Article
Computer Science, Artificial Intelligence
K. Aditya Shastry, H. A. Sanjay
Summary: Data pre-processing is a technique that transforms raw data into a useful format for machine learning, with feature selection and feature extraction being significant components. This study proposes a hybrid strategy using modified Genetic Algorithm and weighted Principal Component Analysis for selecting and extracting features from agricultural datasets, resulting in significant improvements in benchmark and real-world farming datasets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Surani Matharaarachchi, Mike Domaratzki, Saman Muthukumarana
Summary: This article proposes an extended feature selection method which performs well in high dimensional classification problems. Simulated experiments demonstrate that the proposed method can effectively reduce the number of features in different datasets.
PEERJ COMPUTER SCIENCE
(2022)
Article
Engineering, Multidisciplinary
Zhaopeng He, Tielin Shi, Jianping Xuan
Summary: A novel deep learning method based on multi-sensor feature fusion was proposed for milling tool wear prediction. The method extracted signal features in different domains and used correlation analysis to determine the optimal features. Experimental results showed that the proposed method outperformed other comparative methods in predictive performance.
Article
Computer Science, Artificial Intelligence
Zhe Zhao, Bangyong Sun
Summary: The autoencoder (AE) based method has become essential in hyperspectral anomaly detection. However, due to the strong generalised capacity of AE, abnormal samples are often reconstructed along with the normal background samples. To effectively separate anomalies from the background, a memory-augmented autoencoder (MAENet) is proposed. The MAENet reduces the AE capability for abnormal sample reconstruction while maintaining the background reconstruction performance.
CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Zhengxin Li, Feiping Nie, Jintang Bian, Danyang Wu, Xuelong Li
Summary: In the field of data mining, unsupervised feature selection is an important topic. Spectral-based methods rely on similarity matrices to depict data structure, but noise features and computational cost can affect their performance. To address this, a simple and efficient unsupervised model is proposed using PCA and regularization to select discriminative features. Experimental results on synthetic and real-world data sets demonstrate the effectiveness of the proposed method.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Engineering, Multidisciplinary
Junxian Shen, Feiyun Xu
Summary: The health monitoring system for equipment is crucial for industrial production. This study proposes a feature selection and fusion method based on poll mode and optimized Weighted Kernel Principal Component Analysis (WKPCA) method to dig deeper for effective features and improve the separability of fault samples.
Article
Thermodynamics
Aosong Liang, Yunpeng Hu, Guannan Li
Summary: This study investigates the impact of different anomaly detection methods on chiller sensor fault detection and validates the results using an experimental dataset. The results show that anomaly detection methods can improve the quality of the original training data and KPCA has higher fault detection efficiency compared to PCA. IF-KPCA and Kmeans-KPCA further enhance the fault detection efficiency on top of KPCA.
INTERNATIONAL JOURNAL OF REFRIGERATION
(2023)
Article
Biochemical Research Methods
Yue Hu, Jin-Xing Liu, Ying-Lian Gao, Junliang Shang
Summary: A new method is proposed for feature selection, utilizing tensor robust principal component analysis with an introduced L-2, L-1- norm regularization term and solved using the ADMM algorithm. Experimental results demonstrate that this method outperforms other methods in filtering genes closely associated with disease.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Erick Odhiambo Omuya, George Onyango Okeyo, Michael Waema Kimwele
Summary: This study investigates the application of feature selection and classification in various fields, addressing the challenges of high dimensionality in datasets and the negative impact of irrelevant and redundant attributes on classification algorithms. To improve classification performance, a hybrid filter model based on principal component analysis and information gain is proposed and applied to machine learning techniques, demonstrating enhanced accuracy, precision, and recall.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Zahra Atashgahi, Ghada Sokar, Tim van der Lee, Elena Mocanu, Decebal Constantin Mocanu, Raymond Veldhuis, Mykola Pechenizkiy
Summary: A novel unsupervised feature selection method QuickSelection is proposed in this paper, which introduces the concept of neuron strength in sparse neural networks and combines it with sparsely connected denoising autoencoders to derive the importance of all input features. The method achieves the best trade-off of classification and clustering accuracy, running time, and maximum memory usage on benchmark datasets, with considerable speed increase, memory reduction, and the least amount of energy consumption compared to other state-of-the-art autoencoder-based feature selection methods.
