International Journal of Informatics and Communication Technology (IJ-ICT)
Not a member yet
494 research outputs found
Sort by
Advancements in brain tumor classification: a survey of transfer learning techniques
Get PDFThis survey article presents a critical review of the state-of-the-art transfer learning (TL) methodologies applied in the field of brain tumor classification, with a special emphasis on their various contributions and associated performance metrics. We will discuss various pre-processing approaches, the underlying fine-tuning strategies, whether used purely or in an end-to-end training manner, and multi-modal applications. The current study specifically highlights the application of VGG16 and residual network (ResNet) methods for feature extraction, demonstrating that leveraging highorder features in magnetic resonance imaging (MRI) images can enhance accuracy while reducing training. We further analyze fine-tuning methods in relation to their role in optimizing model layers for small, domain-specific datasets, finding them particularly effective in enhancing performance on the small brain tumor dataset. It will look into end-to-end training, which means fine-tuning models that have already been trained on large datasets to make them better. It will also present multimodal TL as a way to use both MRI and computed tomography (CT) scan data to get better classification results. Comparing different pre-trained models can provide a better understanding of the strengths and weaknesses associated with the particular brain tumor classification task. This review aims to analyze the advancements in TL for medical image analysis and explore potential avenues for future research and development in this crucial field of medical diagnostics
Advancing semiconductor integration: 3D ICs and Perylene-N as superior liner material for minimizing TSV clamour coupling
Get PDFThe semiconductor industry faces substantial challenges with planar integration (2D ICs), prompting a significant shift towards vertical IC integration, known as three-dimensional IC (3D ICs). This deliberate slant not only amplifies bandwidth and boosts system action but also effectively reduces power consumption through scaling. 3D ICs intricately coordinate IC chips by vertically stacking them and establishing electrical connections using through silicon vias (TSVs). TSV clamour coupling emerges as a critical factor influencing system performance, particularly between signalcarrying TSVs (ETSV) and victim TSVs. This study showcases significant advancements in electrical integrity by effectively minimizing clamour coupling from TSVs to the silicon substrate. This is achieved through the application of CMOS-compatible dielectric materials as liner structures. Various proposed structures have been meticulously analyzed across an assortment of parameters, encompassing electrical signals and high frequencies. Moreover, the study rigorously investigates clamour coupling across different types of TSVs, including ETSV, thermal TSV (TTSV), and heat sources. Perylene-N emerges as a standout performer among the tested liner materials, demonstrating superior clamour coupling performance across all proposed models, even at higher frequencies such as THz. In this study a novel dielectric material Perylene-N compared with the conventional SiO2 (silicon dioxide). Notably, Perylene-N exhibited a remarkable 33 dB improvement in noise coupling performance at terahertz (THz) frequencies. The results were thoroughly verified and validated in the research work
Automated rice leaf disease detection using artificial intelligence deep learning
Get PDFAs one of the top five rice-producing countries, India relies heavily on rice for both economic management and food needs. To ensure healthy rice plant growth, early detection of diseases and timely treatment are essential. Since manual disease detection is time-consuming and labor-intensive, an automated approach is more practical. This work presents a deep neural network (DNN)-based artificial intelligence (AI) method for recognizing rice leaf diseases. The method detects three common diseases: leaf smut, bacterial leaf blight, and brown spot, as well as healthy images. The approach uses an AI-based attention network and semantic batch normalized DeepNet (AN-SBNDN) combined with a channel attention mechanism to improve disease detection accuracy. Experiments with rice leaf datasets and comparison with conventional networks like residual attention network (Res ATTEN) and dynamic speeded up robust features (DSURF) validate the effectiveness of the method. Key performance metrics include average accuracy, time, precision, and recall, achieved at 21%, 44%, 26%, and 31%, respectively
Parameter-optimized routing protocols for targeted broadcast messages in smart campus environments
Get PDFThe spread of handheld mobile devices integrated with multiple sensors makes it easy for these devices to interact with each other. These interactions are useful in a variety of applications such as monitoring and notification systems that can be adopted in smart campuses. The performance of these applications depends primarily on the network infrastructure and network protocols. In cases of failure, smart campus requires the provision of effective alternatives that can handle essential services. Hence, this work uses the Wi-Fi mobile ad hoc network (MANET) as an alternative backup to the traditional infrastructure. The dynamic nature of such a network relies on individuals' mobility, this leads to a lack of end-to-end connectivity. To overcome this challenge, delay-tolerant networking (DTN) has been adopted as its primary approach to routing information inside campus. Spray and wait, binary spray and wait (BSW), and probabilistic flooding protocols are deeply assessed to ensure sustained communications in the working environment. The protocols’ parameters are comprehensively investigated and optimized. Moreover, the performance metrics that are used in the evaluation are messages consumption, node responsiveness, and coverage. The findings showed that the optimal protocol and its parameters is reliant upon the specific application and resources available
A survey on ransomware detection using AI models
Get PDFData centers and cloud environments are compromised as they are at great risk from ransomware attacks, which attack data integrity and security. Through this survey, we explore how AI, especially machine learning and deep learning (DL), is being used to improve ransomware detection capabilities. It classifies ransomware types, highlights active groups such as Akira, and evaluates new DL techniques effective at real-time data analysis and encryption handling. Feature extraction, selection methods, and essential parameters for effective detection, including accuracy, precision, recall, F1-score and receiver operating characteristic (ROC) curve, are identified. The findings point to the state of the art and the state of the art in AI based ransomware detection and underscore the need for robust, real-time models and collaborative research. The statistical and graphical analyses help researchers and practitioners understand existing trends and directions for future development of efficient ransomware detection systems to strengthen cybersecurity in data centers and cloud infrastructures
