Annals of Data Science最新文献

{"title":"Exponentiated Odd Lindley-X Power Series Class of Distributions: Properties and Applications","authors":"Fastel Chipepa,&nbsp;Nonhle Mdziniso,&nbsp;Shahid Mohammad,&nbsp;Sher Chhetri","doi":"10.1007/s40745-025-00608-w","DOIUrl":"10.1007/s40745-025-00608-w","url":null,"abstract":"<div><p>In this article, we introduce a new family of probability distributions, the Exponentiated Odd Lindley-X Power Series (EOL-XPS) class, which is derived by integrating the power series distribution with the exponentiated odd Lindley-X family. We establish several statistical properties of this new class, including moments, the moment-generating function, the quantile function, mean deviations, order statistics, and Rényi entropy. As a special case, we derive the probability density function and cumulative distribution function of the Exponentiated Odd Lindley-Weibull Poisson (EOL-WP) distribution, using the Weibull-Poisson distribution as the baseline. To assess the robustness of the proposed model, we conduct a Monte Carlo simulation study to evaluate the performance of maximum likelihood estimation for parameter estimation. Furthermore, we apply the EOL-WP model to COVID-19 and Kevlar datasets, demonstrating its flexibility and practical relevance in modeling complex data.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"13 4","pages":"935 - 962"},"PeriodicalIF":0.0,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s40745-025-00608-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148614786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Variable Selection and Variable Integration for Categorical Dummy Variables in Regression Analysis","authors":"Mototsugu Fukushige","doi":"10.1007/s40745-025-00607-x","DOIUrl":"10.1007/s40745-025-00607-x","url":null,"abstract":"<div><p>Mode selection or variable selection in regression analysis is considered one of the most popular problems to study in empirical research and various theoretical and simulation studies have been conducted. When dealing with categorical data, coding with multiple dummy variables is usually used. However, with this approach, variable selection criteria cannot be systematically applied when some categories are integrated into one category. Most researchers see variables with coefficients near size and integrate them and select variables by minimizing information criteria or other model selection criteria. The underlying reason for the implementation of such cumbersome procedures is related to the lack of systematic variable selection procedures. Finding a way to list all the candidates of the combinations of explanatory variables and their integrated versions remains a major problem. Even if all of them can be listed, estimating the regression models for all the variations of combinations takes too much time, which is another practical problem. To effectively address these issues, in this work, we examined the possibility of utilizing variable selection criteria for ordered categorical data by estimated ordering. We also conducted a simulation study to check the performance of some information criteria.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"13 4","pages":"909 - 934"},"PeriodicalIF":0.0,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s40745-025-00607-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148613332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"New Algorithms for Constructing Magic Squares with Ancient Chinese Wisdoms","authors":"Dee Bruce Sun,&nbsp;Stephen Sun","doi":"10.1007/s40745-025-00598-9","DOIUrl":"10.1007/s40745-025-00598-9","url":null,"abstract":"&lt;div&gt;&lt;p&gt;The construction of magic squares is an ancient mathematical problem with numerous applications in areas such as information transmission in network security. While there have been various approaches to constructing high-order magic squares, they are typically computationally intensive. We explore and adapt approaches with roots in Chinese classics to develop a set of simple yet comprehensive methods for constructing magic squares of &lt;i&gt;any&lt;/i&gt; order. For generating odd-order magic squares, the traditional \"Siamese method\" (The Siamese Method(or De la Loubère Method): starting from the central box of the first row with the number 1 (or the first number of any arithmetic progression), the fundamental movement for filling the boxes is diagonally up and right (↗), one step at a time. When a move would leave the square, it is wrapped around to the last row or first column, respectively. If a filled box is encountered, one moves vertically down one box (↓) instead, then continuing as before.) was introduced over 300 years ago by Simon de la Loubère, a French ambassador to Thailand in later seventeenth century. While workable, the Siamese