Review
Computer Science, Information Systems
Jianghui Cai, Jing Hao, Haifeng Yang, Xujun Zhao, Yuqing Yang
Summary: Semi-supervised clustering (SSC) is a technique that integrates semi-supervised learning and clustering analysis to improve clustering performance by incorporating prior information. This paper provides a comprehensive review of SSC, organized into different categories and discusses their performance, suitable scenarios, and ways to add supervising information. It also summarizes successful applications of SSC in various fields and provides application caveats and development trends. This review and analysis of SSC can benefit researchers in providing an overall understanding, research topics, and analysis of existing methods.
INFORMATION SCIENCES
(2023)
Article
Biochemical Research Methods
Qiao Ning, Zedong Qi, Yue Wang, Ansheng Deng, Chen Chen
Summary: Glutarylation plays a crucial role in cell functions, and accurately identifying substrates and sites is a major challenge. A new FCCCSR algorithm is proposed to select reliable negative samples from unlabeled data, combined with feature extraction to enhance the prediction accuracy of glutarylation sites.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Computer Science, Information Systems
Phung The Huan, Pham Huy Thong, Tran Manh Tuan, Dang Trong Hop, Vu Duc Thai, Nguyen Hai Minh, Nguyen Long Giang, Le Hoang Son
Summary: Data partition with high confidence has been a major focus of researchers in Soft Computing for many years. Safe semi-supervised fuzzy clustering has been widely used to tackle this problem, but it often takes a long time and may produce unreasonable results. In this research, a new algorithm called TS3FCM is proposed to address the computational time issue by finding trusted labeled data and performing semi-supervised fuzzy clustering in isolated processes. The key contributions of the paper include a new objective function and a new semi-supervised fuzzy clustering model. Experimental results show that TS3FCM runs faster while maintaining reasonable clustering quality compared to the related algorithms.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Avgoustinos Vouros, Eleni Vasilaki
Summary: This study addresses the problem of data clustering with unidentified feature quality and limited labelled data. A K-Means variant is proposed to combine unsupervised sparse clustering and semi-supervised methods, which effectively identifies informative features from uninformative ones and achieves high performance on synthetic and real world datasets.
PATTERN RECOGNITION LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Tianshu Yang, Nicolas Pasquier, Frederic Precioso
Summary: A novel semi-supervised consensus clustering algorithm is proposed in this article, which utilizes closed pattern mining technique to generate a recommended consensus solution without the need for inputting the number of generated clusters k, and can improve the quality of clustering results.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Chemistry, Multidisciplinary
Mona Suliman Alzuhair, Mohamed Maher Ben Ismail, Ouiem Bchir
Summary: In this paper, a novel semi-supervised deep clustering approach named SC-DEC is proposed to address the limitations exhibited by existing semi-supervised clustering approaches. The proposed approach leverages a deep neural network architecture to generate fuzzy membership degrees that better reflect the true partition of the data. Experimental results show that utilizing minimal previous knowledge about the data can improve the overall clustering performance.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Jingnan Li, Chuan Lin, Ruizhang Huang, Yongbin Qin, Yanping Chen
Summary: This paper proposes an intention-guided deep semi-supervised document clustering model called IGSC. IGSC uses a deep metric learning network to address the limitations of traditional deep semi-supervised clustering models and utilizes an intention matrix to guide the clustering process, resulting in improved clustering performance that aligns with the user's intention.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Computer Science, Information Systems
Fariba Salehi, Mohammad Reza Keyvanpour, Arash Sharifi
Summary: The study introduced a new algorithm called SMKFC-ER, which focuses on external knowledge related to labeled data and combines entropy and relative entropy for semi-supervised multiple kernel fuzzy clustering. The use of relative entropy and entropy helps the semi-supervised section share more consistent concepts, control cluster fuzziness, and regularly determine kernel weights for the unsupervised section.
INFORMATION SCIENCES
(2021)
Article
Automation & Control Systems
Tran Manh Tuan, Mai Dinh Sinh, Tran Dinh Khang, Phung The Huan, Tran Thi Ngan, Nguyen Long Giang, Vu Duc Thai
Summary: This paper proposes an improvement to semi-supervised fuzzy clustering methods by using multiple fuzzifiers. The proposed models show higher performance compared to related models, as evaluated on different datasets.
