Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

29,922 Article Results

Life balloon: a paradigm shift in earthquake safety-intelligent IoT detection and protection system for optimal resilience

10.11591/ijai.v15.i1.pp987-997
Tawfiq Alrawashdeh , Sumaya Abusaleh , Malek Z. Alksasbeh , Khalid Alemerien
Internet of things (IoT) applications for environmental monitoring have greatly improved due to advances in hardware and software technologies. Given the significant economic and societal impacts of earthquakes, there is an increasing need to develop effective earthquake early warning systems (EEWS). However, designing such intelligent systems remains challenging because of inefficient classification methods and limitations in high-fidelity sensing capabilities. To reduce the devastating effects of earthquakes, this paper proposes an earthquake detection and protection system. The system’s primary function is to detect seismic signals and activate a specially designed airbag (life balloon) unit that protects occupants in apartment buildings. In addition, the unit helps maintain necessary oxygen levels, thereby improving occupant safety during seismic events. The proposed system also includes a communication method that transmits critical information about the affected area to relevant parties. Early data transmission enables rapid response and guides the efficient deployment of required resources, making aftershock management more effective. By combining advanced sensor technologies with efficient communication methods, the proposed system aims to enhance safety and emergency management while providing comprehensive protection and support during seismic events. Experimental results show that the proposed method achieves approximately 95% sensitivity and 94.2% accuracy.
Volume: 15
Issue: 1
Page: 987-997
Publish at: 2026-02-01

Enhancing medical language models with big data technologies

10.11591/ijai.v15.i1.pp289-299
Ayoub Allali , Ibtihal Abouchabaka , Najat Rafalia
In this study, we present an end-to-end, big-data–driven framework for continuously enriching and fine-tuning large language models (LLMs) with the latest professional and scientific medical knowledge. Streaming updates from premier sources such as The New England Journal of Medicine (NEJM) are ingested via an Apache Kafka cluster for low-latency delivery and durably archived in a three-node Apache Hadoop (Hadoop distributed file system (HDFS)) system. Each new article is preprocessed into high dimensional embeddings and indexed in a Milvus vector database to enable sub-second semantic retrieval over millions of records. At query or batch time, our retrieval-augmented generation (RAG) module retrieves the top-k relevant embeddings from Milvus and injects them into prompts for DeepSeek-R1, GPT-4o-mini, and Llama 3, models which are hosted, fine tuned, and served via Ollama on an NVIDIA GeForce RTX 3050 Ti GPU for efficient inference and continual learning. The enriched outputs are seamlessly delivered to end users through a Telegram bot programmed in Python using the Telebot library, linking the RAG-enhanced LLMs to an intuitive chat interface. Our Kafka, HDFS, Milvus, RAG, LLM, or Telegram bot pipeline demonstrably improves factual accuracy and topical currency of AI-generated medical insights across clinical decision support, patient engagement and education, drug discovery and development, virtual health assistants, and mental health support, laying the groundwork for truly intelligent, responsive, and data-driven healthcare solutions.
Volume: 15
Issue: 1
Page: 289-299
Publish at: 2026-02-01

Exploring artificial intelligence adoption challenges: bridging the technology gap for marketing advancements

10.11591/ijai.v15.i1.pp56-65
Susi Susanti Tindaon , Jonas Meylan Freddy Banurea
Artificial intelligence (AI) holds immense potential to enhance marketing strategies for micro, small, and medium enterprises (MSMEs). However, significant barriers, including financial constraints, limited digital literacy, and inadequate government support, hinder its widespread adoption. This study compares AI adoption challenges across two Indonesian MSME ecosystems; tourism-oriented Bali and agrarian Garut using the diffusion of innovation (DOI) lens and complementary model⎯technology organization environment (TOE), technology acceptance model (TAM), unified theory of acceptance and use of technology (UTAUT). The findings reveal that MSMEs in Bali exhibit characteristics of "early adopters" and "early majority," driven by the demands of the tourism sector. In contrast, MSMEs in Garut align more with "late majority" and "laggard" profiles, constrained by infrastructural and resource limitations. By mapping these regional disparities onto the DOI curve, this study provides actionable insights for policymakers. It advocates for the development of ecosystem-aware, tailored strategies to bridge the digital divide and foster inclusive digital transformation, enabling MSMEs across diverse regions to leverage AI for a sustainable competitive advantage.
Volume: 15
Issue: 1
Page: 56-65
Publish at: 2026-02-01

