Items where Author is "Garay, Helena"

2025

Article Subjects > Nutrition Europe University of Atlantic > Research > Scientific Production
Fundación Universitaria Internacional de Colombia > Research > Scientific Production
Ibero-american International University > Research > Scientific Production
Ibero-american International University > Research > Articles and Books
Universidad Internacional do Cuanza > Research > Scientific Production
University of La Romana > Research > Scientific Production Open English Mango is one of the most beloved fruits and plays an indispensable role in the agricultural economies of many tropical countries like Pakistan, India, and other Southeast Asian countries. Similar to other fruits, mango cultivation is also threatened by various diseases, including Anthracnose and Red Rust. Although farmers try to mitigate such situations on time, early and accurate detection of mango diseases remains challenging due to multiple factors, such as limited understanding of disease diversity, similarity in symptoms, and frequent misclassification. To avoid such instances, this study proposes a multimodal deep learning framework that leverages both leaf and fruit images to improve classification performance and generalization. Individual CNN-based pre-trained models, including ResNet-50, MobileNetV2, EfficientNet-B0, and ConvNeXt, were trained separately on curated datasets of mango leaf and fruit diseases. A novel Modality Attention Fusion (MAF) mechanism was introduced to dynamically weight and combine predictions from both modalities based on their discriminative strength, as some diseases are more prominent on leaves than on fruits, and vice versa. To address overfitting and improve generalization, a class-aware augmentation pipeline was integrated, which performs augmentation according to the specific characteristics of each class. The proposed attention-based fusion strategy significantly outperformed individual models and static fusion approaches, achieving a test accuracy of 99.08%, an F1 score of 99.03%, and a perfect ROC-AUC of 99.96% using EfficientNet-B0 as the base. To evaluate the model’s real-world applicability, an interactive web application was developed using the Django framework and evaluated through out-of-distribution (OOD) testing on diverse mango samples collected from public sources. These findings underline the importance of combining visual cues from multiple organs of plants and adapting model attention to contextual features for real-world agricultural diagnostics. metadata Mohsin, Muhammad; Hashmi, Muhammad Shadab Alam; Delgado Noya, Irene; Garay, Helena; Abdel Samee, Nagwan and Ashraf, Imran mail UNSPECIFIED, UNSPECIFIED, irene.delgado@uneatlantico.es, helena.garay@uneatlantico.es, UNSPECIFIED, UNSPECIFIED (2025) Dual-modality fusion for mango disease classification using dynamic attention based ensemble of leaf & fruit images. Scientific Reports, 15 (1). ISSN 2045-2322

Article Subjects > Engineering Europe University of Atlantic > Research > Scientific Production
Ibero-american International University > Research > Scientific Production
Ibero-american International University > Research > Articles and Books
Universidad Internacional do Cuanza > Research > Scientific Production
University of La Romana > Research > Scientific Production Open English Introduction: The rapid expansion of generated data through social networks has introduced significant challenges, which underscores the need for advanced methods to analyze and interpret these complex systems. Deep learning has emerged as an effective approach, offering robust capabilities to process large datasets, and uncover intricate relationships and patterns. Methods: In this systematic literature review, we explore research conducted over the past decade, focusing on the use of deep learning techniques for community detection in social networks. A total of 19 studies were carefully selected from reputable databases, including the ACM Library, Springer Link, Scopus, Science Direct, and IEEE Xplore. This review investigates the employed methodologies, evaluates their effectiveness, and discusses the challenges identified in these works. Results: Our review shows that models like graph neural networks (GNNs), autoencoders, and convolutional neural networks (CNNs) are some of the most commonly used approaches for community detection. It also examines the variety of social networks, datasets, evaluation metrics, and employed frameworks in these studies. Discussion: However, the analysis highlights several challenges, such as scalability, understanding how the models work (interpretability), and the need for solutions that can adapt to different types of networks. These issues stand out as important areas that need further attention and deeper research. This review provides meaningful insights for researchers working in social network analysis. It offers a detailed summary of recent developments, showcases the most impactful deep learning methods, and identifies key challenges that remain to be explored. metadata El-Moussaoui, Mohamed; Hanine, Mohamed; Kartit, Ali; Gracia Villar, Mónica; Garay, Helena and de la Torre Díez, Isabel mail UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, monica.gracia@uneatlantico.es, helena.garay@uneatlantico.es, UNSPECIFIED (2025) A systematic review of deep learning methods for community detection in social networks. Frontiers in Artificial Intelligence, 8. ISSN 2624-8212

