Practical Data Analysis Research Articles

Application of machine learning techniques for churn prediction in the telecom business

The telecom business generates a significant amount of data on a daily basis due to its massive client base. Acquiring a fresh client base is more expensive than retaining existing customers, whereas churn refers to customers transitioning from one company to another within a specified timeframe. Telecom managers and analysts investigate why customers cancel their subscriptions and analyze the behavior patterns of customers who have stopped using the services. In this work employs categorization methodologies to determine the instances of leave subscriptions and gathers the rationales behind client leave subscriptions in the telecommunications sector. The primary objective of this work is to examine various machine learning algorithms necessary for creating customer churn prediction (CP) models and identifying the reasons for churn. This work aims to provide retention strategies and plans to address churn. This work utilizes machine learning (ML) technique such as random forests (RF) to collect and classify client data for leave subscriptions. These results compare with other ML algorithm such as support vector machines (SVM), gradient boosting (GB), Extreme Gradient Boosting (XGBoost), and light gradient boosting machines (LGBM), The business model provides a practical analysis of customer churn data, enabling accurate forecasts of customers likely to churn. This allows business management to take timely action to prevent churn and minimize profit loss. In this work obtains an accuracy of 98.1 % by utilizing the random forest classifier for churn prediction. The classifier matrix has obtained a precision of 92.8 % and a recall factor of 92.7 %, resulting in an overall accuracy of 95.6 %. Similarly, our research endeavors enhance churn prediction, encompass additional business domains, and furnish prediction models to retain their current consumers, improve customer service, and efficiently prevent churn.

Measuring the center and the dimension of our Galaxy using globular clusters

Abstract Historical measurements, such as the measurement of the size of our Galaxy and distance to its center, are based on relatively simple assumptions, that allow high-school students and undergraduate college students to explore and replicate the basic analysis of astronomical data. Here I present a practical data analysis exercise, that was developed to select the Slovenian participants to the International Olympiad on Astronomy and Astrophysics (IOAA). The exercise guides students to discover the globular cluster distribution within our Galaxy, use basic knowledge of spherical astronomy and coordinate transformation, explore the night sky and develop basic skills for the analysis of astronomical data.

Heterogenous relapse and efficacy endpoint definitions for neuromyelitis optica spectrum disorder studies: A systematic review

BackgroundOver the last years, multiple studies have been dedicated to evaluate the efficacy of different treatment options for Neuromyelitis Optica Spectrum Disorder (NMOSD). However, there is a wide variety of endpoints employed across these studies. Our goal is to conduct a systematic review describing the endpoints utilized in studies related to NMOSD. MethodsMedline, Embase, and Cochrane were searched from inception to May 2023, to identify studies analyzing treatment options in patients with NMOSD. We collected data on baseline study characteristics and all efficacy outcomes available. ResultsWe included 127 studies and identified approximately 40 different efficacy endpoints, categorized into 1) relapse, 2) disability, 3) visual acuity, and 4) surrogate outcomes. Most studies were retrospective (54.3 %) and aimed at attack prevention (81.4 %). The most common relapse-related outcomes were annualized relapse rate (73.2 %), followed by relapse rate (50.4 %), and relapse-free rate (36.2 %). The relapse definition also varied widely among studies, with only 73 (57.4 %) studies explicitly addressing the definition used. The most common disability outcome was the Expanded Disability Scale (97.6 %), followed by the Modified Rankin Scale (7.9 %). Visual Acuity Score was employed in 14.2 % of studies, followed by Visual Evoked Potentials (6.3 %). Imaging was the most common surrogate (20.5 %), followed by the fraction of B cells (18.1 %). ConclusionPublications were heterogeneous in measuring efficacy, with different use of endpoints and relapse definitions. Standardization across studies would improve data analysis and application in clinical practice.

Review about the Permutation Approach in Hypothesis Testing

Today, permutation tests represent a powerful and increasingly widespread tool of statistical inference for hypothesis-testing problems. To the best of our knowledge, a review of the application of permutation tests for complex data in practical data analysis for hypothesis testing is missing. In particular, it is essential to review the application of permutation tests in two-sample or multi-sample problems and in regression analysis. The aim of this paper is to consider the main scientific contributions on the subject of permutation methods for hypothesis testing in the mentioned fields. Notes on their use to address the problem of missing data and, in particular, right-censored data, will also be included. This review also tries to highlight the limits and advantages of the works cited with a critical eye and also to provide practical indications to researchers and practitioners who need to identify flexible and distribution-free solutions for the most disparate hypothesis-testing problems.