Malware detection using convolutional neural network-di strategy polar fox optimization algorithm

10.11591/ijai.v15.i1.pp140-153
Parvathi Sathenahalli Jayaprakash , Yogeesh Ambalagere Chandrashekaraiah
Malware attacks have escalated significantly with an increase of internet users and connected devices. With the rise of various types of malwares released by the hackers, constructing new competitive methods are necessary to identify the advanced malware. However, conventional malware detection struggles to identify new and evolving malware variants accurately because of its dependence on handcrafted features and static-signature based methods. To address this problem, this research proposes convolutional neural network (CNN) based di strategy polar fox optimization algorithm (DSPFOA) for malware detection to fine-tune the CNN parameters effectively which later assists to overcome the limitations of CNN. The model integrates the sine chaotic mapping and Cauchy operator mutation as DSPFOA prevents the model from local optima issue, and extends search space solution, also enhance convergence. This ensures that the CNN learns highly discriminative features which makes the system more accurate and robust in detecting both known and evolving malware variants. The CNN DSPFOA achieves a high accuracy of 99.65 and 99.76% by utilizing BIG2015 and Malimg dataset respectively compared to existing methods like masked self-supervised model with swin transformer (MalSort).
Volume: 15
Issue: 1
Page: 140-153
Publish at: 2026-02-01

Unmanned aerial vehicle systems: key management and intrusion response techniques

10.11591/ijai.v15.i1.pp972-986
Reshma C. Sonawane , A. Muthukrishnan
Protecting information is crucial for unmanned aerial vehicle (UAV) network communications, particularly during delicate tasks such as surveillance and reconnaissance. While encryption safeguards data privacy, managing and distributing keys in UAV settings poses significant challenges due to the vehicles' mobility and limited processing power. This research proposes an efficient key management scheme utilizing elliptic curve cryptography (ECC) and bilinear pairings, complemented by a lightweight intrusion detection system (IDS). The method employs behavior-based anomaly detection, using cluster-based watchdogs and trust assessment to identify and isolate harmful nodes. Additionally, applying compression techniques before encryption helps to reduce transmission load. Simulation in NS2 demonstrates performance improvements of 6-10% in throughput, 4-6% in packet delivery ratio (PDR), and a 13-17% reduction in delay compared to elliptic curve cryptography Diffie–Hellman (ECCDH) and pairwise encryption methods.
Volume: 15
Issue: 1
Page: 972-986
Publish at: 2026-02-01

Evaluation of artificial intelligence algorithms to estimate water quality parameters using satellite images

10.11591/ijai.v15.i1.pp559-567
Julio Cesar Anaya-Valenzuela , Gloria Yaneth Florez-Yepes , Yeison Alberto Garcés-Gómez
The Ciénaga de la Virgen (Virgen Swamp) is a coastal lagoon in Cartagena de Indias that provides multiple ecosystem services in northern Bolívar. This ecosystem has faced anthropogenic pressure from city growth and improper water resource management, including wastewater and agrochemical discharges. Consequently, environmental authorities must monitor certain sites within the water body and extrapolate the data across its entire expanse. In this study, predictive tools are applied to determine water quality parameters such as chlorophyll-a (CL-a), dissolved oxygen (DO), total suspended solids (TSS), and salinity. This is achieved by correlating traditionally obtained data with the spectral response of medium-resolution satellite images, adjusted using artificial intelligence (AI) algorithms. Support vector machine (SVM) algorithms were used for regression, random forests (RF), and artificial neural networks (ANN), achieving an accuracy of 79% for CL-a, 95% for DO, 89% for TSS, and 96% for salinity. Validation was performed using mean absolute percentage error (MAPE) statistical metrics and root mean square error (RMSE).
Volume: 15
Issue: 1
Page: 559-567
Publish at: 2026-02-01

Classification of Cihateup duck egg fertility using convolutional neural network EfficientNet-B3