Article
Computer Science, Artificial Intelligence
Qingyang Tan, Ling-Xiao Zhang, Jie Yang, Yu-Kun Lai, Lin Gao
Summary: This paper proposes a mesh-based variational autoencoder architecture that can handle meshes with irregular connectivity and nonlinear deformations. By introducing sparse regularization and spectral graph convolution operations, the visual quality and reconstruction ability of the extracted deformation components are improved. The paper also develops a neural shape editing method, achieving shape editing and deformation component extraction in a unified framework.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Information Systems
Hina Afzal, Furqan Rustam, Wajdi Aljedaani, Muhammad Abubakar Siddique, Saleem Ullah, Imran Ashraf
Summary: The emergence of fake job postings as a cyber-crime has posed alarming threats to both job seekers and companies. Despite the existence of machine learning-based approaches for detecting fake job posts, their accuracy is low and they show skewed performance on imbalanced data. This study addresses these limitations by using selective features through Chi-square and principal component analysis (PCA). The study also investigates the impact of dataset imbalance through the synthetic minority oversampling technique (SMOTE). The proposed model, which incorporates SMOTE with Chi-square-based selective features, achieves a 0.99 accuracy, outperforming individual machine learning models and existing state-of-the-art models. K-fold cross-validation confirms the robustness of the results.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Jintang Bian, Dandan Zhao, Feiping Nie, Rong Wang, Xuelong Li
Summary: In this article, a novel robust principal component analysis (RPCA) model is proposed to mitigate the impact of outliers and conduct feature selection simultaneously. By adopting sigma-norm as reconstruction error and applying l(2,0)-norm constraint to subspace projection, as well as proposing an efficient iterative optimization algorithm, robust reconstruction and feature selection are achieved.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Engineering, Industrial
Davide Cacciarelli, Murat Kulahci
Summary: Principal component analysis (PCA) is a commonly used unsupervised learning method with broad applications in both descriptive and inferential analytics. It is widely used for representation learning to extract key features from a dataset and visualize them in a lower dimensional space. This paper explores the relationship between PCA and autoencoders (AEs) and demonstrates their similarity in the case of linear AEs (LAEs) through examples.
QUALITY ENGINEERING
(2023)
Article
Computer Science, Information Systems
Hao Zheng, Liyong Fu, Qiaolin Ye
Summary: Robust principal component analysis (PCA) has been proven effective in data reconstruction and recognition tasks. However, existing methods often suffer from performance and robustness issues. To address this, we propose a new method called flexible capped PCA (FCPCA) that uses capped L2,p-norm distance metric to minimize reconstruction errors. Experimental results demonstrate that FCPCA outperforms existing methods in terms of power and flexibility.
INFORMATION SCIENCES
(2022)
Article
Mathematical & Computational Biology
Abhigyan Nath, S. Karthikeyan
INTERDISCIPLINARY SCIENCES-COMPUTATIONAL LIFE SCIENCES
(2017)
Article
Biotechnology & Applied Microbiology
Abhigyan Nath, Karthikeyan Subbiah
Article
Computer Science, Artificial Intelligence
Anoop Kumar Tiwari, Abhigyan Nath, Karthikeyan Subbiah, Kaushal Kumar Shukla
INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE
(2017)
Article
Biology
Abhigyan Nath, S. Karthikeyan
JOURNAL OF THEORETICAL BIOLOGY
(2018)
Article
Computer Science, Artificial Intelligence
Abhigyan Nath, Karthikeyan Subbiah
Article
Biology
Abhigyan Nath
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2019)
Article
Biology
Abhigyan Nath, Gopal Krishna Sahu
JOURNAL OF THEORETICAL BIOLOGY
(2019)
Article
Biotechnology & Applied Microbiology
D. C. Mishra, Poonam Sikka, Sunita Yadav, Jyotika Bhati, S. S. Paul, A. Jerome, Inderjeet Singh, Abhigyan Nath, Neeraj Budhlakoti, A. R. Rao, Anil Rai, K. K. Chaturvedi
Article
Veterinary Sciences
Poonam Sikka, Abhigyan Nath, Shyam Sundar Paul, Jerome Andonissamy, Dwijesh Chandra Mishra, Atmakuri Ramakrishna Rao, Ashok Kumar Balhara, Krishna Kumar Chaturvedi, Keerti Kumar Yadav, Sunesh Balhara
FRONTIERS IN VETERINARY SCIENCE
(2020)
Article
Biochemistry & Molecular Biology
Manish Kumar Tripathi, Abhigyan Nath, Tej P. Singh, A. S. Ethayathulla, Punit Kaur
Summary: The accumulation of massive data in Cheminformatics databases has made big data and artificial intelligence indispensable in drug design. The development of newer algorithms and architectures has fulfilled the specific needs of various drug discovery processes, while deep learning neural networks have resulted in a paradigm shift in chemical information mining.
