Similarity Measure Research Articles

JLeNeT: Jaccard LeNet for Parkinson's disease detection and severity level classification using voice signal in IoT environment.

The neurodegenerative disorder called Parkinson's disease (PD) is one of the most common diseases now a day. In this research, PD is detected and severity classification is done using the proposed Jaccard LeNet (JLeNet) with the help of voice signal in the IoT environment. Here, the IoT simulation is done. Initially, from which voice signal is collected and the routing process is done by the proposed Chimp Wild Geese Algorithm (ChWGA). This ChWGA is the combination of the Wild Geese Algorithm (WGA) and Chimp Optimization Algorithm (ChOA). Finally, at Base Station (BS), PD is detected and classified. The input voice signal is fed for pre-processing conducted by an adaptive Kalman filter. Following this, feature extraction and feature selection are conducted, where Harmonic mean similarity helps in feature selection. Here, PD is detected using JLeNet, which is the hybridization of LeNet with the Jaccard similarity measure. In this work, routing metrics of energy and delay are superior and recorded with the values of 0.309 J and 0.434 ms for the ChWGA. Moreover, the proposed method attains an Accuracy of 0.910, True Positive Rate (TPR) of 0.903, and True Negative Rate (TNR) of 0.918.

Pythagorean Fuzzy Overlap Functions and Corresponding Fuzzy Rough Sets for Multi-Attribute Decision Making

As a non-associative connective in fuzzy logic, the analysis and research of overlap functions have been extended to many generalized cases, such as interval-valued and intuitionistic fuzzy overlap functions (IFOFs). However, overlap functions face challenges in the Pythagorean fuzzy (PF) environment. This paper first extends overlap functions to the PF domain by proposing PF overlap functions (PFOFs), discussing their representable forms, and providing a general construction method. It then introduces a new PF similarity measure which addresses issues in existing measures (e.g., the inability to measure the similarity of certain PF numbers) and demonstrates its effectiveness through comparisons with other methods, using several examples in fractional form. Based on the proposed PFOFs and their induced residual implication, new generalized PF rough sets (PFRSs) are constructed, which extend the PFRS models. The relevant properties of their approximation operators are explored, and they are generalized to the dual-domain case. Due to the introduction of hesitation in IF and PF sets, the approximate accuracy of classical rough sets is no longer applicable. Therefore, a new PFRS approximate accuracy is developed which generalizes the approximate accuracy of classical rough sets and remains applicable to the classical case. Finally, three multi-criteria decision-making (MCDM) algorithms based on PF information are proposed, and their effectiveness and rationality are validated through examples, making them more flexible for solving MCDM problems in the PF environment.

Advancing AI-driven thematic analysis in qualitative research: a comparative study of nine generative models on Cutaneous Leishmaniasis data

BackgroundAs part of qualitative research, the thematic analysis is time-consuming and technical. The rise of generative artificial intelligence (A.I.), especially large language models, has brought hope in enhancing and partly automating thematic analysis.MethodsThe study assessed the relative efficacy of conventional against AI-assisted thematic analysis when investigating the psychosocial impact of cutaneous leishmaniasis (CL) scars. Four hundred forty-eight participant responses from a core study were analysed comparing nine A.I. generative models: Llama 3.1 405B, Claude 3.5 Sonnet, NotebookLM, Gemini 1.5 Advanced Ultra, ChatGPT o1-Pro, ChatGPT o1, GrokV2, DeepSeekV3, Gemini 2.0 Advanced with manual expert analysis. Jamovi software maintained methodological rigour through Cohen’s Kappa coefficient calculations for concordance assessment and similarity measurement via Python using Jaccard index computations.ResultsAdvanced A.I. models showed impressive congruence with reference standards; some even had perfect concordance (Jaccard index = 1.00). Gender-specific analyses demonstrated consistent performance across subgroups, allowing a nuanced understanding of psychosocial consequences. The grounded theory process developed the framework for the fragile circle of vulnerabilities that incorporated new insights into CL-related psychosocial complexity while establishing novel dimensions.ConclusionsThis study shows how A.I. can be incorporated in qualitative research methodology, particularly in complex psychosocial analysis. Consequently, the A.I. deep learning models proved to be highly efficient and accurate. These findings imply that the future directions for qualitative research methodology should focus on maintaining analytical rigour through the utilisation of technology using a combination of A.I. capabilities and human expertise following standardised future checklist of reporting full process transparency.