Estimation and Management of Wind Energy in Ikorodu Lagos Lagoon, Using Weibull Parametric Measurement to Electricity Production in Nigeria

This paper is an approach on the estimation and approximation management of wind energy production around the lagoon axis of Ikorodu, Lagos state, Nigeria. The article concentrates on the availability of renewable resource such as wind to generate electrical energy in the greater Ikorodu metropolis of Lagos state Nigeria. Here, probability distribution function is used to generate the wind data. In this paper, three distinct methods are presented; Data time series analysis, Weibull probability function, and theoretical comparison with analytical concept. This research uses two important parameters for analyzing wind data: shape factor “k” and scale factor “c” from Weibull distribution function. The theoretical uses mathematical equations of popular methods such as: (a) Moment method, (b) Empirical Formula, or statistical standard deviation, (c) Peak likelihood, (d) modified peak likelihood, (e) double modified peak likelihood (f) graphical method or smallest mean square, and (g) energy sequence factor. The results obtained are tested to optimize the value from the Weibull parameters by adopting five techniques: (i) root average square error methodology, X2, power of agreement , MAPE, and RRMSE. The results expatiated on the practical and theoretical techniques design to confront the outcome of wind energy harnessed per 1.5 km2. Here, a differential optimization technique is used to determine the precision report. This serve as the basis of error litmus check existing between the wind energy determined by theory of statistical and mathematical Weibull Parametric function and the practical time-series data analysis in LSTM. Again, the wind data (speed and energy/power) were measured and recorded between January 2020 to December 2023 in the Ijede-Ikorodu Lagoon area of Lagos State. The optimized value for the shape factor k and scale factor c parametric measurement and management for maximizing the output electrical energy are obtained by using a well robust Weibull distribution function techniques and by absolutely determining and selecting the best position and location for installing the wind/wave alternators/generator. These generators come with turbines as a single unit. The measurement of the yearly average wind speed and average wind power are 10.09 ms-1 and 10.1 KWm-2, concurrently.

88. Veterinary practice data analysis: A novel approach to measure the impact of coccidiosis on broiler health and welfare

88. Veterinary practice data analysis: A novel approach to measure the impact of coccidiosis on broiler health and welfare

SOCIAL AND ECONOMIC STATUS OF PUBLIC SERVANTS: EXAMINING THE CASES OF THE UK, CANADA, GERMANY, NEW ZEALAND, AND KAZAKHSTAN

Civil servants play a crucial role in the management and functioning of modern society. Their social and economic status reflects not only their well-being but also broader trends in government structures. This research article examines the socioeconomic status of civil servants in five different countries: Canada, Great Britain, Germany, New Zealand, and Kazakhstan. Through a comparative analysis of policy, practice, and empirical data, this study aims to clarify the factors influencing the status of civil servants, as well as their consequences on the implementation of public policy, the provision of public services, and the development of society. The purpose of the article is to determine the socioeconomic status of civil servants in five countries: Canada, Great Britain, Germany, New Zealand, and Kazakhstan, as well as to identify job satisfaction. The results show that civil servants in Canada, Germany, and New Zealand are more likely to be satisfied with their socio-economic status and service in the public service system. Those who work in the civil service in the UK and Kazakhstan are less satisfied with their work in the civil service system. The study also highlights the unique challenges and opportunities faced by civil servants in each country, emphasizing the importance of context-oriented approaches to enhance their status and promote effective public administration.

Client and employee engagement measures predict resignations of veterinary support staff: An analysis of UK companion animal practice data.

