithm tailored specifically for UAV fog nodes deployed in rescue missions. While the network lifespan optimization subproblem is efficiently solved via a one-dimensional swapping method. Following that, We introduce a novel optimization strategy for UAV fog node placement in dynamic networks during evacuation scenarios, with a primary focus on ensuring robust connectivity and maximal coverage for mobile users, while extending the network's lifespan. Finally, we introduce Adaptive Whale Optimization Algorithm (WOA) for fog node deployment in a dynamic network. Its agility, rapid convergence, and low computational demands make it an ideal fit for high-pressure environments.

cs.NI↗

arXiv2024

Selective Task offloading for Maximum Inference Accuracy and Energy efficient Real-Time IoT Sensing Systems

The recent advancements in small-size inference models facilitated AI deployment on the edge. However, the limited resource nature of edge devices poses new challenges especially for real-time applications. Deploying multiple inference models (or a single tunable model) varying in size and therefore accuracy and power consumption, in addition to an edge server inference model, can offer a dynamic system in which the allocation of inference models to inference jobs is performed according to the current resource conditions. Therefore, in this work, we tackle the problem of selectively allocating inference models to jobs or offloading them to the edge server to maximize inference accuracy under time and energy constraints. This problem is shown to be an instance of the unbounded multidimensional knapsack problem which is considered a strongly NP-hard problem. We propose a lightweight hybrid genetic algorithm (LGSTO) to solve this problem. We introduce a termination condition and neighborhood exploration techniques for faster evolution of populations. We compare LGSTO with the Naive and Dynamic programming solutions. In addition to classic genetic algorithms using different reproduction methods including NSGA-II, and finally we compare to other evolutionary methods such as Particle swarm optimization (PSO) and Ant colony optimization (ACO). Experiment results show that LGSTO performed 3 times faster than the fastest comparable schemes while producing schedules with higher average accuracy.

cs.LG↗

arXiv2023

IP-UNet: Intensity Projection UNet Architecture for 3D Medical Volume Segmentation

CNNs have been widely applied for medical image analysis. However, limited memory capacity is one of the most common drawbacks of processing high-resolution 3D volumetric data. 3D volumes are usually cropped or downsized first before processing, which can result in a loss of resolution, increase class imbalance, and affect the performance of the segmentation algorithms. In this paper, we propose an end-to-end deep learning approach called IP-UNet. IP-UNet is a UNet-based model that performs multi-class segmentation on Intensity Projection (IP) of 3D volumetric data instead of the memory-consuming 3D volumes. IP-UNet uses limited memory capability for training without losing the original 3D image resolution. We compare the performance of three models in terms of segmentation accuracy and computational cost: 1) Slice-by-slice 2D segmentation of the CT scan images using a conventional 2D UNet model. 2) IP-UNet that operates on data obtained by merging the extracted Maximum Intensity Projection (MIP), Closest Vessel Projection (CVP), and Average Intensity Projection (AvgIP) representations of the source 3D volumes, then applying the UNet model on the output IP images. 3) 3D-UNet model directly reads the 3D volumes constructed from a series of CT scan images and outputs the 3D volume of the predicted segmentation. We test the performance of these methods on 3D volumetric images for automatic breast calcification detection. Experimental results show that IP-Unet can achieve similar segmentation accuracy with 3D-Unet but with much better performance. It reduces the training time by 70\% and memory consumption by 92\%.

eess.IV↗

arXiv2023

Task Offloading for Smart Glasses in Healthcare: Enhancing Detection of Elevated Body Temperature

Wearable devices like smart glasses have gained popularity across various applications. However, their limited computational capabilities pose challenges for tasks that require extensive processing, such as image and video processing, leading to drained device batteries. To address this, offloading such tasks to nearby powerful remote devices, such as mobile devices or remote servers, has emerged as a promising solution. This paper focuses on analyzing task-offloading scenarios for a healthcare monitoring application performed on smart wearable glasses, aiming to identify the optimal conditions for offloading. The study evaluates performance metrics including task completion time, computing capabilities, and energy consumption under realistic conditions. A specific use case is explored within an indoor area like an airport, where security agents wearing smart glasses to detect elevated body temperature in individuals, potentially indicating COVID-19. The findings highlight the potential benefits of task offloading for wearable devices in healthcare settings, demonstrating its practicality and relevance.