The debate on electricity in the eighteenth century: a multilayered digital perspective

Abstract Late-eighteenth century science tends to dismiss the search for the true causes of natural phenomena and tries instead to offer a quantifiable and eventually mathematical account of them. By taking the debates on electricity during that period as a case study, this paper aims to ascertain whether, and to what extent, later more quantified and even mathematized approaches are directly supported or continuously connected with earlier approaches. In order to do so, we take into account a relatively large pool of early modern works devoted to electricity and contextualize them within a much larger corpus of natural philosophy works. By using a computational approach, we then represent these two corpora as networks based on similarity measures derived from how language is used in each work. Network representation helps us to investigate the place and impact of several works within our corpus, extract information on possible trends, and select key samples for close reading. Based on our results, we surmise that the effort towards mathematization does not arise from having dismissed the quest for the actual physical causes of phenomena, nor is it necessarily correlated with a more skeptical or instrumental attitude towards the knowledge attainable in natural philosophy.

Disambiguating a Soft Metagenomic Clustering.

Clustering is a popular technique used for analyzing amplicon sequencing data in metagenomics. Specifically, it is used to assign sequences (reads) to clusters, each cluster representing a species or a higher level taxonomic unit. Reads from multiple species often sharing subsequences, combined with lack of a perfect similarity measure, make it difficult to correctly assign reads to clusters. Thus, metagenomic clustering methods must either resort to ambiguity, or make the best available choice at each read assignment stage, which could lead to incorrect clusters and potentially cascading errors. In this article, we argue for first generating an ambiguous clustering and then resolving the ambiguities collectively by analyzing the ambiguous clusters. We propose a rigorous formulation of this problem and show that it is NP-Hard. We then propose an efficient heuristic to solve it in practice. We validate our approach on several synthetically generated datasets and two datasets consisting of 16S rDNA sequences from the microbiome of rat guts.

Policy Similarity Measure for Two-Player Zero-Sum Games

Policy space response oracles (PSRO) is an important algorithmic framework for approximating Nash equilibria in two-player zero-sum games. Enhancing policy diversity has been shown to improve the performance of PSRO in this approximation process significantly. However, existing diversity metrics are often prone to redundancy, which can hinder optimal strategy convergence. In this paper, we introduce the policy similarity measure (PSM), a novel approach that combines Gaussian and cosine similarity measures to assess policy similarity. We further incorporate the PSM into the PSRO framework as a regularization term, effectively fostering a more diverse policy population. We demonstrate the effectiveness of our method in two distinct game environments: a non-transitive mixture model and Leduc poker. The experimental results show that the PSM-augmented PSRO outperforms baseline methods in reducing exploitability by approximately 7% and exhibits greater policy diversity in visual analysis. Ablation studies further validate the benefits of combining Gaussian and cosine similarities in cultivating more diverse policy sets. This work provides a valuable method for measuring and improving the policy diversity in two-player zero-sum games.

AI-determined similarity increases likability and trustworthiness of human voices.

Modern artificial intelligence (AI) technology is capable of generating human sounding voices that could be used to deceive recipients in various contexts (e.g., deep fakes). Given the increasing accessibility of this technology and its potential societal implications, the present study conducted online experiments using original data to investigate the validity of AI-based voice similarity measures and their impact on trustworthiness and likability. Correlation analyses revealed that voiceprints - numerical representations of voices derived from a speaker verification system - can be used to approximate human (dis)similarity ratings. With regard to cognitive evaluations, we observed that voices similar to one's own voice increased trustworthiness and likability, whereas average voices did not elicit such effects. These findings suggest a preference for self-similar voices and underscore the risks associated with the misuse of AI in generating persuasive artificial voices from brief voice samples.