This study aimed to identify factors associated with veterinary support staff resigning from UK companion animal veterinary practices. Cross-sectional employment data from support staff working across a group of veterinary practices on 31 December 2020 were retrospectively analysed. The support staff were categorised as those who remained at, or resigned from, their practice over the following 12-month period. The proposed risk factors for future resignations were analysed using multivariable binary logistic regression modelling. In total, 398 of 2068 (19.2%) veterinary support staff resigned during the study period. Factors associated with greater odds of future resignations included lower staff engagement (employee net promoter score; odds ratio [OR] = 0.99; 95% confidence interval [CI] 0.99-1.00; p < 0.001), lower client practice engagement (net promoter score; OR = 0.98; 95% CI 0.97-1.00; p = 0.012), shorter tenure (OR = 0.92; 95% CI 0.90-0.95; p < 0.001), recent parental leave (OR = 2.37; 95% CI 1.38-4.07; p = 0.002) and younger age (OR = 0.99; 95% CI 0.98-1.00; p = 0.017). Data were collected retrospectively and were not recorded for research purposes. Employee and client net promoter scores are important predictors of resignations and could be useful tools for managers. Recent parental leave is a predictor of support staff attrition, and 'return to work' interventions may present an opportunity to improve retention.

Bayesian fractional polynomial approach to quantile regression and variable selection with application in the analysis of blood pressure among US adults

Although the fractional polynomials (FPs) can act as a concise and accurate formula for examining smooth relationships between response and predictors, modelling conditional mean functions observes the partial view of a distribution of response variable, as distributions of many response variables such as blood pressure (BP) measures are typically skew. Conditional quantile functions with FPs provide a comprehensive relationship between the response variable and its predictors, such as median and extremely high-BP measures that may be often required in practical data analysis generally. To the best of our knowledge, this is new in the literature. Therefore, in this article, we develop and employ Bayesian variable selection with quantile-dependent prior for the FP model to propose a Bayesian variable selection with parametric non-linear quantile regression model. The objective is to examine a non-linear relationship between BP measures and their risk factors across median and upper quantile levels using data extracted from the 2007 to 2008 National Health and Nutrition Examination Survey (NHANES). The variable selection in the model analysis identified that the non-linear terms of continuous variables (body mass index, age), and categorical variables (ethnicity, gender, and marital status) were selected as important predictors in the model across all quantile levels.

Leveraging Quadratic Polynomials in Python for Advanced Data Analysis

Objectives This study aims to provide a comprehensive overview of the role of quadratic polynomials in data modeling and analysis, particularly in representing the curvature of natural phenomena. Methods We begin with a fundamental explanation of quadratic polynomials and describe their general forms and theoretical significance. We then explored the application of these polynomials in regression analysis, detailing the process of fitting quadratic models to the data using Python libraries NumPy and Matplotlib. The methodology also included calculation of the coefficient of determination (R-squared) to evaluate the polynomial model fit. This study utilizes illustratively generated data to demonstrate the application of quadratic polynomials in Python for robust data analysis. Results Using practical examples accompanied by Python scripts, this study demonstrated the application of quadratic polynomials to analyze data patterns. These examples illustrate the utility of quadratic models in applied analytics. Conclusions This study bridges the gap between theoretical mathematical concepts and practical data analysis, thereby enhancing the understanding and interpretation of the data patterns. Furthermore, its implementation in Python, released under MIT’s license, offers an accessible tool for public use.

Lévy Flight Model of Gaze Trajectories to Assist in ADHD Diagnoses.

The precise mathematical description of gaze patterns remains a topic of ongoing debate, impacting the practical analysis of eye-tracking data. In this context, we present evidence supporting the appropriateness of a Lévy flight description for eye-gaze trajectories, emphasizing its beneficial scale-invariant properties. Our study focuses on utilizing these properties to aid in diagnosing Attention-Deficit and Hyperactivity Disorder (ADHD) in children, in conjunction with standard cognitive tests. Using this method, we found that the distribution of the characteristic exponent of Lévy flights statistically is different in children with ADHD. Furthermore, we observed that these children deviate from a strategy that is considered optimal for searching processes, in contrast to non-ADHD children. We focused on the case where both eye-tracking data and data from a cognitive test are present and show that the study of gaze patterns in children with ADHD can help in identifying this condition. Since eye-tracking data can be gathered during cognitive tests without needing extra time-consuming specific tasks, we argue that it is in a prime position to provide assistance in the arduous task of diagnosing ADHD.

