Article
Computer Science, Theory & Methods
Padraig Cunningham, Sarah Jane Delany
Summary: The article provides an overview of Nearest Neighbour classification techniques, focusing on similarity assessment mechanisms, computational issues in identifying nearest neighbours, and methods for reducing the dimension of the data. New sections on similarity measures for time-series, retrieval speedup, and intrinsic dimensionality have been added, along with an Appendix containing Python code for key methods.
ACM COMPUTING SURVEYS
(2021)
Article
Computer Science, Artificial Intelligence
Francisco J. Castellanos, Jose J. Valero-Mas, Jorge Calvo-Zaragoza
Summary: The k-nearest neighbor rule is well-known for its high performance and versatility, but it also suffers from low efficiency due to the exhaustive search required for each new query. To address this issue, data reduction techniques have been used, but their application to complex structural data has been limited. A new adaptation of the reduction through homogeneous clusters algorithm for string data shows significant improvements in classification performance and reduction rates compared to the set-median version.
Article
Computer Science, Hardware & Architecture
Huru Hasanova, Muhammad Tufail, Ui-Jun Baek, Jee-Tae Park, Myung-Sup Kim
Summary: In this article, a machine learning based Sine Cosine Weighted K-Nearest Neighbour (SCA_WKNN) algorithm is proposed for heart disease prediction, which learns from data stored in blockchain. The proposed algorithm achieves higher accuracy compared to other algorithms. Blockchain-based storage also achieves higher throughput.
COMPUTERS & ELECTRICAL ENGINEERING
(2022)
Article
Astronomy & Astrophysics
Hewei Zhang, Qin Li, Yanxing Yang, Ju Jing, Jason T. L. Wang, Haimin Wang, Zuofeng Shang
Summary: Solar flares, particularly M- and X-class flares, are often associated with coronal mass ejections and are crucial sources of space weather effects. Forecasting these flares is vital to mitigate their destructive impacts. This study introduces statistical and machine-learning methods to predict the flare index of active regions, improving accuracy especially for large flare indexes. By ranking the importance of SHARP parameters using the Borda count method, the study provides insights into predicting solar flares.
ASTROPHYSICAL JOURNAL SUPPLEMENT SERIES
(2022)
Article
Computer Science, Artificial Intelligence
Suvita Rani Sharma, Birmohan Singh, Manpreet Kaur
Summary: The binary versions of Rao algorithms are proposed for solving feature selection problems in Parkinson's disease datasets, optimizing the k parameter of the k-nearest neighbour classifier. The performance of these algorithms is evaluated through 30 independent runs with a 10-fold cross-validation procedure and compared with state of the art methods, with significance analysis conducted using the Friedman rank test.
Article
Computer Science, Information Systems
S. A. R. Zaidi
Summary: This paper provides an overview of Nearest Neighbour (NN) methods, discussing their theoretical background, algorithms, implementations, and key applications. It also examines the challenges related to 5G and beyond wireless networks that can be addressed using NN classification techniques.
Article
Environmental Sciences
Yasin Wahid Rabby, Md Belal Hossain, Joynal Abedin
Summary: This study evaluates and compares the performance of three machine learning models (KNN, RF, and XGBoost) for landslide susceptibility mapping in Rangamati District, Bangladesh, and finds that XGBoost has the best performance.
GEOCARTO INTERNATIONAL
(2022)
Article
Biology
Marilou Bodde, Alex Makunin, Diego Ayala, Lemonde Bouafou, Abdoulaye Diabate, Uwem Friday Ekpo, Mahamadi Kientega, Gilbert Le Goff, Boris K. Makanga, Marc F. Ngangue, Olaitan Olamide Omitola, Nil Rahola, Frederic Tripet, Richard Durbin, Mara K. N. Lawniczak
Summary: The ANOSPP amplicon panel is used for large-scale monitoring of Anopheles species diversity. The NNoVAE method, which combines Nearest Neighbours (NN) and Variational Autoencoders (VAE), can accurately assign species identity by analyzing k-mers. In testing, NNoVAE exhibits high accuracy in classifying samples and identifying unexpected species.
Article
Transportation Science & Technology
Elham Saffari, Mehmet Yildirimoglu, Mark Hickman
Summary: This study aims to estimate the MFD for a large-scale urban network by combining probe vehicle data with an unknown penetration rate and full-scale approximate traffic data based on loop detector data. The Bayesian fusion method outperforms the baseline method in average flow and density estimations, especially showing significant improvement in average density estimations.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2022)
Article
Mathematics, Interdisciplinary Applications
Evgeny Levi, Radu Craiu
Summary: Scientists utilize large datasets to tackle complex problems and use approximate methods like Approximate Bayesian Computation (ABC) or Bayesian Synthetic Likelihood (BSL) to accelerate computation. However, the number of simulations required remains a limiting factor.