method is cumbersome and lacks an underlying logic (Benjamin et al. in Coll Math J 45:92–100, 2014). We employ a concept derived from the circular structure of Five Elements (The &lt;b&gt;Five Elements&lt;/b&gt; (Wuxing) is one of the two fundamental Theories define Chinese civilization. It is characterized essentially by five components to interpret everything. In endless cycles, Five Elements are mutually generative, and at the same time mutually restrictive. For the generation relationship, &lt;b&gt;Metal&lt;/b&gt; generates &lt;b&gt;Water&lt;/b&gt; generates &lt;b&gt;Wood&lt;/b&gt; generates &lt;b&gt;Fire&lt;/b&gt; generates &lt;b&gt;Earth&lt;/b&gt; generates Metal generates Water generates Wood……(see Fig. 1B below); For the restraint relationship, Metal destructs Wood destructs Earth destructs Water destructs Fire destructs Metal destructs Wood destructs Earth generates Water …… This key concept of dynamic and constant balancing, unbalancing, rebalancing process applies universally to explaining natural phenomena, cosmology, medicine, politics, and human affairs.) in Chinese culture to establish a prototype for constructing odd-order magic squares. This approach significantly simplifies the algorithm, and reveals the underlying mathematical logic and mechanism of the Siamese method. We also adapt the composition of the Bagua (The &lt;b&gt;Yin-Yang&lt;/b&gt; is another fundamental theory of Chinese thoughts, originating from Yijing. It is characterized with the Yin-Yang duality and interdependence of two for all things in the universe. Where Yin represents the passive, dark, cold, feminine, and &lt;b&gt;receptive&lt;/b&gt; aspects of nature; Yang represents the active, bright, hot, masculine, and &lt;b&gt;expansive&lt;/b&gt; aspects of nature. Yin and Yang are contrast yet complementary. The Yin Sphere and Yang Sphere are in opposition, eternally embracing together, moving toward and transforming into each other","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"12 6","pages":"1775 - 1798"},"PeriodicalIF":0.0,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145537721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Uncovering University Application Patterns Through Graph Representation Learning","authors":"Hendrik Santoso Sugiarto,&nbsp;Yozef Tjandra","doi":"10.1007/s40745-025-00611-1","DOIUrl":"10.1007/s40745-025-00611-1","url":null,"abstract":"<div><p>In university admissions, interaction networks naturally emerge between prospective students and available majors. Understanding hidden patterns in such a vast network is crucial for decision-making but poses technical challenges due to its complexity and data limitations. Many existing models rely heavily on user profiling, raising privacy concerns and making data collection difficult. Instead, this work extracts meaningful insights using only the adjacency information of the network, avoiding the need for personal data. We leverage Graph Convolutional Networks (GCN) to generate compact representations for major recommendation and clustering tasks. Our GCN-based approach outperforms classical methods such as popularity-based and Non-negative Matrix Factorization (NMF), as well as the neural Generalized Matrix Factorization (GMF) model, achieving up to 61.06% and 12.17% improvements in smaller (dimension 40) and larger (dimension 80) embeddings, respectively. Furthermore, hierarchical clustering on these embeddings reveals implicit patterns in student preferences, particularly regarding fields of study and geographic locations, even without explicit data on these attributes. These findings demonstrate that meaningful insights can be derived from interaction networks while mitigating privacy concerns associated with user profiling.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"12 4","pages":"1343 - 1368"},"PeriodicalIF":0.0,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145163350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Hybrid Approach for Stock Market Price Forecasting Using Long Short-Term Memory and Seahorse Optimization Algorithm","authors":"Burak Gülmez","doi":"10.1007/s40745-025-00609-9","DOIUrl":"10.1007/s40745-025-00609-9","url":null,"abstract":"<div><p>Stock market price forecasting is a challenging task due to the complexity and volatility of market dynamics. This paper proposes a novel approach that combines the strengths of Long Short-Term Memory (LSTM) networks and the Seahorse Optimization (SHO) algorithm for stock market price forecasting. The LSTM-SHO model is compared with other LSTM models optimized using Genetic Algorithm (GA) and different dimensionalities (1D, 2D, 3D), as well as an Artificial Neural Network (ANN) model. The results show that the LSTM-SHO model outperforms the other models in terms of mean squared error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (R2). The hyperparameter tuning process using SHO significantly improves the forecasting accuracy of the LSTM model. The proposed approach demonstrates its potential in developing more accurate and robust stock market forecasting models, which can aid investors and analysts in making informed decisions. The findings of this study have important implications for investors, analysts, and policymakers, and contribute to the existing body of literature on stock market forecasting.