cs.NI↗

arXiv2023

Context-Aware Service Recommendation System for the Social Internet of Things

The Social Internet of Things (SIoT) enables interconnected smart devices to share data and services, opening up opportunities for personalized service recommendations. However, existing research often overlooks crucial aspects that can enhance the accuracy and relevance of recommendations in the SIoT context. Specifically, existing techniques tend to consider the extraction of social relationships between devices and neglect the contextual presentation of service reviews. This study aims to address these gaps by exploring the contextual representation of each device-service pair. Firstly, we propose a latent features combination technique that can capture latent feature interactions, by aggregating the device-device relationships within the SIoT. Then, we leverage Factorization Machines to model higher-order feature interactions specific to each SIoT device-service pair to accomplish accurate rating prediction. Finally, we propose a service recommendation framework for SIoT based on review aggregation and feature learning processes. The experimental evaluation demonstrates the framework's effectiveness in improving service recommendation accuracy and relevance.

cs.IR↗

arXiv2023

Blockchain Application on the Internet of Vehicles (IoV)

With the rapid development of the Internet of Things (IoT) and its potential integration with the traditional Vehicular Ad-Hoc Networks (VANETs), we have witnessed the emergence of the Internet of Vehicles (IoV), which promises to seamlessly integrate into smart transportation systems. However, the key characteristics of IoV, such as high-speed mobility and frequent disconnections make it difficult to manage its security and privacy. The Blockchain, as a distributed tamper-resistant ledge, has been proposed as an innovative solution that guarantees privacy-preserving yet secure schemes. In this paper, we review recent literature on the application of blockchain to IoV, in particular, and intelligent transportation systems in general.

cs.NI↗

arXiv2023

Efficient Fog Node Placement using Nature-Inspired Metaheuristic for IoT Applications

Managing the explosion of data from the edge to the cloud requires intelligent supervision such as fog node deployments, which is an essential task to assess network operability. To ensure network operability, the deployment process must be carried out effectively in terms of two main factors: connectivity and coverage. The network connectivity is based on fog node deployment which determines the physical topology of the network while the coverage determines the network accessibility. Both have a significant impact on network performance and guarantee the network QoS. Determining an optimum fog node deployment method that minimizes cost, reduces computation and communication overhead, and provides a high degree of network connection coverage is extremely hard. Therefore, maximizing coverage as well as preserving network connectivity is a non-trivial problem. In this paper, we proposed a fog deployment algorithm that can effectively connect the fog nodes and cover all edge devices. Firstly, we formulate fog deployment as an instance of multi-objective optimization problems with a large search space. Then, we leverage Marine Predator Algorithm (MPA) to tackle the deployment problem and prove that MPA is well-suited for fog node deployment due to its rapid convergence and low computational complexity compared to other population-based algorithms. Finally, we evaluate the proposed algorithm on a different benchmark of generated instances with various fog scenario configurations. The experimental results demonstrate that our proposed algorithm is capable of providing very promising results when compared to state-of-the-art methods for determining an optimal deployment of fog nodes.

cs.NI↗

arXiv2022

Artificial Intelligence for Suicide Assessment using Audiovisual Cues: A Review

Death by suicide is the seventh leading death cause worldwide. The recent advancement in Artificial Intelligence (AI), specifically AI applications in image and voice processing, has created a promising opportunity to revolutionize suicide risk assessment. Subsequently, we have witnessed fast-growing literature of research that applies AI to extract audiovisual non-verbal cues for mental illness assessment. However, the majority of the recent works focus on depression, despite the evident difference between depression symptoms and suicidal behavior and non-verbal cues. This paper reviews recent works that study suicide ideation and suicide behavior detection through audiovisual feature analysis, mainly suicidal voice/speech acoustic features analysis and suicidal visual cues. Automatic suicide assessment is a promising research direction that is still in the early stages. Accordingly, there is a lack of large datasets that can be used to train machine learning and deep learning models proven to be effective in other, similar tasks.

cs.AI↗