Article
Computer Science, Artificial Intelligence
Maciej Kusy, Piotr A. Kowalski
Summary: This paper presents a method for reducing the architecture of the probabilistic neural network (PNN) by clustering data and selecting nearest neighbors. Experimental results show that the reduced PNN achieves higher accuracy than the original network and existing methods in most classification tasks.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Murat Osmanoglu, Salih Demir, Bulent Tugrul
Summary: Cloud computing allows users to outsource databases and computing functionalities to avoid maintenance costs, providing universal data access. However, security and privacy concerns exist. Encrypting data can help overcome these concerns, but may impact operations. Collaborative approaches among cloud service providers may yield more accurate query results.
PEERJ COMPUTER SCIENCE
(2022)
Article
Computer Science, Information Systems
Krishna Gopal Sharma, Yashpal Singh
Summary: This article introduces the important research area of classification in machine learning, particularly binary classification. It presents a binary classifier called KDV, which is compared to the general classifier KNN to demonstrate its advantages. The article also examines the accuracy of KDV using various cross-validation methods, and suggests that KDV has research potential in the field of machine learning.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Shahadat Uddin, Ibtisham Haque, Haohui Lu, Mohammad Ali Moni, Ergun Gide
Summary: This paper studies different variants of the k-nearest neighbour (KNN) algorithm and compares their performance in disease prediction. By implementing and experimenting on eight datasets, the study found that accuracy values ranged from 64.22% to 83.62%, with Hassanaat KNN showing the highest accuracy. The study also proposes a relative performance index based on accuracy, precision, and recall measures, identifying Hassanaat KNN as the best performing variant.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Artificial Intelligence
Abdul Wahid, Annavarapu Chandra Sekhara Rao
Summary: This paper proposes a Relative Density-based Outlier Factor algorithm for identifying outliers, which analyzes test points through two stages. Experimental results show that the algorithm has higher rank power than baseline methods on real-world datasets.
Article
Education & Educational Research
Juan Ramon Rico-Juan, Cristina Cachero, Hermenegilda Macia
Summary: The study found that collaborative peer assessment significantly improved students' self-assessment accuracy, while peer assessment scores were more accurate and improved with the number of assessments received. Instructors need to balance between students' improved understanding and time constraints when deciding on the design of assessment activities.
ASSESSMENT & EVALUATION IN HIGHER EDUCATION
(2022)
Article
Computer Science, Artificial Intelligence
Antonio Javier Gallego, Juan Ramon Rico-Juan, Jose J. Valero-Mas
Summary: The paper introduces the caKD+ algorithm which combines various techniques to improve the efficiency of kNN search, outperforming 16 state-of-the-art methods on 10 datasets.
PATTERN RECOGNITION
(2022)
Article
Health Care Sciences & Services
Julio Cabrero-Garcia, Juan Ramon Rico-Juan, Antonio Oliver-Roig
Summary: The study found that the Global Activity Limitation Indicator (GALI) is closely associated with multiple participation domains and performs differently across different age groups, but not with gender. The relative importance of participation domains also varies among different age groups. Compared to self-rated health, GALI shows a better ability to reflect restrictions in multiple participation domains.