Predictive model for converting leads into repeat order customer using machine learning
Get PDFIn the competitive business landscape, customer relationship management (CRM) is pivotal for managing customer relationships. Lead generation and customer retention are critical aspects of CRM as they contribute to sustaining business growth and profitability. Also, identifying and converting leads into repeat customers is essential for optimizing revenue and minimizing promotional costs. This study focuses on developing a predictive model using machine learning techniques to convert leads into repeat order customers in conventional businesses. Leveraging data from a motorcycle distribution company in Jakarta and Tangerang, the study compares the performance of various models for predicting repeat orders. This includes individual models like DeepFM, random forest, and gradient boosting decision tree models. Additionally, it explores the effectiveness of stacking these models using logistic regression as a meta-learner. Furthermore, the study implements backward feature elimination for feature selection and hyperband for hyperparameter tuning to enhance model performance. The results indicate that Stacking model using base model default configuration stands out as the most robust, achieving the highest scores in accuracy (0.95), area under the curve receiver-operating characteristic curve (AUC-ROC) (0.67), log loss (0.19), weighted average precision (0.95), weighted average recall (0.95), and weighted average F1- score (0.92), effectively handling the imbalanced dataset
Collaborative singular value decomposition with user-item interaction expansion for first-time user and item recommendations
Get PDFIn today's digital landscape, recommendation systems are essential for delivering personalized content and improving user engagement across various platforms. However, a key challenge known as the cold-start problem—where limited user-item interaction data hampers the ability to generate accurate recommendations—remains a significant obstacle, particularly for new users and items. To address this issue, this paper introduces an enhanced methodology combining collaborative singular value decomposition (Co-SVD) with an innovative approach to reduce data sparsity. The objective of this research is to improve recommendation accuracy in sparse data environments by leveraging collaborative information in the user-item interaction matrix. Extensive experiments conducted on an e-commerce dataset validate the superiority of the proposed Enhanced Co-SVD model over traditional Co-SVD, content-based filtering, and random recommendation methods across multiple metrics. Our approach demonstrates particular strength in cold-start scenarios, providing precise recommendations with minimal user interaction data. These findings have important implications for e-marketing, personalized user experiences, and overall business success in online environments
The integration of discrete contourlet transform in OFDM framework for future wireless communication
Get PDFIn the upcoming era, the forthcoming sixth-generation (6G) wireless communication network will demand highly efficient technology to support extensive capacity, ultra-high speeds, low latency, scalability, and adaptability. While the current fifth-generation (5G) wireless communication system relies on OFDM technology, the evolution towards a beyond 5G wireless communication system necessitates a new OFDM framework. This study introduces a novel OFDM system that integrates the discrete Contourlet transform. A comparative analysis has been conducted among the proposed system, conventional OFDM, and curvelet-based OFDM systems. The results indicate that the proposed system offers improvements in bit error rate (BER), reduced computational complexity, decreased peak-to-average power ratio (PAPR), and enhanced power spectrum density (PSD) when contrasted with both the traditional and curvelet-based systems
Planar hexagonal patch multiple input multiple output 4x4 antenna for UWB applications
Get PDFThe combination of Multiple Input Multiple Output (MIMO) antennas and Ultra-Wideband (UWB) technology offers several advantages, including reduced interference, improved isolation, and optimized dual paths. These benefits extend the range and enhance signal quality. However, designing UWB-MIMO antennas presents challenges, such as achieving low mutual coupling for high isolation and creating small-sized antennas suitable for portable devices while being effective for UWB frequencies in a MIMO configuration. The proposed antenna is a 4x4 planar MIMO antenna with a hexagon-shaped patch, a partial ground plane featuring an inverted L-stub on the left side, and a plus-shaped slot in the centre ground. It has dimensions of 32 x 32 x 1.6 mm³ and is capable of achieving a wide bandwidth of 3-12.5 GHz. The antenna's performance measurements are impressive: return loss below -10 dB at frequencies of 3-12.5 GHz, mutual coupling below -16.5 dB, Envelope Correlation Coefficient (ECC) bellow 0.005, Diversity gain of more than 9.97, Total Active Reflection Coefficient (TARC) below -10 dB. Based on these results, the proposed antenna offers excellent performance for UWB applications, featuring high efficiency, minimal interference between antenna elements, and optimal diversity performance
Enhancing database query interpretation: a comparative analysis of semantic parsing models
Get PDFThe rapid proliferation of NoSQL databases in various domains necessitates effective parsing models for interpreting NoSQL queries, a fundamental aspect often overlooked in database management research. This paper addresses the critical need for a comprehensive understanding of existing semantic parsing models tailored for NoSQL query interpretation. We identify inherent issues in current models, such as limitations in precision, accuracy, and scalability, alongside challenges in deployment complexity and processing delays. This review is pivotal, shedding light on the intricacies and inefficiencies of existing systems, thereby guiding future advancements in NoSQL database querying. This methodical comparison of these models across key performance metrics-precision, accuracy, recall, delay, deployment complexity, and scalability-reveals significant disparities and areas for improvement. By evaluating these models against both individual and combined parameters, we identify the most effective methods currently available. The impact of this work is far-reaching, providing a foundational framework for developing more robust, efficient, and scalable parsing models. This, in turn, has the potential to revolutionize the way NoSQL databases are queried and managed, offering significant improvements in data retrieval and analysis. Through this paper, we aim to bridge the gap between theoretical model development and practical database management, paving the way for enhanced data processing capabilities in diverse NoSQL database applications