Traditional Statistical Approaches Research Articles

Domain Adaptation-Based Machine Learning Framework for Customer Churn Prediction Across Varing Distributions

In today’s fiercely competitive business environment, the ability to accurately predict customer churn is essential for enhancing customer retention and reducing financial losses. Traditional statistical approaches, although beneficial, often struggle to perform effectively across diverse customer data domains due to variability in data distributions. This research introduces a refined method for predicting customer churn that utilizes domain adaptation techniques to overcome these challenges. Specifically, it employs the Correlation Alignment (CORAL) method to synchronize the feature distributions between the source and target datasets, significantly improving the logistic regression model’s capacity to apply insights across various customer segments. The process involves segmenting the customer data into clearly defined clusters using the k-means algorithm, which helps pinpoint and adjust distributional discrepancies, thus boosting the model’s accuracy. Early results indicate that incorporating domain adaptation not only bolsters the model’s applicability across different domains but also drastically minimizes the covariance differences—from a substantial initial gap to nearly zero. This strategic approach demonstrates substantial potential to revolutionize how businesses anticipate and manage customer behavior, providing a more adaptable and effective framework for addressing the complex challenges of customer churn.

Identifying infrastructural gap areas for smart and sustainable tribal village development: A data science approach from India

In the present era of rapid economic development, successive governments in India have placed priority on ‘inclusive economic growth and social justice’. After independence, India chose to adopt mixed economy model and central role for development planning and resource allocation across sectors have been done by the State's planning machinery. It has been the resolve of the Government of India to achieve development and welfare of all social groups including scheduled tribes in the country. In this pursuit, several tribal development programmes have been undertaken from time to time entailing enormous human, financial and material resources. However, data driven planning involving data analytics has not been in use for overall planning process for tribal development instead traditional statistical approach has been mostly in practice. The aim of this paper is to bring to the fore the usefulness of data driven gap based planning using data analytics tools. The endeavour is to put in place a scientific mechanism to identify and prioritize the infrastructural gaps in tribal dominated areas for the State of Telangana using twenty-two different parameters. Expert workshop conducted to cluster twenty-two parameters into seven broad clusters viz Finance Infrastructure, Village Material Infrastructure, Household Material Infrastructure, Information Infrastructure, Social Infrastructure, Education Infrastructure, Health Infrastructure. Moreover, the paper aims to connect the smart and sustainable village development considering various parameters such Information and Social Infrastructure. AHP and TOPSIS multi-criteria approaches have been applied to identify and prioritize the infrastructural gap areas for all nine districts and 1663 tribal dominated villages of Telangana. Analysis reveals that Mahbubnagar being the top ranked district while Adilabad being lowest ranked district in the Telangana in terms of gaps in infrastructure facilities. Villages in all nine districts have been clustered into three categories viz. Good, Fair and Poor. Findings and recommendations have been provided in the paper.

Machine Learning in Pain Medicine: An Up-To-Date Systematic Review.

IntroductionPain is the unpleasant sensation and emotional experience that leads to poor quality of life for millions of people worldwide. Considering the complexity in understanding the principles of pain and its significant impact on individuals and society, research focuses to deliver innovative pain relief methods and techniques. This review explores the clinical uses of machine learning (ML) for the diagnosis, classification, and management of pain.MethodsA systematic review of the current literature was conducted using the PubMed database library.ResultsTwenty-six papers related to pain and ML research were included. Most of the studies used ML for effectively classifying the patients’ level of pain, followed by use of ML for the prediction of manifestation of pain and for pain management. A less common reason for performing ML analysis was for the diagnosis of pain. The different approaches are thoroughly discussed.ConclusionML is increasingly used in pain medicine and appears to be more effective compared to traditional statistical approaches in the diagnosis, classification, and management of pain.

How Would We Cycle Today If We Had the Weather of Tomorrow? An Analysis of the Impact of Climate Change on Bicycle Traffic

Bicycle usage is significantly affected by weather conditions. Climate change is, therefore, expected to have an impact on the volume of bicycle traffic, which is an important factor in the planning and design of bicycle infrastructures. To predict bicycle traffic in a changed climate in the city of Berlin, this paper compares a traditional statistical approach to three machine learning models. For this purpose, a cross-validation procedure is developed that evaluates model performance on the basis of prediction accuracy. XGBoost showed the best performance and is used for the prediction of bicycle counts. Our results indicate that we can expect an overall annual increase in bicycle traffic of 1–4% in the city of Berlin due to the changes in local weather conditions caused by global climate change. The biggest changes are expected to occur in the winter season with increases of 11–14% due to rising temperatures and only slight increases in precipitation.