The application of propensity score methods in observational studies

IntroductionIn research, it is crucial to accurately estimate treatment effects and analyze experimental results. Common methods include comparing outcome differences between different groups and using linear regression models for analysis. However, observational studies may have significantly different distributions of confounding variables between control and treatment groups, leading to errors in estimating treatment effects.MethodsThe propensity score methods can address this issue by weighting or matching samples to approximate the scenario of a randomized experiment and allow for more accurate estimation of treatment this paper.ResultsWe use propensity score methods to analyze three datasets from observational studies and draw conclusions different from those in the original text. Furthermore, we simulate three scenarios, and the results demonstrate the superiority of propensity score methods over methods such as linear regression in addressing selection bias.DiscussionTherefore, it is essential to thoroughly consider the characteristics of the data and select appropriate methods to ensure reliable conclusions in practical data analysis.

Features of the settlement of international cyber disputes through ADR in the context of the legislation of the BRICS countries

This paper examines the comparative effectiveness of various alternative dispute resolution (ADR) methods in resolving international cyber disputes. Drawing on a review of academic literature, analysis of practical case studies and statistical data, it identifies key procedural strengths and limitations of arbitration, mediation, ombudsmen, and online dispute resolution for common cyber conflict scenarios. It concludes on the optimal tailoring of different ADR techniques for cybercrime, hacking attacks, data breaches, and e-commerce disputes. The paper proposes multiple innovations to enhance cyber ADR efficacy, including hybrid models, specialized procedural standards, and enforcement mechanisms. It discusses integrating ADR into national cybersecurity strategies using the BRICS platform as an example. The research aims to inform optimization of flexible, confidential, and technically expert out-of-court approaches to manage the proliferation of cross-border cyber disputes. Key findings show mediation’s utility for cybercrime across jurisdictions but need for law enforcement coordination. International arbitration is appropriate for cyber B2B disputes while ombuds aid consumer recourse. Early neutral evaluation assists cybersecurity breach diagnosis but requires enforcement. Tailored arbitration rules, substantively flexible guidelines, and incentivizing voluntary ADR adoption are advised.

A Modified LSTM Neural Network Based on Boosting and Sufficient Dimensionality Reduction

Stock price prediction is a common scenario in time series data forecasting, providing effective guidance for investment decisions. Long Short-Term Memory (LSTM) is a widely used model for stock price prediction, yet the selection of its hyperparameters remains an unresolved issue. In this paper, we address this challenge by employing model averaging instead of model selection. Specifically, we adaptively solve the hyperparameter selection problem by utilizing a distance covariance-weighted method, effectively balancing the bias and variance of the predictive model. Additionally, we propose an enhanced model that employs a boosting approach based on sufficiently reducing dimensionality through a multifactor model. This approach captures stock price sequence information beyond volatility. Practical data analysis demonstrates that the proposed method exhibits significant advantages over the original LSTM model in terms of mean square error or absolute error. Furthermore, the proposed framework can be applied to hyperparameter selection in other time series prediction models, such as autoregressive integrated moving averages (ARIMA), including the selection of autoregressive and partial autocorrelation orders.

Investigation of the Impact of Work Incentives on Employee Work Quality

The theme of article is dedicated to investigating the impact of work incentives on the quality of employees’ work. During the research, various methods and approaches to incentivization are analyzed, and their effectiveness in terms of improving labor productivity accuracy and work quality is assessed. Special attention is given to the comparative analysis of material and non-material forms of incentives and their influence on employee motivation and satisfaction. The article is based on a theoretical literature review on the topic and practical data analysis obtained from various organizations. The research aims to identify the most effective methods of work incentives, ways to adapt to different organizational conditions, and individual employee needs. As a result of the work, specific recommendations for personnel management practices are presented, and criteria for evaluating the effectiveness of various incentive methods are determined. These recommendations and criteria can be used by organizations to enhance the quality of employee work and ensure the company’s development.

Использование Big Data при назначении и проведении судебно-экологических экспертиз

The article explores contemporary issues and questions related to the application of forensic environmental expertise within the framework of preliminary investigations, and evaluates its role in law enforcement activities. The research formulated and proposed the main tasks of forensic environmental expertise at the current stage, highlighted its features, and identified problems associated with the imperfection of the current methodological support for this type of forensic examination. Based on the analysis of empirical material and law enforcement practice data, the problem of insufficient awareness among investigators and inquiry officers regarding the current prospects for using specialized knowledge in investigating environmental crimes was identified. To address the identified problems and optimize the entire process of investigating this group of criminal cases, measures were proposed for the implementation of technologies based on the use of Big Data in the activities of law enforcement officers and expert institutions.