INTERNATIONAL JOURNAL OF FUZZY SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Raphael Elimeli Nuhoho, Chen Wenyu, Adu Asare Baffour
Summary: This paper introduces a new semi-supervised learning network that optimizes the model by integrating unlabelled data, reducing the reliance on low-level features. Experimental analysis shows that this method performs better in image classification tasks and has strong generalization ability.
Article
Computer Science, Interdisciplinary Applications
Tue Boesen, Eldad Haber, G. Michael Hoversten
Summary: A new graph-Laplacian based semi-supervised clustering method is proposed in this research, which is capable of handling large datasets and yielding satisfactory results in oil prospectivity analysis.
COMPUTERS & GEOSCIENCES
(2021)
Article
Engineering, Electrical & Electronic
Zhaorui Zhu, Quanxue Gao
Summary: With the increasing interest in multi-view clustering due to the diversity of data modalities, this paper proposes a valid semi-supervised multi-view spectral clustering algorithm. By incorporating prior knowledge, utilizing tensor minimization, and applying cannot-link constraints, the algorithm outperforms current methods in terms of stability and accuracy. Experimental results on various datasets demonstrate the algorithm's effectiveness and potential applications.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Review
Computer Science, Information Systems
Kamal Taha
Summary: Retrieving, analyzing, and processing large data can be challenging. Semi-supervised and un-supervised learning methods are more advantageous than supervised learning methods. This paper provides a review and taxonomy of over 200 state-of-the-art algorithms for semi-supervised and un-supervised learning.
INFORMATION SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Yao Ma, Hongbo Shi, Shuai Tan, Yang Tao, Bing Song
Summary: In this article, a consistency regularization autoencoder (CRAE) framework based on encoder-decoder network is proposed to address the problem caused by limited labeled samples. The experiments show that the proposed method is effective for process fault diagnosis when labeled samples are limited.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Information Systems
Mohammad Misbahuddin, Syed Mustafa Ali Zaidi
Summary: This paper presents a semi-supervised machine learning approach utilizing clustering to distinguish malicious attacks from normal traffic in network data. By extracting features and applying supervised learning algorithms, the study achieved successful classification of attacks and normal traffic.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2021)
Article
Chemistry, Multidisciplinary
Muhammad Nadeem, Ali Arshad, Saman Riaz, Syeda Wajiha Zahra, Ashit Kumar Dutta, Sultan Almotairi
Summary: This paper introduces a network architecture that utilizes the efficient technique of semi-supervised clustering to design an intrusion detection system. By observing users' responses inside and outside the cloud server, different rules and mechanisms are enforced to prevent attacks on the cloud server. The research conducts experiments in different scenarios and compares the results with other papers.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Muhammad Nadeem, Ali Arshad, Saman Riaz, Syeda Wajiha Zahra, Ashit Kumar Dutta, Sultan Almotairi
Summary: This paper discusses cloud computing and data security, and proposes a secure architecture and algorithm implementation to prevent replay attacks.
APPLIED SCIENCES-BASEL
(2022)
Review
Chemistry, Multidisciplinary
Afia Zafar, Muhammad Aamir, Nazri Mohd Nawi, Ali Arshad, Saman Riaz, Abdulrahman Alruban, Ashit Kumar Dutta, Sultan Almotairi
Summary: This paper provides an understanding of the importance and role of pooling layers in convolutional neural networks (CNNs). Pooling layers reduce the resolution of feature maps, decrease spatial dimensions, minimize computational costs, and prevent overfitting.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Muhammad Nadeem, Ali Arshad, Saman Riaz, Syeda Wajiha Zahra, Ashit Kumar Dutta, Abdulrahman Alruban, Badr Almutairi, Sultan Almotairi
Summary: This paper investigates the issue of data protection in cloud computing, proposing an efficient encryption algorithm to enhance data security and address the challenge of attackers decrypting encrypted data. The research utilizes a two-layer encryption approach to improve data security. A comparison of different studies and technologies is conducted, leading to conclusions based on the results.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Syeda Wajiha Zahra, Ali Arshad, Muhammad Nadeem, Saman Riaz, Ashit Kumar Dutta, Zaid Alzaid, Rana Alabdan, Badr Almutairi, Sultan Almotairi
Summary: This paper presents the principles and methods of cloud cryptography, which protect data through bit-reversing and salting mechanisms, and encrypt data using Caesar cipher and cipher matrix algorithms. By comparing with other algorithms, it demonstrates the superiority of the proposed algorithm.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Ali Arshad, Muhammad Nadeem, Saman Riaz, Syeda Wajiha Zahra, Ashit Kumar Dutta, Zaid Alzaid, Rana Alabdan, Badr Almutairi, Sultan Almotairi