Characterizing the evolution of the Yangtze River Delta multi-port system using compositional data techniques

ABSTRACT The Yangtze River Delta multi-port system (YMPS) receives increasing attention due to its relevance in world trade and excellent competitiveness in the container traffic market. To insight into the development pattern of YMPS, this paper proposes a method that combines the Hierarchical Clustering with compositional data (CoDa) exploratory tools (i.e., biplot and dendrogram) to explore the temporal and spatial evolution of the YMPS from 1992 to 2019. CoDa describes parts of some whole (i.e., frequency and percentage), conveying relative information in the ratios between its components. Container traffic share in a multi-port region is typical CoDa. Traditional statistical approaches to CoDa could lead to spurious correlations and erroneous conclusions. However, using suitable CoDa techniques, such as the centered log-ratio (clr) transformation, can effectively avoid these misinterpretations. The novel method can simultaneously find the temporal and spatial characteristics. The findings indicate that the development of the YMPS has gone through four stages and the evolution of the YMPS is characterized by a tendency towards a ‘multi-core development’ and faces a differentiated pattern of ‘peripheral port challenges’. The analysis further improves the port system’s evolutionary model and explains the underlying reason for the YMPS development. CoDa techniques also provide a new perspective for the temporal and spatial evolution of the transport discipline.

Performance measurement for offline inspections under variable interactions and inspection errors in low-volume production

The assessment of the performance of inspection strategies is a crucial element in the design phase of product quality inspections of manufacturing companies. The aspects that inspection designers need to consider include: (1) the typology of quality inspection, (2) the inspection variables involved, (3) the potential interaction between variables and (4) the presence of inspection errors. In particular, low-volume inspection design is critical due to the lack of historical data and the inadequacy of traditional statistical approaches. By considering these issues, this paper proposes a novel approach to support inspection designers in the prediction of offline quality inspection performance. The development of a probabilistic model based on the analysis of the possible variable interactions and inspection errors and the definition of some performance measures may successfully help designers in the early design stages of inspection process planning. The approach is supported by a practical application in the Additive Manufacturing field.

1023The uses and abuses of regression models: a new approach to teaching regression analysis

Abstract Focus of Presentation Multivariable regression models are widely used in epidemiological data analysis. Traditional teaching often focusses on technical aspects with insufficient attention paid to the purposes for which regression methods are used. Findings We have addressed these issues in a new short course that provides an introduction to regression analysis in the context of learning about causal inference, beginning from the standpoint that the majority of research questions in epidemiology are causal in nature. This approach leads naturally to using regression models in two different ways, firstly for direct estimation of a causal effect, under an assumption of constancy of the effect across strata of confounders, and secondly for prediction of outcomes, as a necessary step in the estimation of causal effects via g-computation. Conclusions/Implications Approaching the teaching of regression methods within a causal inference framework helps to dispel confusion created by traditional statistical approaches that imply the existence of “true models” and encourage the building of models in a way that is unclear about the purpose for which they will be used, for example seeking to identify “risk factors” in an exploratory manner. Key messages The teaching and practice of regression methods in epidemiology can be enhanced by emphasising the key differences between three distinct analytic purposes: description, prediction, causal. Regression models may play a role in all three but the way in which models are developed and interpreted differs between them.

