Development of a Framework for Predicting Students' Academic Performance in STEM Education using Machine Learning Methods DOI Open Access

Rustam Abdrakhmanov,

Ainur Zhaxanova,

Malika Karatayeva

et al.

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Jan. 1, 2024

In the continuously evolving educational landscape, prediction of students' academic performance in STEM (Science, Technology, Engineering, Mathematics) disciplines stands as a paramount component for stakeholders aiming at enhancing learning methodologies and outcomes. This research paper delves into sophisticated analysis, employing Machine Learning (ML) algorithms to predict achievements, focusing explicitly on multifaceted realm education. By harnessing robust dataset drawn from diverse backgrounds, incorporating myriad factors such historical data, socioeconomic demographics, individual interactions, study innovates by transcending traditional parameters. The meticulously evaluates several machine models, juxtaposing their efficacies through rigorous methodologies, including Random Forest, Support Vector Machines, Neural Networks, subsequently advocating an ensemble approach bolster accuracy. Critical insights reveal that customized pathways, preemptive identification at-risk candidates, nuanced understanding contributing influencers are significantly enhanced ML framework, offering transformative lens strategies. Furthermore, confronts ethical quandaries challenges data privacy emerging wake advanced analytics education, proposing holistic guideline stakeholders. exploration not only underscores potential revolutionizing predictive strategies education but also advocates continuous model optimization, embracing symbiotic integration between pedagogical technological advancements, thereby redefining trajectories paradigms.

Language: Английский

Analyzing coverage area and channel bonding performance in 60 GHz MIMO systems for wireless local area networks DOI Creative Commons

Sisahu Tolcha Badane,

Tolcha Lemma Sime,

Gemechu Yadeta

et al.

Deleted Journal, Journal Year: 2025, Volume and Issue: 7(4)

Published: March 21, 2025

Language: Английский

Citations

0
A Transparent and Flexible Absorber for Electromagnetic Interference Suppression, Designed for 5G Communication and Sub-6G Applications DOI Open Access
Senfeng Lai,

Huiyuan Fu,

Junxian Tan

et al.

Electronics, Journal Year: 2025, Volume and Issue: 14(7), P. 1350 - 1350

Published: March 28, 2025

As 5G technology rapidly advances, the extension of spectrum into millimeter-wave bands enables higher data speeds and reduced latency. However, this frequency expansion introduces significant electromagnetic interference (EMI) issues, particularly in environments with dense equipment base stations. To tackle these challenges, paper presents a multilayer transparent ultra-wideband microwave absorber (MA) using indium tin oxide (ITO) that operates between 4 26 GHz. This optimized MA design successfully achieves absorption from 4.07 to 25.07 GHz, encompassing both Sub-6 GHz n258 bands, relative bandwidth 144% minimal thickness 0.129λL (where λL is free-space wavelength at lowest cutoff frequency). For TE TM polarization incidence angles ranging 0° 45°, demonstrates exceptional performance, maintaining exceeding 120%. Notably, for 60° 70°, can sustain an capacity greater than 100%. By integrating principles impedance matching, surface current theory, equivalent circuit simulation fitting, mechanism further analyzed, thereby confirming reliability design. offers wideband absorption, optical transparency, wide-angle characteristics, demonstrating great potential applications stealth, EMI suppression, compatibility (EMC) communications.

Language: Английский

Citations

0
Reconfigurable Intelligent Surface‐Empowered Non‐orthogonal Multiple Access: Outage and ABER Analysis of Smart and Blind Transmissions DOI
Vinoth Babu Kumaravelu,

Anjana Babu Sujatha,

Helen Sheeba John Kennedy

et al.

Published: May 9, 2025

Language: Английский

Citations

0
An Overview of Channel Modeling and Propagation Measurements in IRS‐Based Wireless Communication Systems DOI
Mohan Kumar,

Ammar Summaq,

Sunil Chinnadurai

et al.

Published: May 9, 2025

Language: Английский

Citations

0
Analysis of potential 5G transmission methods concerning Bit Error Rate DOI
Ali H. Abdulwahhab, Ali H. Abdulwahhab

AEU - International Journal of Electronics and Communications, Journal Year: 2024, Volume and Issue: 184, P. 155407 - 155407

Published: June 18, 2024

Language: Английский

Citations

3
State-of-the-Art Review of Deep Learning Methods in Fake Banknote Recognition Problem DOI Open Access
Ualikhan Sadyk, Rashid Baimukashev, Cemil Turan

et al.

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Jan. 1, 2024

In the burgeoning epoch of digital finance, exigency for fortified monetary transactions is paramount, underscoring need advanced counterfeit deterrence methodologies. The research paper provides an exhaustive analysis, delving into profundities employing sophisticated deep learning (DL) paradigms in battle against fiscal fraudulence through fake banknote detection. This comprehensive review juxtaposes traditional machine approaches with avant-garde DL techniques, accentuating conspicuous superiority latter terms accuracy, efficiency, and diminution human oversight. Spanning multiple continents currencies, discourse highlights universal applicability potency DL, incorporating convolutional neural networks (CNNs), recurrent (RNNs), generative adversarial (GANs) discerning most cryptic counterfeits, a feat unachievable by obsolete technologies. meticulously dissects architectures, processes, operational facets these systems, offering insights their strata, pooling heuristics, backpropagation, loss minimization algorithms, alluding to consequential roles feature extraction intricate pattern recognition - quintessentials authenticating banknotes. Furthermore, exploration broaches ethical privacy concerns stemming from including data bias over-reliance on technology, suggesting harmonization algorithmic advancements robust legislative frameworks. Conclusively, this seminal posits that while techniques herald revolutionary competence recognition, continuous research, multi-faceted strategies are imperative adapting ever-evolving chicanery malefactors.

