Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection

Permanent URI for this collectionhttps://hdl.handle.net/11147/7148

Browse

Filters

2010 - 201912020 - 20264

Settings

Scopus Q: search.filters.scopusQuartile.Q2Subject: search.filters.subject.Clustering

Search Results

Now showing 1 - 5 of 5
  • Article
    FW-S3PFCM: Feature-Weighted Safe-Semi Possibilistic Fuzzy C-Means Clustering
    (Springer, 2026) Khezri, Shirin; Aghazadeh, Nasser; Hashemzadeh, Mahdi; Golzari Oskouei, Amin
    The safe semi-supervised fuzzy c-means clustering (S3FCM) method is a well-known clustering method that can produce successful results by incorporating prior knowledge of the class distribution. Its process is fast and simple but still has two limitations. The first issue is that it gives equal weight to all data features, while in real-world applications, different features usually have different importance. Secondly, S3FCM is very sensitive to noise and outliers. This paper proposes an extension of the S3FCM, entitled FW-S3PFCM, to mitigate these shortcomings. The proposed method uses a local feature weighting scheme to consider the different feature weights in the clustering process. Additionally, a possibilistic version of the S3FCM is designed to reduce the sensitivity to noise and outliers. The effectiveness of the proposed method is comprehensively evaluated on various benchmark datasets, and its performance is compared with the state-of-the-arts methods. To practically asses the FW-S3FCM, a real-world dataset of brain MRI images and its segmentation performance are analyzed as well. The average Accuracy, F1-score, Sensitivity, and Precision measures obtained by FW-S3FCM are 0.9682, 0.9826, 0.9743, and 0.9925, respectively, which are better than the competitors' performance.
  • Article
    A Robust Possibilistic Semi-Supervised Fuzzy Clustering Algorithm With Neighborhood-Aware Feature Weighting
    (Springer Heidelberg, 2025) Moghaddam, Arezou Najafi; Aghazadeh, Nasser; Hashemzadeh, Mahdi; Oskouei, Amin Golzari
    The Semi-Supervised Fuzzy C-Means (SSFCM) method integrates class distribution information with fuzzy logic to overcome the challenges of semi-supervised clustering methods. While the inclusion of label information in the objective function improves the quality of the clustering method, semi-supervised fuzzy techniques still encounter important limitations, including (1) sensitivity to noise and outliers, (2) uniform feature importance, (3) neglecting the influences of neighborhood in the clustering process. In this paper, an improved semi-supervised clustering algorithm is presented to address these challenges. First, the algorithm reduces the sensitivity to noise and outliers by integrating the possibilistic fuzzy C-means algorithm into the SSFCM method. Second, a dynamic feature weighting method assigns different weights to the features in each cluster, which improves the performance of the algorithm in imbalanced datasets. Third, the proposed algorithm introduces a neighborhood mechanism that incorporates the neighbor's trade-off weighting and feature weighting strategy considering a strong metric. Finally, a robust kernel metric is used to further improve the performance on complex and nonlinear datasets. Extensive experiments are conducted on several benchmark datasets to evaluate the performance of the proposed method. The results show that the proposed method outperforms the current state-of-the-art techniques. The implementation source codes of the proposed method are publicly available at https://github.com/Amin-Golzari-Oskouei/RPSSFC-NAFW.
  • Article
    Citation - Scopus: 5
    Event Distortion-Based Clustering Algorithm for Energy Harvesting Wireless Sensor Networks
    (Springer, 2022) Al-Qamaji, A.; Atakan, B.
