Saeed Mostafaei

ORCID: 0000-0003-3722-5011
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Imbalanced Data Classification Techniques
  • Machine Learning and Data Classification
  • Anomaly Detection Techniques and Applications
  • Superconducting and THz Device Technology
  • Electricity Theft Detection Techniques
  • Data Mining Algorithms and Applications
  • Radio Frequency Integrated Circuit Design
  • Seismic Performance and Analysis
  • Structural Health Monitoring Techniques
  • Microwave Engineering and Waveguides
  • Explainable Artificial Intelligence (XAI)
  • Seismic Waves and Analysis

Islamic Azad University Kerman
2024

Islamic Azad University of Kermanshah
2024

Amirkabir University of Technology
2020-2023

In this paper, a compact and simple structure of an elliptic microstrip lowpass filter (LPF) is designed for harmonic suppression in microwave quadrature hybrid coupler (QHC) applications. A radial resonator rectangular are used to produce LPF. The proposed LPF on the outer sides branch line coupler, which has improved suppression. Furthermore, artificial neural networks (ANNs) incorporated improve design process. best obtained using ANN model. size, only occupies 16.4 mm × 7.3 equals 0.164...

10.1155/2024/8722642 article EN cc-by Active and Passive Electronic Components 2024-03-11

In this paper, it is aimed to perform clustering analysis on a large set of near-fault ground motions based the intensity velocity pulses. For purpose, pulses these are extracted asymmetric Gaussian chirplet model (AGCM), adapted dictionary-free orthogonal matching pursuit (ADOMP) algorithm, and Newton method. The location first AGCM atom considered as real pulse, its combination with adjacent atoms reconstructs pulse motions. addition, hierarchical k-means used cluster into two, three, five...

10.1061/(asce)em.1943-7889.0001845 article EN Journal of Engineering Mechanics 2020-07-30

In two-class classification problems, due to the bias of machine learning algorithms towards majority class, from imbalanced data is one most challenging tasks. some studies, it has been shown that imbalance problem not only reason and other factors related nature data, such as small disjunct borderline rare examples, overlap between classes, shift are main reasons for this difficulty. To improve performance algorithms, here in paper, a hybrid sampling method proposed. First, feature space...

10.2139/ssrn.4379932 article EN 2023-01-01
Coming Soon ...