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"13 5","pages":"1075 - 1101"},"PeriodicalIF":0.0,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s40745-025-00609-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149098346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rumor Governance Under Uncertain Conditions: An Evolutionary Game Theory Analysis","authors":"Xuefan Dong,&nbsp;Lei Tang","doi":"10.1007/s40745-025-00606-y","DOIUrl":"10.1007/s40745-025-00606-y","url":null,"abstract":"<div><p>In the rapidly evolving landscape of online information dissemination, managing rumors has become an imperative challenge for governments worldwide. This study employs a tripartite evolutionary game model to examine the behavior evolution of the government, online media, and netizens in the process of rumor propagation under uncertain conditions. The innovation of the model lies in considering the probability of successful rumor detection under government regulation, the uncertainty of rumor dissemination by online media and netizens, and introducing a dynamic government penalty mechanism. Through simulation and analysis, we identify the evolutionarily stable strategies of each participant under different scenarios and provide specific governance strategies for each party involved. The results reveal that appropriate government penalties, proactive regulation by online media, and rational choices by netizens can effectively curb rumor spreading. In uncertain environments, adopting flexible policies and dynamic adjustment mechanisms is crucial for effective rumor governance. The results reveal that appropriate government penalties, proactive regulation by online media, and rational choices by netizens can effectively curb rumor spreading. In uncertain environments, adopting flexible policies and dynamic adjustment mechanisms is crucial for effective rumor governance. This study not only enriches the application of evolutionary game theory but also offers practical strategic recommendations for policymakers to address the challenges of rumor propagation.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"12 3","pages":"1073 - 1111"},"PeriodicalIF":0.0,"publicationDate":"2025-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145161941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal Individual Selection Algorithm Based on Layer Proximity and Branch Distance Functions","authors":"An Yingjian,&nbsp;La Ping","doi":"10.1007/s40745-025-00600-4","DOIUrl":"10.1007/s40745-025-00600-4","url":null,"abstract":"<div><p>Automatic generation of test cases using heuristic methods is a hot research topic nowadays. Although its advantages are obvious, it is slightly insufficient in the selection of optimal individuals. Aiming at the existing problems in the evaluation and selection of the optimal individual, this paper proposes a test case evaluation algorithm based on the comprehensive analysis of the characteristics of layer proximity and branch distance function, which is a joint structure of “layer proximity and branch distance function”. The basic idea of this algorithm is that when selecting pilot individuals in the evolutionary process, we first select the individuals with high proximity between the actual execution path and the target path, and then select the individuals with the smallest branching distances among these individuals, so as to obtain the individuals with the optimal piloting ability. Experiments show that the proposed algorithm can quickly find the optimal test cases, especially for the test case generation of multi-layer nested programs.