A Novel Framework for Mining Social Media Data Based on Text Mining, Topic Modeling, Random Forest, and DANP Methods

The huge volume of user-generated data on social media is the result of the aggregation of users’ personal backgrounds, past experiences, and daily activities. This huge size of the generated data, the so-called “big data,” has been studied and investigated intensively during the past few years. In spite of the impression one may get from the media, a great deal of data processing has not been uncovered by existing techniques of data engineering and processing. However, very few scholars have tried to do so, especially from the perspective of multiple-criteria decision-making (MCDM). These MCDM methods can derive influence relationships and weights associated with aspects and criteria, which can hardly be achieved by traditional data analytics and statistical approaches. Therefore, in this paper, we aim to propose an analytic framework to mine social networks, feed the meaningful information via MCDM methods based on a theoretical framework, derive causal relationships among the aspects of the theoretical framework, and finally compare the causal relationships with a social theory. Latent Dirichlet allocation (LDA) will be adopted to derive topic models based on the data retrieved from social media. By clustering the topics into aspects of the social theory, the probability associated with each aspect will be normalized and then transformed to a Likert-type 5-point scale. Afterwards, for every topic, the feature importance of all other topics will be derived using the random forest (RF) algorithm. The feature importance matrix will be transformed to the initial influence matrix of the decision-making trial and evaluation laboratory (DEMATEL). The influence relationships among the aspects and criteria and influence weights can then be derived by using the DEMATEL-based analytic network process (DANP). The influence weight versus each criterion can be derived by using DANP. To verify the feasibility of the proposed framework, Taiwanese users’ attitudes toward air pollution will be analyzed based on the value–belief–norm (VBN) theory by using social media data retrieved from Dcard (dcard.tw). Based on the analytic results, the causal relationships are fully consistent with the VBN framework. Further, the mutual influences derived in this work that were seldom discussed by earlier works, i.e., the mutual influences between altruistic concerns and egoistic concerns, as well as those between altruistic concerns and biosphere concerns, are worth further investigation in future.

Machine learning approaches in predicting ambulatory same day discharge patients after total hip arthroplasty

BackgroundWith continuing financial and regulatory pressures, practice of ambulatory total hip arthroplasty is increasing. However, studies focusing on selection of optimal candidates are burdened by limitations related to traditional statistical...

BLC mini-series: New statistical approaches and research practices for bilingualism research

The past decade has witnessed some dramatic methodological changes in the wider disciplines of psycholinguistics, psychology, and experimental linguistics. One such set of changes comprises the development of open and transparent research practices, which have increasingly been adopted in response to concerns that empirical results often fail to replicate and may not generalise across samples and experimental conditions (Gibson & Fedorenko, 2013; Maxwell, Lau, & Howard, 2015; McElreath & Smaldino, 2015; Yarkoni, 2020). Another important set of changes concerns the use of sophisticated statistical techniques, such as mixed-effects models (Baayen, Davidson, & Bates, 2008) and Bayesian analyses (Vasishth, Nicenboim, Beckman, Li & Kong, 2018), which can provide much more information about magnitudes of effects and sources of variation than the more traditional statistical approaches.

Radioactive waste sampling for characterisation - A Bayesian upgrade

Presented in this paper is a methodology for combining a Bayesian statistical approach with Data Quality Objectives (a structured decision-making method) to provide increased levels of confidence in analytical data when approaching a waste boundary. Development of sampling and analysis plans for the characterisation of radioactive waste often use a simple, one pass statistical approach as underpinning for the sampling schedule. Using a Bayesian statistical approach introduces the concept of Prior information giving an adaptive sample strategy based on previous knowledge. This aligns more closely with the iterative approach demanded of the most commonly used structured decision-making tool in this area (Data Quality Objectives) and the potential to provide a more fully underpinned justification than the more traditional statistical approach. The approach described has been developed in a UK regulatory context but is translated to a waste stream from the Fukushima Daiichi Nuclear Power Station to demonstrate how the methodology can be applied in this context to support decision making regarding the ultimate disposal option for radioactive waste in a more global context.

Abstract 165: Enhanced processing of genomic sequencing data for pediatric cancers: GPUs and machine learning techniques for variant detection