10.11591/ijai.v15.i1.pp798-809
Evi Dewi Sri Mulyani , Teuku Mufizar , Dani Rohpandi , Ayu Djuliani , Egi Rahmatulloh , Rinaldi Satia Aulia Rahmat
Accurate detection of egg fertility is crucial to improve hatching success in duck farming. Conventional candling methods rely heavily on human expertise, making them subjective and error-prone. This study proposes an automated classification system for Cihateup duck egg fertility using candling images and a convolutional neural network (CNN) based on the EfficientNet-B3 architecture. Image enhancement techniques, including contrast limited adaptive histogram equalization (CLAHE), unsharp masking, and adaptive thresholding, were applied to improve image quality and feature visibility. The dataset consisted of fertile and infertile egg images captured at two incubation stages: the first 24 hours and the 8th–15th days. Data were split into training, validation, and testing sets with a ratio of 70:15:15. Experimental results show that image enhancement significantly improves classification performance. Without enhancement, the model achieved an accuracy of 49% with an area under curve (AUC) of 0.4226, indicating poor discrimination capability. With image enhancement, the proposed method achieved accuracies of 77% for the first 24 hours dataset and 80% for the 8th–15th days dataset, with AUC values of 0.9962 and 0.9317, respectively. These results demonstrate that EfficientNet-B3 combined with image enhancement provides an effective and computationally efficient solution for automated fertility detection of Cihateup duck eggs.
Volume: 15
Issue: 1
Page: 798-809
Publish at: 2026-02-01

Portable system for real-time traffic volume and speed estimation using YOLOv10

10.11591/ijai.v15.i1.pp300-309
Ida Bagus Sradha Nanda , Masrono Yugihartiman , Eko Primadi Hendri , I Made Suartika
Accurate traffic data is essential for effective transportation planning and policymaking. However, in many regions, especially those lacking intelligent infrastructure, data collection remains dependent on manual methods that are labor-intensive, time-consuming, and susceptible to human error. While advanced systems such as closed-circuit television (CCTV) and area traffic control systems (ATCS) offer automation, their high cost and infrastructure requirements limit widespread adoption. This study proposes a portable, low-cost, and real-time traffic monitoring system based on the YOLOv10 object detection algorithm. The system operates using only a smartphone-grade camera (1080 p, 60 fps) and a standard laptop, eliminating the need for expensive installations. It detects, classifies, and counts vehicles as they pass through a predefined region of interest (ROI), and also estimates their speed based on time–distance measurements. Field evaluations using five one-hour urban traffic videos showed excellent agreement with manual counts, achieving a mean absolute percentage error (MAPE) of just 0.30%. Speed estimation trials conducted on sample clips also demonstrated consistent and plausible results. These findings highlight the system’s potential as a scalable and accurate alternative for traffic monitoring in infrastructure-limited environments.
Volume: 15
Issue: 1
Page: 300-309
Publish at: 2026-02-01

Predicting university student dropouts in Latin America using machine learning

10.11591/ijai.v15.i1.pp628-641
Laberiano Andrade-Arenas , Inoc Rubio Paucar , Margarita Giraldo Retuerto , Cesar Yactayo-Arias
In the university context, student dropout has become one of the most recurring problems, both in the short and long term. The objective of this research was to develop a predictive model using the random forest (RF) algorithm to identify patterns associated with university dropout. To achieve this, the knowledge discovery in databases (KDD) methodology was applied, which encompasses the stages of selection, preprocessing, transformation, data mining, and interpretation of results. The RF model demonstrated superior performance compared to other evaluated models, achieving an accuracy of 87%, a precision of 86%, a recall of 85%, an F1-score of 85%, and an receiver operating characteristic (ROC) area under the curve (AUC) of 0.91, highlighting its high predictive capability compared to other techniques analyzed. Therefore, the application of the proposed model is recommended in various university institutions in order to identify potential dropout cases at an early stage.
Volume: 15
Issue: 1
Page: 628-641
Publish at: 2026-02-01

Change detection and classification of satellite images using convolutional neural network