    Wireless sensor networks (WSNs) consist of compact deployed sensor nodes which collectively report their sensed readings about an event to the Base Station (BS). In WSNs, due to the dense deployment, sensor readings can be spatially correlated and it is nonessential to transmit all their readings to the BS. Therefore, for more energy efficient, it is vital to choose which sensor node should report their sensed readings to the BS. In this paper, the event distortion-based clustering (EDC) algorithm is proposed for the spatially correlated sensor nodes. Here, the sensor nodes are assumed to harvest energy from ambient electromagnetic radiation source. The EDC algorithm allows the energy-harvesting sensor nodes to select and eliminate nonessential nodes while maintain an acceptable level of distortion at the BS. To measure the reliability, a theoretical framework of the distortion function is first derived for both single-hop and two-hop communication scenarios. Then, based on the derived theoretical framework, the EDC algorithm is introduced. Through extensive simulations, the performance of the EDC algorithm is evaluated in terms of achievable distortion level, number of alive nodes and harvested energy levels. As a result, EDC algorithm can successfully exploit both the spatial correlation and energy harvesting to improve the energy efficiency while preserving an acceptable level of distortion. Furthermore, the performance comparisons reveal that the two-hop communication model outperforms the single-hop model in terms of the distortion and energy-efficiency. © 2021, The Author(s).
  • Article
    Citation - WoS: 3
    Citation - Scopus: 5
    Event Distortion-Based Clustering Algorithm for Energy Harvesting Wireless Sensor Networks
    (Springer, 2021) Al-Qamaji, Ali; Atakan, Barış
    Wireless sensor networks (WSNs) consist of compact deployed sensor nodes which collectively report their sensed readings about an event to the Base Station (BS). In WSNs, due to the dense deployment, sensor readings can be spatially correlated and it is nonessential to transmit all their readings to the BS. Therefore, for more energy efficient, it is vital to choose which sensor node should report their sensed readings to the BS. In this paper, the event distortion-based clustering (EDC) algorithm is proposed for the spatially correlated sensor nodes. Here, the sensor nodes are assumed to harvest energy from ambient electromagnetic radiation source. The EDC algorithm allows the energy-harvesting sensor nodes to select and eliminate nonessential nodes while maintain an acceptable level of distortion at the BS. To measure the reliability, a theoretical framework of the distortion function is first derived for both single-hop and two-hop communication scenarios. Then, based on the derived theoretical framework, the EDC algorithm is introduced. Through extensive simulations, the performance of the EDC algorithm is evaluated in terms of achievable distortion level, number of alive nodes and harvested energy levels. As a result, EDC algorithm can successfully exploit both the spatial correlation and energy harvesting to improve the energy efficiency while preserving an acceptable level of distortion. Furthermore, the performance comparisons reveal that the two-hop communication model outperforms the single-hop model in terms of the distortion and energy-efficiency.
  • Article
    Citation - WoS: 15
    Citation - Scopus: 21
    Tracking Fast Moving Targets in Wireless Sensor Networks
    (Institution of Electronics and Telecommunication Engineers, 2010) Alaybeyoğlu, Ayşegül; Erciyeş, Kayhan; Kantarcı, Aylin; Dağdeviren, Orhan
    We propose a dynamic distributed algorithm for tracking objects that move fast in a sensor network. In the earlier efforts in tracking moving targets, the current leader node at time t predicts the location only for time t + 1 and if the target moves in high speed, it can pass by a group of nodes very fast without being detected. Therefore, as the target increases its speed, the probability of missing that target also increases. In this study, we propose a target tracking system that predicts future k locations of the target and awakens the -corresponding leader nodes so that the nodes along the trajectory self organize to form the clusters to collect data related to the target in advance and thus reduce the target misses. The algorithm first -provides detection of the target and forms a cluster with the neighboring nodes around it. After the selection of the cluster leader, the coordinates of the target is estimated using localization methods and cooperation -between the cluster nodes under the control of the leader node. The coordinates and the speed of the target are then used to estimate its trajectory. This information in turn provides the location of the nodes along the estimated trajectory which can be awaken, hence providing tracking of the moving object. We describe the algorithm, analyze its efficiency and show by simulations that it performs well to track very fast moving objects with speeds much higher than reported in literature.