Summary: There are techniques and algorithms available for detecting attacks on cloud data, but they cannot protect the data from attackers. Cloud cryptography provides a secure and reliable way to transmit data. This paper develops various data security techniques to completely protect data from attackers.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Muhammad Nadeem, Ali Arshad, Saman Riaz, Syeda Wajiha Zahra, Ashit Kumar Dutta, Moteeb Al Moteri, Sultan Almotairi
Summary: This paper presents an efficient algorithm to protect data from invaders and prevent data misuse. By converting plain text to ASCII and using a Non-Deterministic Bit Generator to generate a key, the key is XORed with the plain text and bitwise toggling is applied. Then, an efficient matrix cipher encryption algorithm is developed to encrypt the data. The key can only decrypt the data with the correct key, ensuring the data security.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Muhammad Nadeem, Ali Arshad, Saman Riaz, Syeda Wajiha Zahra, Shahab S. Band, Amir Mosavi
Summary: Many organizations focus on protecting cloud servers from external attacks, but the majority of risks come from internal sources. While there are algorithms in place to safeguard against attacks, hackers constantly find ways to bypass these security measures. Cloud cryptography provides the best data protection algorithm for secure data exchange between authentic users.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Muhammad Nadeem, Ali Arshad, Saman Riaz, SyedaWajiha Zahra, Muhammad Rashid, Shahab S. Band, Amir Mosavi
Summary: Cloud computing is an attractive and cost-saving model that offers online services to end-users, allowing them to access data from any node. However, cloud security is a major concern due to various malware attacks from internal and external sources. This paper proposes a tool that uses Cloudflare and K-nearest neighbors (KNN) classification to prevent spamming attacks on cloud servers. Cloudflare blocks attacker's IP addresses, while KNN classifiers identify the location of spammers. The article also discusses various prevention techniques, compares with other studies, and draws conclusions based on different results.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Saman Riaz, Ali Arshad, Shahab S. Band, Amir Mosavi
Summary: Personality is characterized by individuals' patterns of feeling, thinking, and behaving. Predicting personality from small video series is an exciting area of research in computer vision. Most existing research has achieved preliminary results in extracting extensive knowledge from visual and audio modalities. To address this limitation, the Deep Bimodal Fusion (DBF) approach is proposed, which predicts five traits of personality using a combination of visual and audio information. The proposed framework utilizes modified convolutional neural networks for visual analysis and employs long short-term memory models to analyze audio representations. By independently determining traits and combining them through weighted fusion, the proposed approach achieves a mean accuracy score of 0.9183, outperforming previous frameworks.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Afia Zafar, Muhammad Aamir, Nazri Mohd Nawi, Ali Arshad, Saman Riaz, Abdulrahman Alruban, Ashit Kumar Dutta, Badr Almutairi, Sultan Almotairi
Summary: Convolutional neural networks are widely used in computer vision and have shown effectiveness in solving image processing problems. Designing the network structure for higher accuracy requires adjusting hyperparameters, which is time-consuming and requires domain knowledge. To overcome this, we propose an evolutionary algorithm-based approach that dynamically enhances the structure of CNNs using optimized hyperparameters, resulting in superior classification accuracy compared to previous methods.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Wanghu Chen, Meilin Zhou, Chenhan Zhai, Mengyang Shen, Pengbo Lv, Ali Arshad
Summary: In this paper, the combination of Generative Adversarial Network (GAN) with Ensemble Learning is introduced for anomaly detection in high-dimensional sparse data. Experimental results show that this approach outperforms traditional GAN-based methods in improving AUC and compared favorably with other representative anomaly detection approaches.
2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)
(2021)
Article
Computer Science, Information Systems
Muhammad Nadeem, Ali Arshad, Saman Riaz, Shahab S. Band, Amir Mosavi
Summary: This article discusses the security issues of cloud computing and defense mechanisms, focusing on monitoring the attack rate of the network using an Intrusion Detection System, and providing various solutions to protect the cloud server from attacks.
Article
Computer Science, Information Systems
Saman Riaz, Ali Arshad, Licheng Jiao
Article
Computer Science, Information Systems
Ali Arshad, Saman Riaz, Licheng Jiao