International Journal of Informatics and Communication Technology (IJ-ICT)
Not a member yet
494 research outputs found
Sort by
Utilizing deep learning algorithms for the resolution of partial differential equations
Get PDFPartial differential equations (PDEs) are mathematical equations that are used to model physical phenomena around us, such as fluid dynamics, electrodynamics, general relativity, electrostatics, and diffusion. However, solving these equations can be challenging due to the problem known as the dimensionality curse, which makes classical numerical methods less effective. To solve this problem, we propose a deep learning approach called deep Galerkin algorithm (DGA). This technique involves training a neural network to approximate a solution by satisfying the difference operator, boundary conditions and an initial condition. DGA alleviates the curse of dimensionality through deep learning, a meshless approach, residue-based loss minimisation and efficient use of data. We will test this approach for the transport equation, the wave equation, the Sine-Gordon equation and the Klein-Gordon equation
Traffic accident classification using IndoBERT
Get PDFTraffic accidents are a widespread concern globally, causing loss of life, injuries, and economic burdens. Efficiently classifying accident types is crucial for effective accident management and prevention. This study proposes a practical approach for traffic accident classification using IndoBERT, a language model specifically trained for Indonesian. The classification task involves sorting accidents into four classes: car accidents, motorcycle accidents, bus accidents, and others. The proposed model achieves a 94% accuracy in categorizing these accidents. To assess its performance, we compared IndoBERT with traditional methods, random forest (RF) and support vector machine (SVM), which achieved accuracy scores of 85% and 87%, respectively. The IndoBERT-based model demonstrates its effectiveness in handling the complexities of the Indonesian language, providing a useful tool for traffic accident classification and contributing to improved accident management and prevention strategies
Extraction of association rules in a diabetic dataset using parallel FP-growth algorithm under apache spark
Get PDFThis research paper focuses on enhancing the frequent pattern growth (FP-growth) algorithm, an advanced version of the Apriori algorithm, by employing a parallelization approach using the Apache Spark framework. Association rule mining, particularly in healthcare data for predicting and diagnosing diabetes, necessitates the handling of large datasets which traditional methods may not process efficiently. Our method improves the FP-growth algorithmβs scalability and processing efficiency by leveraging the distributed computing capabilities of apache spark. We conducted a comprehensive analysis of diabetes data, focusing on extracting frequent itemsets and association rules to predict diabetes onset. The results demonstrate that our parallelized FP-growth (PFP-growth) algorithm significantly enhances prediction accuracy and processing speed, offering substantial improvements over traditional methods. These findings provide valuable insights into disease progression and management, suggesting a scalable solution for large-scale data environments in healthcare analytics.
Enhancing PI controller performance in grid-connected hybrid power systems
Get PDFThe optimal operation of a microgrid was buildup of both uncontrollable (solar, wind) and controllable (batteries, diesel generators) electrical energy sources are enclosed in this paper. By replacing controllers, the variations in wavering power supply caused by load fluctuations are managed. The objective of the research paper is to optimize these controller gain settings for effective use of electrical energy. In this paper integral time square error principle is combined along with the Cuckoo search algorithm (CSA) and particle swarm algorithm (PSA) to obtain the accurate, precise and appropriate results. It enhances the microgrid's steady-state sensitive responsiveness in comparison to trial-and-error techniques, assuring a stable supply of electricity to the load
Machine learning-driven design and performance analysis of microstrip antennas for sub-6 GHz/mm Wave 5G networks
Get PDFIn the realm of modern communication systems, antennas are crucial components, with the microstrip patch antenna being particularly notable for its low profile and seamless integration. Despite its widespread use, designing this antenna involves complex simulations to optimize parameters, requiring significant expertise and consuming considerable time and energy. To streamline this process, machine learning (ML) algorithms are being utilized. This paper introduces an innovative approach that employs ML techniques to design a rectangular microstrip patch antenna operating within the sub-6 GHz frequency range (1-6 GHz) and the millimeter frequency range (28-40 GHz). The antenna design maintains consistent patch dimensions positioned strategically at the center, with a thorough examination of patch length and width to enhance performance. Datasets are meticulously prepared, covering output parameters such as beam area, directivity, gain, and radiation efficiency across the specified frequency ranges. By employing various ML algorithms, this study conducts a comprehensive analysis to identify the most effective algorithm for accurately predicting antenna characteristics. The K-nearest neighbor (KNN) algorithm achieved high accuracy across all parameters: gain at 94.23% under sub-6 GHz and 95.93% under millimeter frequency range, directivity at 99.02% and 98.59%, radiation efficiency at 93.94% and 94.28%, and beam area at 99.07% and 98.59% respectively. These results optimize microstrip antenna designs and enhance understanding of the relationship between design parameters and performance outcomes with ML
Enhancement of liner materials based on nanomaterials to promote sustainability in noise intercourse