10.11591/ijai.v15.i1.pp329-337
Raghavendra Srinivasaiah , Santosh Kumar Jankatti , Manjunath Ramanna Lamani , Niranjana Shravanabelagola Jinachandra
Satellite and airborne imagery, collectively known as earth observation imagery, are images of the earth collected from spaceborne or airborne platforms such as satellites and aircraft. Over the last 100 years, with the fast development of aviation, space exploration, and imaging technologies, the coming together of these technologies has been inevitable. Earth observation imagery has many applications in regional planning, geology, reconnaissance, fishing, meteorology, oceanography, agriculture, biodiversity conservation, forestry, landscape, intelligence, cartography, education, and warfare. With the rise in the number of these airborne and spaceborne imaging platforms being deployed by government and private entities alike, the capability to sift through and analyze vast amounts of data generated by these platforms is the need of the hour. With the exponential improvement in the computational capabilities of computers over the last half a century, analysts are exceedingly moving towards the practice of artificial intelligence, machine learning (ML), and computer vision solutions to automate a large part of the processes employed in analyzing earth observation imagery. This work recommends a workflow to perceive and classify changes in earth observation imagery of a given area by utilizing the vast flexibility that convolutional neural networks (CNN) provide.
Volume: 15
Issue: 1
Page: 329-337
Publish at: 2026-02-01

Comparison of image enhancement methods for pratima theft detection using artificial intelligence

10.11591/ijai.v15.i1.pp213-228
Made Sudarma , Ni Wayan Sri Ariyani , I Putu Agus Eka Darma Udayana , Ida Bagus Gde Pranatayana , Lie Jasa
The theft of pratima in Balinese temples threatens the spiritual and cultural balance of the community. These sacred objects, regarded as manifestations of God in Hinduism, hold profound religious significance, and their loss represents both material and spiritual desecration. To address this issue, this study investigates a security system that leverages image enhancement for low-light detection. Four techniques—contrast limited adaptive histogram equalization (CLAHE), adaptive histogram equalization (AHE), histogram equalization (HE), and gamma correction—were evaluated to improve image quality. CLAHE yielded the lowest mean squared error (MSE) of 21.16 and the highest peak signal-to-noise ratio (PSNR) of 38.13 dB. For object detection, VGG-19 and AlexNet were assessed. The best configuration, VGG-19 with HE, reached 83.33% accuracy and 93.75% recall, and achieved a receiver operating characteristic area under the curve (ROC AUC) of 0.90±0.02 across five runs. Thresholds derived from the ROC analysis were selected using the Youden J statistic to balance sensitivity and specificity. The approach outperformed lightweight and classical baselines in AUC, indicating superior discrimination under low illumination. These findings show that superior image quality does not always align with higher detection accuracy, and they highlight the importance of pairing effective enhancement with robust detectors for temple security. The study contributes practical insights for preserving Balinese cultural and spiritual heritage by strengthening efforts to protect pratima against theft.
Volume: 15
Issue: 1
Page: 213-228
Publish at: 2026-02-01

Efficient data streaming in dynamic vehicular networks: a hybrid controller for seamless connectivity

10.11591/ijai.v15.i1.pp229-236
Prathibha Thimmappa , Mayuri Kundu
The demand for highly efficient data transmission is being increasingly demanded for dynamic vehicular networks, especially in the case of internet-of-vehicle (IoV). The current data transmission methods are known to encounter inefficiencies in terms of unreliable routing and restricted scalability. Evolving studies have found artificial intelligence (AI) based schemes more suitable to address these issues; however, there are no significant innovations towards developing a potential framework that can not only increase data transmission performance but also minimize the analytical overheads of AI. Hence, this paper presents a novel baseline framework by introducing an optimized controller structure at anchor points with the inclusion of novel ideologies of orientation degree and selection of mediating node. The proposed model witnesses 32.3 dB of signal quality, 857 kbps throughput, 81 ms delay, and 171 ms of response time, exhibiting much better performance in contrast to the frequently used data transmission method. The proposed model contributes to a solid foundation for any futuristic AI model for efficient and reliable data transmission in IoV.
Volume: 15
Issue: 1
Page: 229-236
Publish at: 2026-02-01

Brain tumor segmentation and classification using artificial hummingbird optimization algorithm

