Article
Automation & Control Systems
Dominik Olszewski
Summary: The study introduces an enhanced adaptive version of SOM that preserves input data structure and captures data scattering, which has been empirically proven to be superior to other data visualization techniques.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Shuo Li, Fang Liu, Licheng Jiao, Puhua Chen, Lingling Li
Summary: This paper proposes a deep learning-based clustering method called S(3)OCNet, which achieves joint learning of feature extraction and feature clustering through self-supervised learning, thus realizing a single-stage clustering method. The method significantly improves the performance on multiple image classification benchmarks and the effectiveness of the clustering results is validated through feature and image visualization.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Interdisciplinary Applications
Ryosuke Motegi, Yoichi Seki
Summary: This paper proposes an efficient algorithm that automatically selects a suitable number of clusters based on a probability distribution model framework. The algorithm includes a generalized self-organizing map and a dynamically updating method of the map structure. Compared with existing methods, the proposed method is computationally efficient and can accurately select the number of clusters.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2023)
Article
Geosciences, Multidisciplinary
Flora Giudicepietro, Antonietta M. Esposito, Laura Spina, Andrea Cannata, Daniele Morgavi, Lukas Layer, Giovanni Macedonio
Summary: The study focuses on using an unsupervised neural network, Self-Organizing Map (SOM), to cluster seismo-acoustic events and investigate the cause-effect relationships between signals and processes. The experimental conditions were controlled to validate the effectiveness of the neural network in clustering events.
FRONTIERS IN EARTH SCIENCE
(2021)
Review
Environmental Sciences
Sabina Licen, Aleksander Astel, Stefan Tsakovski
Summary: The evaluation of the spatial and temporal distribution of pollutants and the use of Self-Organizing Map (SOM) is crucial for assessing the anthropogenic burden on the environment. The SOM is an artificial neural network that can handle non-linear problems and is used for exploratory data analysis, pattern recognition, and variable relationship assessment. The review provides a description of the SOM operation principle, its application for obtaining pollution patterns, and advice for reporting and extracting valuable information from the model results.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Environmental Sciences
Brittany V. Lancellotti, Kristen L. Underwood, Julia N. Perdrial, Andrew W. Schroth, Eric D. Roy, Carol E. Adair
Summary: Oxygen (O-2) plays a crucial role in regulating soil processes and biogeochemical cycles. However, accurately measuring soil O-2 variability is challenging, leading to the use of soil moisture as a proxy for O-2. However, relying solely on soil moisture measurements may lead to inaccurate estimations of soil O-2 levels.
WATER RESOURCES RESEARCH
(2023)
Article
Engineering, Marine
Peng Yao, Yating Lou, Keming Zhang
Summary: In this paper, a hierarchical two-layer framework is proposed for the cooperative path planning of multi-USV system in complex marine environments. The framework includes low-level path planning for each USV and high-level target allocation for the multi-USV system. An improved self-organizing map (SOM) based on window update is used to avoid collisions and access multiple targets in the optimal sequence. The simulation results validate the feasibility and effectiveness of the proposed method for multi-USV cooperation and path traceability.
Article
Computer Science, Information Systems
Hiroomi Hikawa
Summary: This paper investigates the impact of nested hardware SOM on FPGA implementation tools and finds that the nested architecture provides better results in resource usage and performance.
Article
Computer Science, Artificial Intelligence
Jialiang Xie, Shanli Zhang, Honghui Wang, Dongrui Wu
Summary: This paper proposes a gray wolf optimization-based self-organizing fuzzy multi-objective evolutionary algorithm. It optimizes the initial weights using the gray wolf optimization algorithm and establishes new neighborhood relationships using self-organizing map. The algorithm utilizes the fuzzy differential evolution operator to generate new initial solutions and apply the polynomial mutation operator for refinement. Experimental results demonstrate that the proposed algorithm outperforms other state-of-the-art multi-objective evolutionary algorithms in terms of convergence and diversity.
Article
Engineering, Civil
Yaobin Zhang, Qiulan Zhang, Wenfang Chen, Weiwei Shi, Yali Cui, Leilei Chen, Jingli Shao
Summary: This study utilized a combination of self-organizing map (SOM) and K-means clustering to investigate the hydrogeochemical characteristics at a contaminated site. The results showed that the groundwater characteristics could be divided into different clusters, and the chemistry control mechanisms of these clusters were analyzed using the Gibbs diagram and saturation index method.
JOURNAL OF HYDROLOGY
(2023)
Article
Engineering, Civil
Li-Chiu Chang, Wu-Han Wang, Fi-John Chang
Summary: The study compared the effectiveness of two SOM training strategies, with the S2 strategy demonstrating higher efficiency and effectiveness in constructing regional flood inundation maps.
JOURNAL OF HYDROLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Leonardo A. Dias, Augusto M. P. Damasceno, Elena Gaura, Marcelo A. C. Fernandes
Summary: The study introduces a fully parallel architecture for SOM that significantly improves processing speed and resource efficiency.
Article
Computer Science, Artificial Intelligence
Seng-Khoon Teh, Iris Rawtaer, Ah-Hwee Tan
Summary: In-home sensing of daily living patterns from older adults coupled with machine learning is a promising approach to detect Mild Cognitive Impairment (MCI), and a predictive self-organizing neural network known as fuzzy Adaptive Resonance Associate Map (fuzzy ARAM) has been proposed to detect MCI using in-home sensor data. Fuzzy ARAM showed the highest predictive performance and yielded unique rules for MCI detection compared to other predictive models.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Chemistry, Analytical
Zerun Li, Qinglin Wang, Yufei Zhu, Zuocheng Xing
Summary: This paper proposes a hierarchical self-organizing map model based on high-order cumulants and amplitude moment features for automatic modulation classification, showing advantages in classification accuracy and computational resource consumption.
Article
Green & Sustainable Science & Technology
Mehrbakhsh Nilashi, Shahla Asadi, Rabab Ali Abumalloh, Sarminah Samad, Fahad Ghabban, Eko Supriyanto, Reem Osman
Summary: This study introduces a new approach based on machine learning techniques for assessing sustainability performance, considering ecological sustainability and human sustainability through cluster analysis and prediction learning. By applying SOM and CART techniques to a dataset of indicators from 128 countries, prediction models were generated, and the integration of models improved prediction accuracy.