Abstract Introduction: The evolution of precision oncology will demand the integration of multi-omics, clinical data, demographics and outcomes, which is a complex and computationally intensive endeavor. Accurate and robust molecular characterization of pediatric leukemias require a comprehensive approach involving longitudinal sample collection across a large cohort. Genomic alterations, both germline and somatic, need to be efficiently captured and annotated to identify clinically meaningful trends. Algorithms developed for conducting genomic analysis typically leverage a central processing unit (CPU) environment for data processing and traditional statistical approach, such as Bayes' theorem, for variant detection. More recently, algorithms have been developed that can leverage a graphics processing unit (GPU) architecture and machine learning (ML) technology. In this study we benchmarked these techniques and applied them sequencing data generated from pediatric AML subjects enrolled on the Children's Oncology Group AAML1031 trial (951 samples). Methods: Whole genome sequencing was conducted using the Illumina platform (2X150) and sequenced between 30x-60x coverage. Genome alignment files were generated and subjected to germline and somatic variant calling (NVIDIA's Selene DGXA100 cluster; Parabricks Pipelines). Germline variants were called using Haplotype caller and DeepVariant (Parabricks v3.0). HG002 genome in a bottle sample was used as a positive control to determine precision, recall and F1 score. Somatic variant calling was completed using somatic sniper, VarScan2, Muse, MuTect2, and Strelka (Parabricks v3.5) and SEQCII data was used to assess somatic variant calling performance. Results: Control sample HG002 was used for benchmarking the algorithms tested. At all depths analyzed, 50X, 30X, 15X, and 5X, GoogleDeep Variant had the highest F1 score, precision, and recall compared to Haplotyper for SNPs and InDels. For somatic SNV detection using SEQCII dataset, Strelka2 had the highest F1 score, precision, and recall (0.9435, 0.9445, 0.9425), followed by MuTect2, MUSE, VarScan2, and SomaticSniper. For somatic InDel detection, MuTect2 and Strelka2 performed similar with MuTect2 higher precision, but Strelka2 had higher recall. These pipelines were utilized to analyze 951 pediatric AML samples, and demonstrated superior processing time and identification of variants, both germline and somatic. Discussion: ML algorithms integrated into a GPU computing environment demonstrated greater precision and accuracy for germline variant detection compared to traditional statistical approaches for variant detection. Ongoing efforts are focused at integration of multiple data sources correlated with outcomes and analyzed via ML tools in order to improve future risk stratification and therapeutic selection for superior outcomes. Citation Format: Erin L. Crowgey, Pankaj Vats, Karl Franke, Gary Burnett, Ankit Sethia, Timothy Harkins, Todd E. Druley. Enhanced processing of genomic sequencing data for pediatric cancers: GPUs and machine learning techniques for variant detection [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 165.

Industrial Forecasting with Exponentially Smoothed Recurrent Neural Networks

Time series modeling has entered an era of unprecedented growth in the size and complexity of data which require new modeling approaches. While many new general purpose machine learning approaches have emerged, they remain poorly understand and irreconcilable with more traditional statistical modeling approaches. We present a general class of exponential smoothed recurrent neural networks (RNNs) which are well suited to modeling nonstationary dynamical systems arising in industrial applications. In particular, we analyze their capacity to characterize the nonlinear partial autocorrelation structure of time series and directly capture dynamic effects such as seasonality and trends. Application of exponentially smoothed RNNs to forecasting electricity load, weather data, and stock prices highlight the efficacy of exponential smoothing of the hidden state for multistep time series forecasting. The results also suggest that popular, but more complicated neural network architectures originally designed for speech processing are likely over-engineered for industrial forecasting and light-weight exponentially smoothed architectures, trained in a fraction of the time, capture the salient features while being superior and more robust than simple RNNs and autoregressive models. Additionally, uncertainty quantification of Bayesian exponential smoothed RNNs is shown to provide improved coverage.

New Synonyms Extraction Model Based on a Novel Terms Weighting Scheme

The traditional statistical approach in synonyms extraction is time-consuming. It is necessary to develop a new method to improve the efficiency and accuracy. This research presents a new method in synonyms extraction called Noun Based Distinctive Verbs (NBDV) that replaces the traditional tf-idf weighting scheme with a new weighting scheme called the Orbit Weighing Scheme (OWS). The OWS links the nouns to their semantic space by examining the singular verbs in each context. The new method was compared with important models in the field such as the Skip-Gram, the Continuous Bag of Words, and the GloVe model. The NBDV model was manipulated over the Arabic and English languages and the results showed 47% Recall and 51% Precision in the dictionary-based evaluation and 57.5% Precision in the human experts’ evaluation. Comparing with the synonyms extraction based on tf.idf, the NBDV obtained 11% higher recall and 10% higher precision. Regarding the efficiency, we found that on average, the synonyms extraction of a single noun requires the process of 186 verbs and in 63% of the runs; the number of singular verbs was less than 200. It is concluded that the developed method is efficient and processes the single run in linear time.

