The Science and Information (SAI) Organization publishes open-access peer-reviewed journals in computer science and artificial intelligence.
Deep learning is a subfield of machine learning built on artificial neural networks with multiple layers that automatically learn hierarchical representations of data, reducing the need for manual feature engineering. Architectures include convolutional neural networks for image and spatial data, recurrent neural networks and long short-term memory networks for sequential data, and transformer models, which now underpin most state-of-the-art natural language processing and increasingly computer vision systems. Training deep networks typically relies on large labeled datasets, backpropagation, and gradient-based optimization, along with regularization techniques and specialized hardware such as GPUs and TPUs. A notable 2026 shift in the field favors smaller, specialized models over ever-larger ones, prioritizing reliability, transparency, and efficient inference over raw parameter count. Deep learning drives advances in image recognition, speech processing, machine translation, medical image diagnosis, and generative models for text, images, and audio. As an open-access deep learning journal, IJACSA covers novel deep learning architectures and their evaluation across vision, language, and applied domains.
Published in International Journal of Advanced Computer Science and Applications (IJACSA) · list last refreshed September 2026
Hyperparameter optimisation (HPO) remains a fundamental challenge in deep learning, especially for pre-trained convolutional neural networks (CNNs). While pre-trained models reduce the computational burden of training fr…
QR Codes are widely used in the digital era for storing and sharing information in various applications. However, they are often susceptible to physical damage such as scratches, tears, or fading, which can result in sca…
The rapid increase in defaults in the online lending industry highlights significant flaws in current debtor verification, which largely relies on static, preparable interviews, leading to high non-performing loans. Exis…
Overfitting and limited generalization remain significant challenges for deep learning models, often leading to suboptimal performance on unseen data. To address this, The Divided Ensemble Voting (DEV) method was introdu…
Due to the high volatility and complex interdependencies within financial markets, predicting Forex prices becomes a difficult challenge for investors. Furthermore, the traditional trading models struggle to capture thos…
Machine learning models are typically developed under the “closed-world” assumption, where training and testing data originate from a consistent distribution. However, in real-world scenarios, especially in the retail do…
Wireless capsule endoscopy (WCE) has revolutionized gastrointestinal (GI) diagnostics by offering a patient-friendly imaging and diagnostic tool compared to traditional endoscopic techniques. However, the manual assessme…
This study introduces a novel multi-feature fusion model aimed at improving text similarity calculation in scientific and technological projects. The primary objective is to enhance the accuracy and efficiency of assessi…
Human action recognition has many applications in different scenarios. With the advancement of wireless sensing and the widespread deployment of Wi-Fi devices, the perception technology of Wi-Fi channel state information…
During digital transformation, manufacturing enterprises encounter challenges such as the high cost of smart devices, operational interruptions, and increased technology expenses, raising their financial risks. Addressin…