Practical Qualitative Data Analysis for Public Health Research: A Guide to a Team-Based Approach With Flexible Coding

Qualitative research is important to advance health equity as it offers nuanced insights into structural determinants of health inequities, amplifies the voices of communities directly affected by health inequities, and informs community-based interventions. The scale and frequency of public health crises have accelerated in recent years (e.g., pandemic, environmental disasters, climate change). The field of public health research and practice would benefit from timely and time-sensitive qualitative inquiries for which a practical approach to qualitative data analysis (QDA) is needed. One useful QDA approach stemming from sociology is flexible coding. We discuss our practical experience with a team-based approach using flexible coding for qualitative data analysis in public health, illustrating how this process can be applied to address multiple research questions simultaneously or asynchronously. We share lessons from this case study, while acknowledging that flexible coding has broader applicability across disciplines. Flexible coding provides an approachable step-by-step process that enables collaboration among coders of varying levels of experience to analyze large datasets. It also serves as a valuable training tool for novice coders, something urgently needed in public health. The structuring enabled through flexible coding allows for prioritizing urgent research questions, while preparing large datasets to be revisited many times, facilitating secondary analysis. We further discuss the benefit of flexible coding for increasing the reliability of results through active engagement with the data and the production of multiple analytical outputs.

Research on classroom teaching innovation based on “BOPPPS+Chaoxing On-line Learning Platform”-Take “Linux operating system” as an example

The purpose of this paper is to explore the application and effect of classroom teaching innvation based on "BOPPPS+ChaoXing on-line Plant" in the course of “Linux Operating System”. By introducing the BOPPPS teaching mode and using the “Chaoxing On-line Learning Platform” platform, this paper aims to improve students' learning effect, enhance classroom interaction and promote the improvement of teaching quality. This paper first introduces the basic concepts of BOPPPS teaching mode and “Chaoxing On-line Learning Platform”, and then expounds in detail the process of implementing this teaching innvation in the course of “Linux Operating System”, and evaluates the effect of the innvation through practical cases and data analysis. Finally, this paper summarizes the effect of teaching innvation and puts forward the further improvement direction.

A characteristic extraction method for VoicePrint slice statistics base on joint time-frequency processing

Mel-Frequency Cepstral Coefficients (MFCC) and its differential coefficients are widely used as a typical non-linear spectral envelope feature in passive sonar target recognition. MFCC emphasizes the frequency domain information at the low-frequency end of the target, while the differential coefficients add dynamic characteristics of MFCC in the time domain. However, the utilization of time information is not sufficient. In response to this issue, we propose two solutions: one is the characteristic fusion method of Weighted Mel-Frequency Cepstral Coefficient based on Bhattacharyya Distance (BD-WMFCC); the other is the characteristic extraction method of VoicePrint Slice Statistics (VPSS) based on time–frequency joint processing. BD-WMFCC combines weighted differential coefficients with MFCC using Bhattacharyya distance, maximizing the utilization of time information. VPSS is a linear spectral feature that utilizes the stability of the spectral intensity within a certain bandwidth over continuous time. It extracts the steady-state characteristics of the target's spectrum with respect to time changes at equidistant bandwidths. Simulation and real data analysis have shown that VPSS demonstrates outstanding separability and classification performance in both non-linear and linear spectral features, with strong stability. In practical data analysis, the accuracy of VPSS features under the AlexNet algorithm is 3.8 % and 5.7 % higher than the spectral (Fre) and Mel features, respectively.

Efficient model selection for predictive pattern mining model by safe pattern pruning

Predictive pattern mining is an approach used to construct prediction models when the input is represented by structured data, such as sets, graphs, and sequences. The main idea behind predictive pattern mining is to build a prediction model by considering unified inconsistent notation sub-structures, such as subsets, subgraphs, and subsequences (referred to as patterns), present in the structured data as features of the model. The primary challenge in predictive pattern mining lies in the exponential growth of the number of patterns with the complexity of the structured data. In this study, we propose the safe pattern pruning method to address the explosion of pattern numbers in predictive pattern mining. We also discuss how it can be effectively employed throughout the entire model building process in practical data analysis. To demonstrate the effectiveness of the proposed method, we conduct numerical experiments on regression and classification problems involving sets, graphs, and sequences.