V ex IR2V ec : An Architecture-Neutral Embedding Framework for Binary Similarity

Binary similarity involves determining whether two binary programs exhibit similar functionality with applications in vulnerability detection, malware analysis, and copyright detection. However, variations in compiler settings, target architectures, and deliberate code obfuscations significantly complicate the similarity measurement by effectively altering the syntax, semantics, and structure of the underlying binary. To address these challenges, we propose V ex IR2V ec , a robust, architecture-neutral approach based on VEX-IR to solve binary similarity tasks. V ex IR2V ec consists of three key components: a peephole extractor, a normalization engine (V ex INE), and an embedding model (V ex N et ). The process to build program embeddings starts with the extraction of sequences of basic blocks, or peepholes , from control-flow graphs via random walks, capturing structural information. These generated peepholes are then normalized using V ex INE, which applies compiler-inspired transformations to reduce architectural and compiler-induced variations. Embeddings of peepholes are generated using representation learning techniques, avoiding Out-Of-Vocabulary (OOV) issues. These embeddings are then fine-tuned with V ex N et , a feed-forward Siamese network that maps functions into a high dimensional space for diffing and searching tasks in an application-independent manner. We evaluate V ex IR2V ec against five baselines — BinDiff, DeepBinDiff, SAFE, BinFinder, and histograms of opcodes — on a dataset comprising \(2.7M\) functions and \(15.5K\) binaries from \(7\) projects compiled across \(12\) compilers targeting x86 and ARM architectures. The experiments span four adversarial settings — cross-optimization, cross-compilation, cross-architecture, and obfuscations — that are typically exploited by malware and vulnerabilities. In diffing experiments, V ex IR2V ec outperforms the nearest baseline in these four scenarios by \(40\%\) , \(18\%\) , \(21\%\) , and \(60\%\) , respectively. In the searching experiment, V ex IR2V ec achieves a mean average precision of \(0.76\) , the nearest baseline, by \(46\%\) . Our framework is highly scalable and is built as a lightweight, multi-threaded, parallel library using only open-source tools. V ex IR2V ec is \(\approx 3.1\) – \(3.5\times\) faster than the closest baselines and orders-of-magnitude faster than other tools.

A systematic review of low-rank and local low-rank matrix approximation in big data medical imaging

Abstract The large volume and complexity of medical imaging datasets pose significant challenges for storage, transmission, and processing. To address these issues, low-rank matrix approximation (LORMA) and its derivative, local LORMA (LLORMA), have shown promising potential. This paper presents a comprehensive literature review of the application of LORMA and LLORMA across various imaging modalities and examines the challenges and limitations of existing methods. Notably, since 2015, there has been a significant shift toward a preference for LLORMA in the medical imaging field, demonstrating its effectiveness in capturing complex structures in medical data compared to LORMA. Given the limitations of shallow similarity methods in LLORMA, we propose incorporating advanced semantic image segmentation to improve the accuracy of similarity measurement. We further explain how this approach can be utilized to identify similar patches and assess its feasibility in medical imaging applications. We observe that LORMA and LLORMA have primarily been applied to unstructured medical data, and we suggest extending their use to other types of medical data, including structured and semi-structured formats. This paper also explores how LORMA and LLORMA can be adapted for regular data with missing entries, considering the impact of inaccuracies in predicting these missing values and their consequences. In addition, we examine the effect of patch size and suggest using random search (RS) to identify the optimal patch size. To further enhance feasibility, we propose a hybrid approach combining Bayesian optimization and RS, which could improve the application of LORMA and LLORMA in medical imaging.

A Similarity Matrix for Preserving Haplotype Diversity Amongst Parents in Genomic Selection.

In genomic selection, balancing genetic gain with the preservation of genetic diversity is a critical challenge, requiring innovative approaches to parent selection. Traditional methods risk losing valuable genetic diversity by not fully accounting for the complex patterns of haplotype distribution. To address this, we developed a novel haplotype similarity measure that estimates the genetic similarity amongst offspring from parent pairs by analysing segregating marker patterns and the covariance of additive genetic effects between potential parental gametes. This measure is encapsulated in a novel similarity matrix that quantifies parental genetic relationships and their Mendelian sampling variance, facilitating the selection of parents with diverse haplotypes to maintain genetic diversity. Our method was evaluated through simulation studies and empirical data analysis, indicating that the similarity matrix can help preserve haplotype diversity and potentially improve long-term genetic gains compared to traditional selection methods. These results suggest that the similarity matrix could contribute to more efficient and sustainable genomic selection programs, although further research is necessary to fully understand its impact.

REMatch plus SOS: Machine-learning-accelerated structure prediction for supported metal nanoclusters

Predicting stable structures of nanoclusters is crucial yet computationally demanding. This study presents a machine learning-based methodology designed to accelerate the prediction of stable structures in nanoclusters. By integrating local environment descriptors, with dimensionality reduction, kernel-based similarity measure, and outlier detection, we efficiently screen and select promising configurations, thus accelerating identification of global and local minimum structures. The approach is validated through rigorous optimization, demonstrating its capability to identify low-energy structures while significantly reducing computational costs. This method offers a robust framework for structural screening. Published by the American Physical Society 2025

SC-FSM: a new hybrid framework based on subtractive clustering and fuzzy similarity measures for imbalanced data classification

SC-FSM: a new hybrid framework based on subtractive clustering and fuzzy similarity measures for imbalanced data classification

Cosine similarity and information measures of [formula omitted] spherical fuzzy sets: Application in selecting migration destination

Cosine similarity and information measures of [formula omitted] spherical fuzzy sets: Application in selecting migration destination

STSimM: A new tool for evaluating neuron model performance and detecting spike trains similarity.