A deep dive into fat: Investigating blubber lipidomic fingerprint of killer whales and humpback whales in northern Norway.

In cetaceans, blubber is the primary and largest lipid body reservoir. Our current understanding about lipid stores and uses in cetaceans is still limited, and most studies only focused on a single narrow snapshot of the lipidome. We documented an extended lipidomic fingerprint in two cetacean species present in northern Norway during wintertime. We were able to detect 817 molecular lipid species in blubber of killer whales (Orcinus orca) and humpback whales (Megaptera novaeangliae). The profiles were largely dominated by triradylglycerols in both species and, to a lesser extent, by other constituents including glycerophosphocholines, phosphosphingolipids, glycerophosphoethanolamines, and diradylglycerols. Through a unique combination of traditional statistical approaches, together with a novel bioinformatic tool (LION/web), we showed contrasting fingerprint composition between species. The higher content of triradylglycerols in humpback whales is necessary to fuel their upcoming half a year fasting and energy‐demanding migration between feeding and breeding grounds. In adipocytes, we assume that the intense feeding rate of humpback whales prior to migration translates into an important accumulation of triacylglycerol content in lipid droplets. Upstream, the endoplasmic reticulum is operating at full capacity to supply acute lipid storage, consistent with the reported enrichment of glycerophosphocholines in humpback whales, major components of the endoplasmic reticulum. There was also an enrichment of membrane components, which translates into higher sphingolipid content in the lipidome of killer whales, potentially as a structural adaptation for their higher hydrodynamic performance. Finally, the presence of both lipid‐enriched and lipid‐depleted individuals within the killer whale population in Norway suggests dietary specialization, consistent with significant differences in δ15N and δ13C isotopic ratios in skin between the two groups, with higher values and a wider niche for the lipid‐enriched individuals. Results suggest the lipid‐depleted killer whales were herring specialists, while the lipid‐enriched individuals might feed on both herrings and seals.

Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid

The prediction of time series data applied to the energy sector (prediction of renewable energy production, forecasting prosumers’ consumption/generation, forecast of country-level consumption, etc.) has numerous useful applications. Nevertheless, the complexity and non-linear behaviour associated with such kind of energy systems hinder the development of accurate algorithms. In such a context, this paper investigates the use of a state-of-art deep learning architecture in order to perform precise load demand forecasting 24-h-ahead in the whole country of France using RTE data. To this end, the authors propose an encoder-decoder architecture inspired by WaveNet, a deep generative model initially designed by Google DeepMind for raw audio waveforms. WaveNet uses dilated causal convolutions and skip-connection to utilise long-term information. This kind of novel ML architecture presents different advantages regarding other statistical algorithms. On the one hand, the proposed deep learning model’s training process can be parallelized in GPUs, which is an advantage in terms of training times compared to recurrent networks. On the other hand, the model prevents degradations problems (explosions and vanishing gradients) due to the residual connections. In addition, this model can learn from an input sequence to produce a forecast sequence in a one-shot manner. For comparison purposes, a comparative analysis between the most performing state-of-art deep learning models and traditional statistical approaches is presented: Autoregressive-Integrated Moving Average (ARIMA), Long-Short-Term-Memory, Gated-Recurrent-Unit (GRU), Multi-Layer Perceptron (MLP), causal 1D-Convolutional Neural Networks (1D-CNN) and ConvLSTM (Encoder-Decoder). The values of the evaluation indicators reveal that WaveNet exhibits superior performance in both forecasting accuracy and robustness.

Machine Learning to Classify Suicidal Thoughts and Behaviors: Implementation Within the Common Data Elements Used by the Military Suicide Research Consortium

Suicide rates among military-connected populations have increased over the past 15 years. Meta-analytic studies indicate prediction of suicide outcomes is lacking. Machine-learning approaches have been promoted to enhance classification models for suicide-related outcomes. In the present study, we compared the performance of three primary machine-learning approaches (i.e., elastic net, random forests, stacked ensembles) and a traditional statistical approach, generalized linear modeling (i.e., logistic regression), to classify suicide thoughts and behaviors using data from the Military Suicide Research Consortium’s Common Data Elements (CDE; n = 5,977–6,058 across outcomes). Models were informed by (a) selected items from the CDE or (b) factor scores based on exploratory and confirmatory factor analyses on the selected CDE items. Results indicated similar classification performance across models and sets of features. In this study, we suggest the need for robust evidence before adopting more complex classification models and identify measures that are particularly relevant in classifying suicide-related outcomes.