Article Subjects > Engineering Europe University of Atlantic > Research > Scientific Production
Ibero-american International University > Research > Scientific Production
Ibero-american International University > Research > Articles and Books
Universidad Internacional do Cuanza > Research > Scientific Production
University of La Romana > Research > Scientific Production Closed English Icons are the first visual element users encounter when searching for applications in online store. Icons with eye-catching features can make an app stand out in user searches, playing a crucial role in attracting user attention and influencing selection. This increases the likelihood of downloads, which can expand the user base, improve revenue, and enhance engagement, contributing to the application’s overall success. However, the majority of research focused on evaluating appeal of apps through application icons is empirical in nature and may lack comprehensive data analytical approaches. While empirical research holds its significance, it may still be limited by the size of the dataset analyzed and could also be subjective. This proposed research presents a novel data-analytical methodology to analyze a large dataset of application icons from Google Play to determine their influence on downloads. It clusters the icons using three different techniques: -means clustering with two distinct feature vectors and agglomerative clustering, extracting various visual features from the clusters that are strongly correlated with application installs. Subsequently, validation of results has revealed that factors of varied colors, the dominance of white or black colors, text, and exposure in the icons can be linked to downloads. metadata Bilal, Ahmad; Turab Mirza, Hamid; Ahmad, Adnan; Hussain, Ibrar; Raza, Ali; Garay, Helena; Alemany Iturriaga, Josep and Ashraf, Imran mail UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, helena.garay@uneatlantico.es, josep.alemany@uneatlantico.es, UNSPECIFIED (2025) On the correlation between Google Play Store application icons and downloads. The Computer Journal, 68 (10). pp. 1579-1593. ISSN 0010-4620

Article Subjects > Engineering Europe University of Atlantic > Research > Scientific Production
Fundación Universitaria Internacional de Colombia > Research > Scientific Production
Ibero-american International University > Research > Articles and Books
Universidad Internacional do Cuanza > Research > Scientific Production
University of La Romana > Research > Scientific Production Open English Efficient image retrieval from a variety of datasets is crucial in today's digital world. Visual properties are represented using primitive image signatures in Content Based Image Retrieval (CBIR). Feature vectors are employed to classify images into predefined categories. This research presents a unique feature identification technique based on suppression to locate interest points by computing productive sum of pixel derivatives by computing the differentials for corner scores. Scale space interpolation is applied to define interest points by combining color features from spatially ordered L2 normalized coefficients with shape and object information. Object based feature vectors are formed using high variance coefficients to reduce the complexity and are converted into bag-of-visual-words (BoVW) for effective retrieval and ranking. The presented method encompass feature vectors for information synthesis and improves the discriminating strength of the retrieval system by extracting deep image features including primitive, spatial, and overlayed using multilayer fusion of Convolutional Neural Networks(CNNs). Extensive experimentation is performed on standard image datasets benchmarks, including ALOT, Cifar-10, Corel-10k, Tropical Fruits, and Zubud. These datasets cover wide range of categories including shape, color, texture, spatial, and complicated objects. Experimental results demonstrate considerable improvements in precision and recall rates, average retrieval precision and recall, and mean average precision and recall rates across various image semantic groups within versatile datasets. The integration of traditional feature extraction methods fusion with multilevel CNN advances image sensing and retrieval systems, promising more accurate and efficient image retrieval solutions. metadata Chaki, Jyotismita; Shabir, Aiza; Ahmed, Khawaja Tehseen; Mahmood, Arif; Garay, Helena; Prado González, Luis Eduardo and Ashraf, Imran mail UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, helena.garay@uneatlantico.es, uis.prado@uneatlantico.es, UNSPECIFIED (2025) Deep image features sensing with multilevel fusion for complex convolution neural networks & cross domain benchmarks. PLOS ONE, 20 (3). e0317863. ISSN 1932-6203

Article Subjects > Engineering Europe University of Atlantic > Research > Scientific Production
Ibero-american International University > Research > Scientific Production
Ibero-american International University > Research > Articles and Books
Universidad Internacional do Cuanza > Research > Scientific Production
University of La Romana > Research > Scientific Production Open English Hand-drawn mathematical geometric shapes are geometric figures, such as circles, triangles, squares, and polygons, sketched manually using pen and paper or digital tools. These shapes are fundamental in mathematics education and geometric problem-solving, serving as intuitive visual aids for understanding complex concepts and theories. Recognizing hand-drawn shapes accurately enables more efficient digitization of handwritten notes, enhances educational tools, and improves user interaction with mathematical software. This research proposes an innovative machine learning algorithm for the automatic classification of mathematical geometric shapes to identify and interpret these shapes from handwritten input, facilitating seamless integration with digital systems. We utilized a benchmark dataset of mathematical shapes based on a total of 20,000 images with eight classes circle, kite, parallelogram, square, rectangle, rhombus, trapezoid, and triangle. We introduced a novel machine-learning algorithm CnN-RFc that uses convolution neural networks (CNN) for spatial feature extraction and the random forest classifier for probabilistic feature extraction from image data. Experimental results illustrate that using the CnN-RFc method, the Light Gradient Boosting Machine (LGBM) algorithm surpasses state-of-the-art approaches with high accuracy scores of 98% for hand-drawn shape classification. Applications of the proposed mathematical geometric shape classification algorithm span various domains, including education, where it enhances interactive learning platforms and provides instant feedback to students. metadata Alam, Aneeza; Raza, Ali; Thalji, Nisrean; Abualigah, Laith; Garay, Helena; Alemany Iturriaga, Josep and Ashraf, Imran mail UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, helena.garay@uneatlantico.es, josep.alemany@uneatlantico.es, UNSPECIFIED (2025) Novel transfer learning approach for hand drawn mathematical geometric shapes classification. PeerJ Computer Science, 11. e2652. ISSN 2376-5992