STSimM: A new tool for evaluating neuron model performance and detecting spike trains similarity.

Shotgun and Hi-C Sequencing Datasets for Binning Wheat Rhizosphere Microbiome

Binning is a crucial process in metagenomics studies, where sequenced reads are combined to form longer contigs and assigned to individual genomes. Conventional methods, such as shotgun binning, rely on similarity measurements and abundance profiles across multiple samples. However, cost constraints for sequencing and limited sample collection capacity hinder their effectiveness. High-throughput chromosome conformation capture (Hi-C), a DNA proximity ligation technique, has been adapted to accurately bin metagenome-assembled genomes (MAGs) from a single sample, addressing challenges like chimeric MAGs. In this study, we generated over 190 Gb of metagenomic data from wheat rhizospheres grown in two highly calcareous soils of South Australian region and compared conventional and Hi-C binning methods. Two shotgun metagenomes and Hi-C libraries were generated, assembling 1089 shotgun MAGs across 39 bacterial and one archaeal taxon, including 94 Hi-C based bins. Binning performed using only short read sequences was prone to high contamination, while the addition of Hi-C binning improved MAG quality and identified mobile element-host-infection interaction. This dataset provides important tools for studying microbial communities in wheat rhizosphere soils.

Utilizing conditional generative adversarial networks for data augmentation in logging evaluation

Logging is critical for reservoir and fluid characterization by integrating the geophysical properties of rock layers. Traditional evaluation methods struggle with parameter selection, compromising predictive accuracy and generalizability. Machine learning, particularly conditional generative adversarial networks (CGAN), offers a robust alternative, addressing the disparity between labeled and unlabeled logging data that can lead to evaluation discrepancies. This study applies CGAN to augment data, enhancing input features for improved logging interpretation. We implemented this approach in the Jiyang Depression, Eastern China, using a dataset comprising 194 data points, each consisting of 18 logging curve features and one corresponding total organic carbon measurement from well NY1. The methodology aims to bolster the accuracy and reliability of logging interpretations through targeted data augmentation. We validated the augmented data's reliability through comprehensive analysis, including data characteristic assessments, statistical tests, mutual information analysis, similarity measurements, and consistency testing. The results confirm the efficacy of our data enhancement strategy, providing a robust framework for logging interpretation and future reservoir assessment. The method's effectiveness and reliability suggest broad applicability in fields such as seismic assessment and engineering development. This research not only bridges the gap between labeled and unlabeled data but also harnesses advanced machine learning to enhance geophysical evaluation outcomes. It underscores the importance of innovative data augmentation in the advancement of reservoir characterization and geophysical exploration.

A similarity measure based on subspace distance for spectral clustering

A similarity measure based on subspace distance for spectral clustering

A text-based recommender system for recommending relevant news articles

A text-based recommender system for recommending relevant news articles

Analyzing temporal patterns in frequent emergency department visits among oncology patients using semantic similarity measures.

Analyzing temporal patterns in frequent emergency department visits among oncology patients using semantic similarity measures.

Task-Oriented Channel Attention for Fine-Grained Few-Shot Classification.

The difficulty of fine-grained image classification mainly comes from a shared overall appearance across classes. Thus, recognizing discriminative details, such as the eyes and beaks of birds, is a key to the task. However, this is particularly challenging when training data is limited. To address this, we propose Task Discrepancy Maximization (TDM), a task-oriented channel attention method tailored for fine-grained few-shot classification with two novel modules Support Attention Module (SAM) and Query Attention Module (QAM). SAM highlights channels encoding class-wise discriminative features, while QAM assigns higher weights to object-relevant channels of the query. Based on these submodules, TDM produces task-adaptive features by focusing on channels encoding class-discriminative details and possessed by the query at the same time, for accurate class-sensitive similarity measure between support and query instances. While TDM influences high-level feature maps by task-adaptive calibration of channel-wise importance, we further introduce Instance Attention Module (IAM) operating in intermediate layers of feature extractors to instance-wisely highlight object-relevant channels, by extending QAM. The merits of TDM and IAM and their complementary benefits are experimentally validated in fine-grained few-shot classification tasks. Moreover, IAM is also effective in coarse-grained and cross-domain few-shot classifications.