Get PDFDaily usage of devices has had a major influence on lives and existence, which would be unimaginable without them. Due to this, recent gadget dependability concerns need particular attention. PCs, hand mobile phones, and other computerized household gadgets need integrated circuits (ICs). Individual components must work together to accomplish their tasks and make the circuit operate. Hot carrier effect, oxide breakdown, and other system-level problems result from accommodating several devices in a planar IC. Vertical linking active components in one IC to another IC is a common method of three-dimensional IC integration (3D-IC). The main issue with 3D-IC adoption is electrical interference to neighboring through silicon via (TSV) and active transistors, which substantially reduces system performance. The electrical TSV (ETSV) model, which employs solely electrical signal carrying TSV, and the thermal TSV (TTSV) model, which incorporates thermal TSV during simulation, are used in this research to reduce electrical interference. The electrical signal transporting TSV to the substrate and other TSV was investigated for interference. With other models, this study also shows higher frequency regimes up to 1 THz. We found that the suggested methodology improves 3D-IC development by more than 30% by reducing electrical interference from signal-carrying TSV to other TSV
Design of a model for multistage classification of diabetic retinopathy and glaucoma
Get PDFThis study addresses the escalating prevalence of diabetic retinopathy (DR) and glaucoma, major global causes of vision impairment. We propose an innovative iterative Q-learning model that integrates with fuzzy C-means clustering to improve diagnostic accuracy and classification speed. Traditional diagnostic frameworks often struggle with accuracy and delay in disease stage classification, particularly in discerning complex features like exudates and veins. Our model overcomes these challenges by combining fuzzy C means with Q learning, enhancing precision in identifying key retinal components. The core of our approach is a custom-designed 45-layer 2D convolutional neural network (CNN) optimized for nuanced detection of DR and glaucoma stages. Compared to previous approaches, the performance on the IDRID and SMDG-19 datasets and associated samples shows a 10.9% rise in precision, an 8.5% improvement in overall accuracy, an 8.3% enhancement in recall, a 10.4% larger area under the curve (AUC), a 5.9% boost in specificity, and a 2.9% decrease in latency. This methodology has the potential to bring about significant changes in the field of DR and glaucoma diagnosis, leading to prompt medical interventions and possibly decreasing vision loss. The use of sophisticated machine learning techniques in medical imaging establishes a model for future investigations in ophthalmology and other clinical situations
Automated multi-document summarization using extractive-abstractive approaches
Get PDFThis study presents a multi-document text summarizing system that employs a hybrid approach, including both extractive and abstractive methods. The goal of document summarizing is to create a coherent and comprehensive summary that captures the essential information contained in the document. The difficulty in multi-document text summarization lies in the lengthy nature of the input material and the potential for redundant information. This study utilises a combination of methods to address this issue. This study uses the TextRank algorithm as an extractor for each document to condense the input sequence. This extractor is designed to retrieve crucial sentences from each document, which are then aggregated and utilised as input for the abstractor. This study uses bidirectional and auto-regressive transformers (BART) as an abstractor. This abstractor serves to condense the primary sentences in each document into a more cohesive summary. The evaluation of this text summarizing system was conducted using the ROUGE measure. The research yields ROUGE R1 and R2 scores of 41.95 and 14.81, respectively
A comprehensive analysis of dynamic PAPR reduction schemes in MIMO-OFDM systems
Get PDFInΒ this paper, an attempt develops three different methods, namely, Hybrid Maximal-Minimum (Max-Min) model with Decomposed Selective Mapping (D-SLM) in a UFMC, Modified Enhancement Asymmetric Arithmetic Coding Scheme (M-EAAC) and Dynamic Threshold-based Logarithmic Companding (DTLC) is carried out in Multiple-Input, Multiple-Output Orthogonal Frequency-Division Multiplexing (MIMO-OFDM) technology to enhance the PAPR reduction. These methods allow increased data rate request through threshold limit adjustment in a desired out-of-band (OOB) range, allows data transmission for the selected for the candidate sequences for maximizing the channel utility, data capacity and computational demands and varying threshold limit to analyse the nonlinear companding effect, respectively on D-UFMC-SLM, M-EAAC SCS-TI and DTLC. The extensive analysis shows that the proposed M-EAAC SCS-TI achieves a reduced CCDF PAPR, increased average spectral efficiency and redued Bit Error Rate (BER) than the other proposed DTLC and D-UFMC-SLM methods
One time pad for enhanced steganographic security using least significant bit with spiral pattern
Get PDFData is an important commodity in todayβs digital era. Therefore, data needs to get adequate security to prevent misuse. A common data security practice in the transmission of information is cryptography. Another approach is steganography, which hides secret messages in other media that are not confidential and can be accessed by the public. In this study, the spiral pattern is used for data placement using the least significant bit (LSB) method. Modifications were made to the 2-bits LSB to increase the data capacity that can be hidden. In order to increase security, the data is first converted into a datastream using random numbers as one time pad (OTP). exclusive-OR (XOR) operation is performed on datastream and OTP to get encrypted data to be hidden. The results showed that the image quality of the steganography results at a capacity close to 100% was still fairly good, as indicated by a peak signal-to-noise ratio (PSNR) value greater than 46 dB. Visually, the steganographic image does not look different from the original one. Likewise, the use of random numbers as OTP succeeded in changing the hidden data significantly, as indicated by the avalanche effect value above 50%