Language: Английский

Citations

2
Advanced Decoding Methods for Massive-MIMO Systems Employing Deep Learning DOI
Shashank Kumar, Mangal Singh

Published: Jan. 11, 2024

Massive Multiple Input Output (MIMO) technology has a lot of potential for meeting wireless communication systems' growing need large data rates. This work presents novel decoder design that takes advantage deep learning techniques and is tailored Massive-MIMO systems. The conventional decoding methods encounter challenges in handling the massive scale antennas complex interference scenarios. To address these limitations, we propose an innovative architecture integrates models enhanced signal processing accuracy. By employing neural network-based strategies, our proposed system demonstrates significant improvements performance, achieving remarkable gains spectral efficiency error correction capabilities. Extensive simulations comparative analyses validate efficacy learning-driven decoder, showcasing its adaptability superior performance real-world environments.

Language: Английский

Citations

2
Tourist Tracking Techniques and Their Role in Destination Management: A Bibliometric Study, 2007–2023 DOI Open Access
Yarlein Ivama Julio Guerrero, Francisco Dias

Sustainability, Journal Year: 2024, Volume and Issue: 16(9), P. 3708 - 3708

Published: April 28, 2024

The tourism industry has recently undergone a significant evolution because of the implementation Information and Communication Technologies (ICT). These changes have raised new conceptions destinations, thus giving rise to smart destinations. In this context, digital tracking tourists become great interest. This article aims analyze research landscape on tourist through bibliometric study, which allows evaluation performance publications, recognition prominent SciVal topics, analysis techniques, their use as tool that contributes development Unlike prior studies in area, addition paper focuses only empirical tracking, identifies analyzes linkage thereof with Sustainable Development Goals (SDGs). study was carried out based articles extracted from Scopus database published between 2007 2023. is mainly quantitative, (journals, authors, institutions, countries) science mapping (citation analysis, co-word co-authorship analysis). Results show quality publications made 2007–2023 period indicate most used techniques been GPS devices, social media, app-based GPS/GNSS. addition, it found occurred at same time ICT topic covers largest number associated destination management. It concluded application transformation destinations into they provide useful relevant information market stakeholders for strategies improve decision-making allow sustainable management

Language: Английский

Citations

2
Development and analysis of small Multi Input Multi Output antenna with better isolation for Frequency Range‐2 5th Generation band applications DOI
Tej Raj, Ranjan Mishra, Ankush Kapoor

et al.

International Journal of Communication Systems, Journal Year: 2024, Volume and Issue: 37(16)

Published: July 8, 2024

Summary A simple planar wideband four‐port MIMO antenna of size 2.75𝜆 l × 0.04𝜆 (𝜆 is the wavelength when f 24.26 GHz) with high isolation that offers to 30.2 GHz bandwidth covering 5G FR‐2 spectrums proposed in this work. Total fractional 21.8% has been achieved covers n257 (26.5 29.5 GHz), n258 (24.25 27.5 and n261 (27.5 28.35 spectrums. Popular rectangular radiator‐based 4‐port symmetric stubs loaded parallel feed line tapered semi‐circular cut lower corners patch incorporated exhibits an excellent below −27 dB through entire spectrum extreme realized separation −40 dB. The projected radiator premeditated on RT/duroid 5880 0.508 mm depth permittivity (ε r ) 2.2. maximum radiation efficiency gain are 93.07% 6.48 dBi, respectively. parameters such as envelope correlation coefficient (ECC) were lesser than 0.003, diversity (DG) was around 10 dB, channel capacity loss (CCL) 0.15 bits/s/Hz, mean effective (MEG) between −3.6 −3 throughout impedance band, which exhibit properties, thus, appropriate for 5G/beyond wireless technologies.

Language: Английский

Citations

2
Performance Optimization of 5G–Satellite Integrated Networks for IoT Applications in Smart Cities: A Two-Ray Propagation Model Approach DOI Creative Commons
Mfonobong Uko, Sunday Ekpo, Sunday Enahoro

et al.

Smart Cities, Journal Year: 2024, Volume and Issue: 7(6), P. 3895 - 3913

Published: Dec. 11, 2024

The convergence of 5G terrestrial networks with satellite systems offers a revolutionary approach to achieving global, seamless connectivity, particularly for Internet Things (IoT) applications in urban and rural settings. This paper investigates the implications this 5G–satellite integrated network architecture, specifically through application two-ray propagation model free-space path loss (FSPL) model. By simulating signal characteristics over varying distances, altitudes, environmental parameters, we explore how factors such as transmitter height, altitude, frequency impact received power, loss, channel capacity, outage probability. key findings indicate that power decreases significantly increasing distance, notable oscillations due interference from ground reflections, evident within first 100 km. For example, at 50 km, 300 km altitude yields approximately −115 dBm while 1000 drops around −136 dBm. Higher frequencies (e.g., 32 GHz) exhibit greater than lower 24 GHz), 5 dB difference observed reinforcing need considerations long-range communication design. In terms bandwidth enhances achievable data rates but declines distance diminishing power. At MHz supports up 4500 Mbps, 3000 capacity Mbps. probability analysis shows higher signal-to-noise ratio (SNR) thresholds substantially increase likelihood failures, especially distances exceeding 2000 instance, 15 SNR threshold reaches 25%, compared less 5% threshold. These results underscore critical trade-offs designing IoT networks, balancing bandwidth, frequency, thresholds, altitudes optimal performance across diverse applications. provides valuable insights enhancing connectivity reliability remote underserved regions.

Language: Английский

Citations

2