2024

Article Subjects > Engineering Europe University of Atlantic > Research > Scientific Production
Ibero-american International University > Research > Articles and Books
Universidad Internacional do Cuanza > Research > Scientific Production
University of La Romana > Research > Scientific Production Open English The perception and recognition of objects around us empower environmental interaction. Harnessing the brain’s signals to achieve this objective has consistently posed difficulties. Researchers are exploring whether the poor accuracy in this field is a result of the design of the temporal stimulation (block versus rapid event) or the inherent complexity of electroencephalogram (EEG) signals. Decoding perceptive signal responses in subjects has become increasingly complex due to high noise levels and the complex nature of brain activities. EEG signals have high temporal resolution and are non-stationary signals, i.e., their mean and variance vary overtime. This study aims to develop a deep learning model for the decoding of subjects’ responses to rapid-event visual stimuli and highlights the major factors that contribute to low accuracy in the EEG visual classification task.The proposed multi-class, multi-channel model integrates feature fusion to handle complex, non-stationary signals. This model is applied to the largest publicly available EEG dataset for visual classification consisting of 40 object classes, with 1000 images in each class. Contemporary state-of-the-art studies in this area investigating a large number of object classes have achieved a maximum accuracy of 17.6%. In contrast, our approach, which integrates Multi-Class, Multi-Channel Feature Fusion (MCCFF), achieves a classification accuracy of 33.17% for 40 classes. These results demonstrate the potential of EEG signals in advancing EEG visual classification and offering potential for future applications in visual machine models. metadata Rehman, Madiha; Anwer, Humaira; Garay, Helena; Alemany Iturriaga, Josep; Díez, Isabel De la Torre; Siddiqui, Hafeez ur Rehman and Ullah, Saleem mail UNSPECIFIED, UNSPECIFIED, helena.garay@uneatlantico.es, josep.alemany@uneatlantico.es, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED (2024) Decoding Brain Signals from Rapid-Event EEG for Visual Analysis Using Deep Learning. Sensors, 24 (21). p. 6965. ISSN 1424-8220

Article Subjects > Engineering
Subjects > Psychology Europe University of Atlantic > Research > Scientific Production
Fundación Universitaria Internacional de Colombia > Research > Scientific Production
Ibero-american International University > Research > Scientific Production
Ibero-american International University > Research > Articles and Books
Universidad Internacional do Cuanza > Research > Scientific Production Open English Predicting depression intensity from microblogs and social media posts has numerous benefits and applications, including predicting early psychological disorders and stress in individuals or the general public. A major challenge in predicting depression using social media posts is that the existing studies do not focus on predicting the intensity of depression in social media texts but rather only perform the binary classification of depression and moreover noisy data makes it difficult to predict the true depression in the social media text. This study intends to begin by collecting relevant Tweets and generating a corpus of 210000 public tweets using Twitter public application programming interfaces (APIs). A strategy is devised to filter out only depression-related tweets by creating a list of relevant hashtags to reduce noise in the corpus. Furthermore, an algorithm is developed to annotate the data into three depression classes: ‘Mild,’ ‘Moderate,’ and ‘Severe,’ based on International Classification of Diseases-10 (ICD-10) depression diagnostic criteria. Different baseline classifiers are applied to the annotated dataset to get a preliminary idea of classification performance on the corpus. Further FastText-based model is applied and fine-tuned with different preprocessing techniques and hyperparameter tuning to produce the tuned model, which significantly increases the depression classification performance to an 84% F1 score and 90% accuracy compared to baselines. Finally, a FastText-based weighted soft voting ensemble (WSVE) is proposed to boost the model’s performance by combining several other classifiers and assigning weights to individual models according to their individual performances. The proposed WSVE outperformed all baselines as well as FastText alone, with an F1 of 89%, 5% higher than FastText alone, and an accuracy of 93%, 3% higher than FastText alone. The proposed model better captures the contextual features of the relatively small sample class and aids in the detection of early depression intensity prediction from tweets with impactful performances. metadata Rizwan, Muhammad; Mushtaq, Muhammad Faheem; Rafiq, Maryam; Mehmood, Arif; Diez, Isabel de la Torre; Gracia Villar, Mónica; Garay, Helena and Ashraf, Imran mail UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, monica.gracia@uneatlantico.es, helena.garay@uneatlantico.es, UNSPECIFIED (2024) Depression Intensity Classification from Tweets Using FastText Based Weighted Soft Voting Ensemble. Computers, Materials & Continua, 78 (2). pp. 2047-2066. ISSN 1546-2226