10.11591/ijai.v15.i1.pp429-442
Radhakrishnan Karthikeyan , Arappaleeswaran Muruganandham
The time and medical personnel experience are the only factors that determine whether brain tumors can be manually identified from numerous magnetic resonance imaging (MRI) pictures in medical practice. Many frameworks based on brain tumors are diagnosed using both deep learning and machine learning. This study proposes a Wasserstein deep convolutional generative adversarial network (WDCGAN) optimized using the artificial hummingbird optimization algorithm (AHBOA) for brain tumor segmentation and classification (SCBT). First, the BraTS dataset is used to gather the input data. Then it is pre-processed consuming adaptive self guided filtering (ASGF) and the result is segmented using fuzzy possibilistic C-ordered mean clustering (FPCOMC). After that, features are extracted using the dual tree complex discrete wavelet transform (DT-CDWT). The characteristics of feature extracted are fed to WDCGAN for effectively categorize the various parameters. Then the proposed MATLAB is used to implement the technique, and the performance measurements like F1-score, accuracy, error rate, precision, sensitivity, mean square error, receiver operating characteristic (ROC), and computational time are analyzed. The WDCGAN-AHBOA-SCBT method significantly improves precision in SCBT by integrating adaptive optimization strategies, resulting in 32.18, 32.75, and 32.90% higher precision in contrast to current techniques. This demonstrates that the approach is more accurate and effective, making it a reliable tool for medical diagnosis.
Volume: 15
Issue: 1
Page: 429-442
Publish at: 2026-02-01

Enhanced framework for detecting Vietnamese hate and offensive spans

10.11591/ijai.v15.i1.pp962-971
Dinh-Hong Vu , Tuong Le
The rise of hate and offensive content on social media platforms, such as Facebook and Twitter, has emerged as an escalating concern, especially in Vietnam. Consequently, detecting hate and offensive spans in Vietnamese text is an essential area of research. This study introduces ViHateOff, an advanced framework that combines a hated speech dictionary (HSD) automatically constructed from the Vietnamese hate and offensive spans (ViHOS) dataset with the pre-trained language models for Vietnamese (PhoBERT)-large language model to enhance the detection of offensive expressions. The framework functions through two primary modules. First, it constructs an HSD from the ViHOS dataset, which serves as a reference for identifying hate and offensive language in Vietnamese text. Second, the framework integrates the PhoBERT-large language model with HSD, enhancing the detection of harmful words in the input text. Experimental results demonstrate that the proposed framework significantly outperforms existing state-of-the-art (SOTA), achieving an F1-score of 0.8693 on the all spans subset and 0.8709 on the multiple-spans subset representing relative improvements of over 10% compared to the strongest baseline.
Volume: 15
Issue: 1
Page: 962-971
Publish at: 2026-02-01

Pneumonia classification from chest X-rays using significant feature selection and machine learning

10.11591/ijai.v15.i1.pp592-603
Yugandhar Chodagam , Manjunatha Hiremath
The chest X-ray images of normal lungs differ only subtly from those of lungs with pneumonia, making image-based diagnosis highly challenging. To address this issue, we developed a machine learning (ML)-based, lightweight, end-to-end Python package that processes chest X-ray images, implements robust feature selection methods, and classifies the images using various algorithms. While many studies have focused on improving classification accuracy using newer methods, few have addressed the interpretability of the extracted features or the growing computational demands of complex models. We used four publicly available datasets and extracted first-order, textural, and transform-based radiomic features to test our package. Features were selected using the Shapley additive explanations (SHAP) combined with recursive feature elimination (RFE) and stability selection algorithms. Our final solution contains a method that extracts a finite set of features identified by stability selection and feeds them as inputs into classical ML algorithms. Our model achieved 98% accuracy on the primary dataset, and 97%±1, 96%±2, and 94%±2% accuracy on the other three datasets. Our approach is fast, self-contained, and requires only an ideal set of features, making it suitable for resource-constrained clinical environments.
Volume: 15
Issue: 1
Page: 592-603
Publish at: 2026-02-01
1 2 3
...
1994 1995
Show 41 of 1995

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration