Article
Ecology
Binh Thai Pham, Abolfazl Jaafari, Tran Van Phong, Davood Mafi-Gholami, Mandis Amiri, Nguyen Van Tao, Van-Hao Duong, Indra Prakash
Summary: In this study, a spatially explicit ensemble modeling framework was developed to estimate groundwater potential in Kon Tum Province, Vietnam, using different ensemble learning techniques. The ensemble models outperformed the single NB model in terms of mapping accuracy, with the RFNB model showing the highest accuracy. Feature selection identified key variables for explaining groundwater potential distribution in the region. The proposed methodology and potential maps can assist managers in aligning water use patterns and developing sustainable groundwater management strategies.
ECOLOGICAL INFORMATICS
(2021)
Article
Physics, Multidisciplinary
Shouta Sugahara, Maomi Ueno
Summary: Previous research has shown that the classification accuracies of Bayesian networks obtained by maximizing the conditional log likelihood were higher than those obtained by maximizing the marginal likelihood. However, in cases with small sample sizes and a class variable with multiple parents, the accuracies of exact learning with ML were significantly lower. Introducing an exact learning augmented naive Bayes classifier improved the situation and guaranteed similar class posterior estimation as exact learning Bayesian networks.
Article
Engineering, Biomedical
S. Nanglia, Muneer Ahmad, Fawad Ali Khan, N. Z. Jhanjhi
Summary: Breast cancer, common in both men and women, is difficult to detect in early stages and can be costly and complex to treat, leading to high fatality rates. This paper introduces a heterogeneous ensemble machine learning approach for early detection of breast cancer.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Computer Science, Artificial Intelligence
Anderson Ara, Francisco Louzada
Summary: The main goal of this paper is to introduce a new procedure for a naive Bayes classifier, called alpha skew Gaussian naive Bayes (ASGNB), which utilizes a flexible generalization of the Gaussian distribution on continuous variables. ASGNB is capable of handling asymmetry or bimodal behavior in the data and outperforms other traditional classification methods in terms of predictive performance.
INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING
(2022)
Article
Engineering, Biomedical
Chayashree Patgiri, Amrita Ganguly
Summary: Detection of anomalous cells in blood diseases is crucial, and automatic recognition with robust segmentation and classification methods can improve efficiency. A novel hybrid segmentation method using features extracted from cells for training and testing classifiers shows potential for high performance in classifying normal and sickle cells.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Computer Science, Software Engineering
Gias Uddin, Yann-Gael Guehenuc, Foutse Khomh, Chanchal K. Roy
Summary: Sentiment analysis in software engineering has the potential to support diverse development activities, but current tools may not be fully satisfactory in terms of accuracy. The combination of stand-alone sentiment detectors for fault detection has shown better performance, but there is no such approach for sentiment detection in software artifacts. This study explores the feasibility of developing an ensemble engine by combining the polarity labels of stand-alone sentiment detectors in the field of software engineering. The results show that the existing tools can complement each other, but a majority voting-based ensemble does not improve the accuracy. The developed tool, Sentisead, combining polarity labels and bag of words, outperforms the individual tools. The use of advanced language-based pre-trained transformer models further improves the infrastructure of Sentisead.
ACM TRANSACTIONS ON SOFTWARE ENGINEERING AND METHODOLOGY
(2022)
Article
Computer Science, Information Systems
Wysterlanya K. P. Barros, Matheus T. Barbosa, Leonardo A. Dias, Marcelo A. C. Fernandes
Summary: This work proposes a fully parallel hardware architecture for the Naive Bayes classifier and evaluates it on an FPGA. Experimental results show that the proposed implementation performs well in terms of speed and power consumption compared to other state-of-the-art works.
Article
Genetics & Heredity
Samuel Anyaso-Samuel, Archie Sachdeva, Subharup Guha, Somnath Datta
Summary: This study utilized microbiome samples from urban environments to predict the geographical location of unknown samples, implemented multiple classifiers and a robust ensemble approach, and highlighted the unreliability of relying on a single classification algorithm for metagenomic samples. By combining several classifiers via ensemble approach, the study achieved classification results comparable to the best-performing component classifier.
FRONTIERS IN GENETICS
(2021)
Article
Engineering, Environmental
Ba-Quang-Vinh Nguyen, Yun-Tae Kim
Summary: The study compares the performances of 5 machine learning techniques in predicting landslide probability in Japan and Korea. The CNN model performed the best, while the NB model performed the worst. Statistical tests confirmed the significance of all classified landslide susceptibility maps and differences between maps generated by different ML models.
BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT
(2021)
Article
Environmental Sciences
Sushant K. Singh, Robert W. Taylor, Biswajeet Pradhan, Ataollah Shirzadi, Binh Thai Pham
Summary: This study evaluates the performance of different machine learning models in predicting preferences for sustainable arsenic mitigation. The results show that a Gaussian distribution-based Naive Bayes classifier performs the best, while linear classifiers underperform. Nonlinear or ensemble classifiers can better understand the complex relationships in socio-environmental data and provide accurate and robust prediction models. In cases of limited data, Gaussian Naive Bayes is the best option.
ECOTOXICOLOGY AND ENVIRONMENTAL SAFETY
(2022)
Article
Computer Science, Artificial Intelligence
Amgad M. Mohammed, Enrique Onieva, Michal Wozniak, Gonzalo Martinez-Munoz
Summary: This article discusses the strategy of classifier ensemble pruning, involving optimizing predefined performance criteria to identify subensembles. The study analyzes a set of heuristic metrics to guide the pruning process, with results indicating that ordered aggregation is an effective strategy for improving predictive performance and reducing computational complexities.
PATTERN RECOGNITION
(2022)
Article
Green & Sustainable Science & Technology
Gyeongmin Kim, Jin Hur
Summary: Renewable-power-generating resources offer unlimited clean energy with minimal air pollutants and greenhouse gases, in contrast to fossil fuels which contribute to environmental pollution and climate change. The global capacity of renewable power is increasing, but predicting the output of wind resources remains a key challenge. This study introduces an ensemble prediction model for wind power based on augmented naive Bayes classifiers, which shows lower error rates compared to single prediction models when applied to empirical data from a wind farm in Jeju Island, South Korea.
Article
Agriculture, Dairy & Animal Science
Rafael N. Watanabe, Priscila A. Bernardes, Elieder P. Romanzini, Larissa G. Braga, Thais R. Brito, Ronyatta W. Teobaldo, Ricardo A. Reis, Danisio P. Munari
Summary: Monitoring animal activity in production systems is crucial for obtaining information on animal health, production, and reproduction. This study evaluated the use of accelerometers to predict grazing behavior of Nelore cattle and found that the Random Forest algorithm, combined with resampling techniques, achieved high accuracy in classifying various behaviors. Knowledge of animal behavior can provide insights into their well-being, health, and productivity, making accelerometers a valuable tool for continuous animal monitoring.
Article
Computer Science, Artificial Intelligence
Hongjia Ren, Qiulin Guo
Summary: This paper introduces a Tree Augmented Naive Bayes classifier (TAN) widely used in machine learning and data mining. To enhance the flexibility and classification performance of TAN, a Flexible Tree Augmented Naive Bayes classifier (FTAN) is proposed, which measures attribute dependencies using mutual information contribution rate and filters out weak dependencies through a flexible threshold adjustment. Experimental results on UCI datasets demonstrate the considerable advantages of FTAN in terms of 0-1 loss and class probability root mean square error. The application of FTAN in predicting the favorable distribution area for remaining oil and gas resources in the Junggar Basin shows its effectiveness and superiority, providing a decision-making basis for optimizing drilling strategies and exploration targets.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Hakan Ezgi Kiziloz
Summary: This study formally compares different classifier ensemble methods in the feature selection domain and finds that ensemble methods outperform single classifiers, albeit with longer execution time, and are more effective in minimizing the number of features.
Article
Computer Science, Artificial Intelligence
Mariam Zomorodi-moghadam, Moloud Abdar, Zohreh Davarzani, Xujuan Zhou, Pawel Plawiak, U. Rajendra Acharya
Summary: Coronary artery disease is a major cause of mortality worldwide, and modern computer-aided approaches such as particle swarm optimization have been used for disease prediction and diagnosis. A proposed method for discovering CAD classification rules based on real-world data aims to produce accurate and effective rules for disease detection.
Article
Physics, Multidisciplinary
Mahboobeh Houshmand, Zahra Mohammadi, Mariam Zomorodi-Moghadam, Monireh Houshmand
INTERNATIONAL JOURNAL OF THEORETICAL PHYSICS
(2020)
Article
Neurosciences
Zeinab Sherkatghanad, Mohammadsadegh Akhondzadeh, Soorena Salari, Mariam Zomorodi-Moghadam, Moloud Abdar, U. Rajendra Acharya, Reza Khosrowabadi, Vahid Solari
FRONTIERS IN NEUROSCIENCE
(2020)
Review
Computer Science, Information Systems
Mohamed Hammad, Rajesh N. V. P. S. Kandala, Amira Abdelatey, Moloud Abdar, Mariam Zomorodi-Moghadam, Ru San Tan, U. Rajendra Acharya, Joanna Plawiak, Ryszard Tadeusiewicz, Vladimir Makarenkov, Nizal Sarrafzadegan, Abbas Khosravi, Saeid Nahavandi, Ahmed A. Abd EL-Latif, Pawel Plawiak
Summary: This paper reviews the state-of-the-art machine and deep learning based CAAC expert systems for shockable ECG signal recognition, discussing their strengths, advantages, and drawbacks. Moreover, unique bispectrum and recurrence plots are proposed to represent shockable and non-shockable ECG signals. Deep learning methods are usually more robust and accurate than standard machine learning methods but require big data of good quality for training.
INFORMATION SCIENCES
(2021)
Article
Physics, Multidisciplinary
Davood Dadkhah, Mariam Zomorodi, Seyed Ebrahim Hosseini
Summary: In this study, a new approach was proposed to optimize the teleportation cost in Distributed Quantum Circuits (DQCs) by replacing equivalent circuits with a heuristic approach and using a genetic algorithm to optimize the placement of qubits. Results showed promising outcomes for the proposed method.
INTERNATIONAL JOURNAL OF THEORETICAL PHYSICS
(2021)
Article
Physics, Multidisciplinary
Omid Daei, Keivan Navi, Mariam Zomorodi
Summary: Distributed quantum systems are a well-known method to increase processing power in the quantum processing world by optimizing communication between limited-capacity quantum circuits. This study introduces an efficient method based on quantum gate commuting to reduce the number of teleportations required in performing distributed quantum circuits (DQCs), resulting in a significant decrease in teleportations and execution time.
INTERNATIONAL JOURNAL OF THEORETICAL PHYSICS
(2021)
Article
Quantum Science & Technology
Ismail Ghodsollahee, Zohreh Davarzani, Mariam Zomorodi, Pawel Plawiak, Monireh Houshmand, Mahboobeh Houshmand
Summary: This paper introduces a novel quantum computation model based on matrix representation, proposing a new approach to reduce teleportation cost in distributed quantum circuits. The method includes two phases for optimization, effectively decreasing the cost of teleportations.
QUANTUM INFORMATION PROCESSING
(2021)
Article
Quantum Science & Technology
Moein Sarvaghad-Moghaddam, Mariam Zomorodi
Summary: This paper explores the implementation of n-qubit controlled-U gates in distributed quantum computation, proposing a general protocol for remote implementation with minimal resources, and applying it to implement Toffoli gates in bipartite and tripartite systems. By considering the layout and resource allocation of qubits in network subsystems, the method further optimizes the implementation of gates.
QUANTUM INFORMATION PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Mohammad Beheshti Roui, Mariam Zomorodi, Masoomeh Sarvelayati, Moloud Abdar, Hamid Noori, Pawel Plawiak, Ryszard Tadeusiewicz, Xujuan Zhou, Abbas Khosravi, Saeid Nahavandi, U. Rajendra Acharya
Summary: This paper introduces a novel approach for generating classification rules based on evolutionary computation, with custom crossover and mutation operators for GPU execution, and leveraging parallelism to enhance fitness function performance. Experimental results demonstrate high accuracy and speedup for HCV, Poker, and COVID-19 datasets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Multidisciplinary Sciences
Zohreh Davarzani, Mariam Zomorodi, Mahboobeh Houshmand
Summary: This paper proposed a multi-layer hierarchical architecture for distributing quantum computation in a distributed quantum computing (DQC) system using teleportation. The two-level hierarchical optimization method aims to minimize the number of communications among subsystems by distributing qubits among different parts effectively. Experimental results demonstrated that the proposed approach outperforms previous methods in terms of efficiency.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Information Systems
Davood Dadkhah, Mariam Zomorodi, Seyed Ebrahim Hosseini, Pawel Plawiak, Xujuan Zhou
Summary: This paper proposes a new approach to reduce the teleportation cost and execution time in distributed quantum circuits. By using a combination of genetic algorithm and modified tabu search algorithm, the quantum circuit is partitioned to optimize the number of teleportations and reduce execution time. The proposed approach achieves significant improvements in benchmark circuits and surpasses previous methods.
Article
Computer Science, Information Systems
Mina Abbaszade, Vahid Salari, Seyed Shahin Mousavi, Mariam Zomorodi, Xujuan Zhou
Summary: This paper develops compositional vector-based semantics of positive transitive sentences using Q-NLP and proposes a protocol based on Q-LSTM for sentence translation. The method generalizes to use quantum circuits of sentences as inputs for translation, paving the way for quantum neural machine translation that may demonstrate quadratic speedup and better accuracy over classical methods.
Article
Physics, Multidisciplinary
Omid Daei, Keivan Navi, Mariam Zomorodi-Moghadam
INTERNATIONAL JOURNAL OF THEORETICAL PHYSICS
(2020)