QUALITY OF LIFE RESEARCH
(2022)
Article
Computer Science, Interdisciplinary Applications
Antonio Oliver-Roig, Juan Ramon Rico-Juan, Miguel Richart-Martinez, Julio Cabrero-Garcia
Summary: This study successfully predicted exclusive breastfeeding during in-hospital postpartum stay using machine learning algorithms and explained the model's behavior to support decision making. The results demonstrated the order of importance of specific predictor variables to the outcome.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2022)
Review
Computer Science, Artificial Intelligence
Cristina Cachero, Juan Ramon Rico-Juan, Hermenegilda Macia
Summary: The successful instructional design of self and peer assessment in higher education faces challenges, including the influence of students' personalities on their intention to adopt peer assessment. This study conducted a quasi-experiment with 85 participants in a Computer Engineering program, assessing their personality and acceptance of three modalities of peer assessment. The results showed that the Random Forest algorithm had significantly better predictions for three out of four adoption variables. The study also found that Agreeableness, Extraversion, and Neuroticism were the best predictors for different aspects of peer assessment. The discussion emphasizes the role of low Consciousness in predicting resistance to peer assessment and highlights the positive impact of peer assessment on students with higher Neuroticism. However, the study also suggests that personality variables have a greater impact on student perceptions than the modality of peer assessment.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Carlos de la Fuente, Francisco J. Castellanos, Jose J. Valero-Mas, Jorge Calvo-Zaragoza
Summary: This research presents a new approach to detect frustration in game-play scenarios by automatically extracting meaningful descriptors from individual audio and video sources of information using Deep Neural Networks (DNN). The multimodal proposals introduced in this study outperform other state-of-the-art approaches, achieving error rate improvements of between 40% and 90%.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Alejandro Sanchez-Ferrer, Jose J. Valero-Mas, Antonio Javier Gallego, Jorge Calvo-Zaragoza
Summary: The large amount of debris in the oceans has a significant impact on marine life. Efforts to tackle this problem through human-based campaigns have been insufficient due to the overwhelming amount of litter. Autonomous underwater vehicles (AUVs) have gained interest as a potential solution for locating and collecting garbage. This study explores the use of Mask Region-based Convolutional Neural Networks for automatic marine debris location and classification with limited data availability, achieving state-of-the-art results and suggesting room for further improvement.
PATTERN RECOGNITION LETTERS
(2023)
Article
Computer Science, Artificial Intelligence
Maria Alfaro-Contreras, Jose J. Valero-Mas, Jose M. Inesta, Jorge Calvo-Zaragoza
Summary: Music transcription is crucial for Music Information Retrieval (MIR) as it converts music sources into a structured digital format. The MIR community has approached this problem through two lines of research: Optical Music Recognition (OMR) for music documents and Automatic Music Transcription (AMT) for audio recordings. Although these fields have developed modality-specific frameworks, recent developments in sequence labeling tasks have led to a common output representation, enabling research on multimodal image and audio music transcription.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Adrian Rosello, Jose J. Valero-Mas, Antonio Javier Gallego, Javier Saez-Perez, Jorge Calvo-Zaragoza
Summary: The use of deep learning in computer vision tasks can achieve remarkable results, but it depends on the availability of training data and its relationship with the application scenario. Domain adaptation techniques are crucial in robotics, where there is limited access to targeted environment data. To facilitate research in this area, Kurcuma provides a collection of datasets for kitchen utensil recognition, along with a baseline using domain-adversarial training.
PATTERN ANALYSIS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Jose J. Valero-Mas, Antonio Javier Gallego, Pablo Alonso-Jimenez, Xavier Serra
Summary: This study adapts multiclass prototype generation strategies to the multilabel case and demonstrates through experiments that they significantly improve efficiency and classification performance, especially showing stronger robustness in noisy scenarios.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Maria Alfaro-Contreras, Antonio Rios-Vila, Jose J. Valero-Mas, Jorge Calvo-Zaragoza
Summary: This paper proposes a self-supervised learning-based method for symbol recognition in document images. It trains a neural-based feature extractor with unlabeled documents and performs recognition with only a few reference samples. Experimental results demonstrate that this method achieves high accuracy rates of up to 95% in few-shot settings and outperforms supervised learning approaches using the same amount of data.
PATTERN RECOGNITION LETTERS
(2023)
Article
Computer Science, Artificial Intelligence
Juan Llorca-Schenk, Juan Ramon Rico-Juan, Miguel Sanchez-Lozano
Summary: This paper presents the development of a tool based on machine learning (ML) to solve the critical aspect of porthole die design. The ML-based model, using a large amount of geometrical data from successful designs, outperforms the previous linear regression model in terms of predictive accuracy.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Jose J. Valero-Mas, Antonio Javier Gallego, Juan Ramon Rico-Juan
Summary: SNNs are a representative approach for Few-Shot Image Classification, utilizing weight sharing CNN models to reduce parameters and overfitting. This study assesses the representation capabilities of SNN architectures, introduces techniques such as data augmentation and transfer learning, and achieves high classification rates with limited prototypes per class.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Proceedings Paper
Acoustics
Victor Arroyo, Jose J. Valero-Mas, Jorge Calvo-Zaragoza, Antonio Pertusa
Summary: This research introduces a new output representation to address the limitations of sequence-based A2S recognition framework and provides an initial approximation for dealing with unconstrained polyphony. The proposed method is validated using synthetic audio from string quartets and piano sonatas with intricate polyphonic mixtures, and it improves the state-of-the-art rates for fixed-polyphony.
2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
(2022)