A Network-Based Analysis of a Worksite Canteen Dataset

The provision of wellness in workplaces gained interest in recent decades. A factor that contributes significantly to workers’ health is their diet, especially when provided by canteen services. The assessment of such a service involves questions as food cost, its sustainability, quality, nutritional facts and variety, as well as employees’ health and disease prevention, productivity increase, economic convenience vs. eating satisfaction when using canteen services. Even if food habits have already been studied using traditional statistical approaches, here we adopt an approach based on Network Science that allows us to deeply study, for instance, the interconnections among people, company and meals and that can be easily used for further analysis. In particular, this work concerns a multi-company dataset of workers and dishes they chose at a canteen worksite. We study eating habits and health consequences, also considering the presence of different companies and the corresponding contact network among workers. The macro-nutrient content and caloric values assessment is carried out both for dishes and for employees, in order to establish when food is balanced and healthy. Moreover, network analysis lets us discover hidden correlations among people and the environment, as communities that cannot be usually inferred with traditional or methods since they are not known a priori. Finally, we represent the dataset as a tripartite network to investigate relationships between companies, people, and dishes. In particular, the so-called network projections can be extracted, each one being a network among specific kind of nodes; further community analysis tools will provide hidden information about people and their food habits. In summary, the contribution of the paper is twofold: it provides a study of a real dataset spanning over several years that gives a new interesting point of view on food habits and healthcare, and it also proposes a new approach based on Network Science. Results prove that this kind of analysis can provide significant information that complements other traditional methodologies.

A machine learning approach to predict healthcare-associated infections at intensive care unit admission: findings from the SPIN-UTI project

Identifying patients at higher risk of healthcare-associated infections (HAIs) in intensive care units (ICUs) represents a major challenge for public health. Machine learning could improve patient risk stratification and lead to targeted infection prevention and control interventions. To evaluate the performance of the Simplified Acute Physiology Score (SAPS) II for HAI risk prediction in ICUs, using both traditional statistical and machine learning approaches. Data for 7827 patients from the 'Italian Nosocomial Infections Surveillance in Intensive Care Units' project were used in this study. The Support Vector Machines (SVM) algorithm was applied to classify patients according to sex, patient origin, non-surgical treatment for acute coronary disease, surgical intervention, SAPS II at admission, presence of invasive devices, trauma, impaired immunity, and antibiotic therapy in 48 h preceding ICU admission. The performance of SAPS II for predicting HAI risk provides a receiver operating characteristic curve with an area under the curve of 0.612 (P<0.001) and accuracy of 56%. Considering SAPS II along with other characteristics at ICU admission, the SVM classifier was found to have accuracy of 88% and an AUC of 0.90 (P<0.001) for the test set. The predictive ability was lower when considering the same SVM model but with the SAPS II variable removed (accuracy 78%, AUC 0.66). This study suggested that the SVM model is a useful tool for early prediction of patients at higher risk of HAIs at ICU admission.

Modeling Residential Energy Consumption

Smart meters that allow information to flow between users and utility service providers are expected to foster intelligent energy consumption. Previous studies focusing on demand-side management have been predominantly restricted to factors that utilities can manage and manipulate, but have ignored factors specific to residential characteristics. They also often presume that households consume similar amounts of energy and electricity. To fill these gaps in literature, the authors investigate two research questions: (RQ1) Does a data mining approach outperform traditional statistical approaches for modelling residential energy consumption? (RQ2) What factors influence household energy consumption? They identify household clusters to explore the underlying factors central to understanding electricity consumption behavior. Different clusters carry specific contextual nuances needed for fully understanding consumption behavior. The findings indicate electricity can be distributed according to the needs of six distinct clusters and that utilities can use analytics to identify load profiles for greater energy efficiency.