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"12 3","pages":"1041 - 1054"},"PeriodicalIF":0.0,"publicationDate":"2025-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145161940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"RBON: A Robust Bi-objective Feature Selection Framework Using Hybrid Ensemble Technique and NSGA-II","authors":"Kurman Sangeeta,&nbsp;Sumitra Kisan","doi":"10.1007/s40745-025-00599-8","DOIUrl":"10.1007/s40745-025-00599-8","url":null,"abstract":"<div><p>Stability, a method’s sensitivity to training set changes, is often overlooked in studies, despite its importance in robustness and model performance. Our proposal aims to address this issue without compromising feature reduction. The proposed model uses an NSGA-II multiobjective genetic algorithm to optimise two competing goals, accuracy and number of features. The suggested model functions in two phases. The first is stability retainment, population initialization and, multiobjective optimization being the second stage. A hybrid ensemble technique that integrates both data and functional variation is applied to select stable features. Functional variation is achieved using three different feature selection algorithms: mRMR, ReliefF, and Random forest. The selected stable feature subsets guide the population initialization process of the multi-objective NSGA-II algorithm to optimize two objectives, accuracy and number of features. We considered three datasets, two being gene expressions in a microarray dataset, Leukemia and colon cancer, and one is an imbalanced dataset, Cervical cancer risk factor to validate the proposed method. The results show a significant increase in accuracy and reduced feature subset size with the same or improved stability.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"13 4","pages":"829 - 847"},"PeriodicalIF":0.0,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148613026","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PNAP-YOLO: An Improved Prompts-Based Naturalistic Adversarial Patch Model for Object Detectors","authors":"Jun Li,&nbsp;Chenwu Shan,&nbsp;Liyan Shen,&nbsp;Yawei Ren,&nbsp;Jiajie Zhang","doi":"10.1007/s40745-025-00604-0","DOIUrl":"10.1007/s40745-025-00604-0","url":null,"abstract":"<div><p>Detectors have been extensively utilized in various scenarios such as autonomous driving and video surveillance. Nonetheless, recent studies have revealed that these detectors are vulnerable to adversarial attacks, particularly adversarial patch attacks. Adversarial patches are specifically crafted to disrupt deep learning models by disturbing image regions, thereby misleading the deep learning models when added to into normal images. Traditional adversarial patches often lack semantics, posing challenges in maintaining concealment in physical world scenarios. To tackle this issue, this paper proposes a Prompt-based Natural Adversarial Patch generation method, which creates patches controllable by textual descriptions to ensure flexibility in application. This approach leverages the latest text-to-image generation model—Latent Diffusion Model (LDM) to produce adversarial patches. We optimize the attack performance of the patches by updating the latent variables of LDM through a combined loss function. Experimental results indicate that our method can generate more natural, semantically rich adversarial patches, achieving effective attacks on various detectors.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"12 3","pages":"1055 - 1072"},"PeriodicalIF":0.0,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145161140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Feature Selection with an Improved Flow Direction Algorithm to Improve the Performance of Intrusion Detection Systems","authors":"Farideh Ghahramani,&nbsp;Farhad Soleimanian Gharehchopogh","doi":"10.1007/s40745-025-00602-2","DOIUrl":"10.1007/s40745-025-00602-2","url":null,"abstract":"<div><p>Intrusion Detection Systems (IDSs) rely heavily on critical features to enhance their detection accuracy. Due to numerous irrelevant or redundant features in network data, classification approaches struggle to identify patterns in data without accurately selecting the proper features. It aims to eliminate irrelevant and redundant features to maintain or increase classification accuracy. This paper uses an enhanced binary Flow Direction Algorithm (FDA) for feature selection. The improved FDA proposes two binary models for the feature selection issue. In the first model, an S-shaped transfer function is used, and in the second model, a V-shaped transfer function is employed to convert the continuous version to binary. The FDA algorithm is designed to solve continuous optimization problems; therefore, solutions for feature selection need to be converted into a binary form. A Lévy Flight (LF) operator enhances the FDA algorithm. This operator helps balance exploration and exploitation, addressing the issue of getting trapped in local optima. Two datasets, KDD99 and NSL-KDD, have been used to validate the proposed model. The results indicate that the proposed model's accuracy percentage is higher than the optimization algorithms. Specifically, the accuracy percentages for the proposed model on the KDD99 and NSL-KDD datasets are 97.71% and 98.32%, respectively.</p></div>","PeriodicalId":36280,"journal":{"name":"Annals of Data Science","volume":"13 4","pages":"873 - 907"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148614346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}