MOLECULAR DIVERSITY
(2021)
Article
Biology
Abhigyan Nath
Summary: The inefficiency of current antivirals and the resistance of viruses have led to the demand for novel antiviral agents. Antiviral peptides show promise as a potential avenue for developing effective antiviral drugs, with the ability to halt the progression of viral infections.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2021)
Article
Biotechnology & Applied Microbiology
Roopshikha Sahu, Amisha Yadav, Abhigyan Nath
Summary: This study developed a machine learning-based prediction model for MRTD of anti-retroviral drugs. Through feature selection algorithm and representative training/testing set, a subset of top features was extracted, achieving good predictive performance.
MINERVA BIOTECHNOLOGY AND BIOMOLECULAR RESEARCH
(2021)
Article
Biochemical Research Methods
Abhigyan Nath, Andre Leier
BMC BIOINFORMATICS
(2020)
Article
Biology
Abhigyan Nath, S. Karthikeyan
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2017)
Article
Biology
Kunal Bhattacharya, Shikha Mahato, Satyendra Deka, Nongmaithem Randhoni Chanu, Amit Kumar Shrivastava, Pukar Khanal
Summary: Chemoresistance, a major challenge in cancer treatment, is associated with the cellular glutathione-related detoxification system. A study has identified GSTP1 enzyme as critical in the inactivation of anticancer drugs and suggests the need for GSTP1 inhibitors to combat chemoresistance. Through molecular docking and simulations, the study found that quercetin 7-O-beta-D-glucoside showed promise as a potential candidate for addressing chemoresistance in cancer patients.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2024)
Article
Biology
Manwi Shankar, Majji Sai Sudha Rani, Priyanka Gopi, P. Arsha, Prateek Pandya
Summary: This study investigates the interaction between the food dye BBY and the serum protein BSA. The results show that BBY binds to a specific site on BSA through hydrophobic interactions, affecting the structural stability of the protein. These findings enhance our understanding of the molecular-level interactions between BBY and BSA.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2024)
Article
Biology
Chi Zhang, Qian Gao, Ming Li, Tianfei Yu
Summary: In this study, we propose a graph neural network-based autoencoder model, AGraphSAGE, that effectively predicts protein-protein interactions across diverse biological species by integrating gene ontology.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2024)
Article
Biology
Kangjie Wu, Liqian Xu, Xinxiang Li, Youhua Zhang, Zhenyu Yue, Yujia Gao, Yiqiong Chen
Summary: Named Entity Recognition (NER) is a crucial task in natural language processing (NLP) and big data analysis, with wide application range. This paper proposes an improved neural network method for NER of rice genes and phenotypes, which can learn semantic information in the context without feature engineering. Experimental results show that the proposed model outperforms other models.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2024)
Article
Biology
Suman Hait, Sudip Kundu
Summary: Interactions between amino acids in proteins are crucial for stability and structural integrity. Thermophiles have more and more stable interactions to survive in extreme environments. Different types of interactions are enriched in different structural regions.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2024)
Article
Biology
Kountay Dwivedi, Ankit Rajpal, Sheetal Rajpal, Virendra Kumar, Manoj Agarwal, Naveen Kumar
Summary: This study aims to identify biomarkers for non-small cell lung cancer (NSCLC) using copy number variation (CNV) data. A novel deep learning architecture, XL1R-Net, is proposed to improve the classification accuracy for NSCLC subtyping. Twenty NSCLC-relevant biomarkers are uncovered using explainable AI (XAI)-based feature identification. The results show that the identified biomarkers have high classification performance and clinical relevance. Additionally, twelve of the biomarkers are potentially druggable and eighteen of them have a high probability of predicting NSCLC patients' survival likelihood according to the Drug-Gene Interaction Database and the K-M Plotter tool, respectively. This research suggests that investigating these seven novel biomarkers can contribute to NSCLC therapy, and the integration of multiomics data and other sources will help better understand NSCLC heterogeneity.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2024)
Article
Biology
Pengli Lu, Wenqi Zhang, Jinkai Wu
Summary: Researchers have developed a computational method, AMPCDA, to predict circRNA-disease associations using predefined metapaths, achieving high predictive accuracy. This method effectively combines node embeddings with higher-order neighborhood representations and provides valuable guidance for revealing new disease mechanisms in biological research.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2024)