Data Science Algorithms Research Articles

Towards a System Dynamics Framework for Human–Machine Learning Decisions: A Case Study of New York Citi Bike

The growing number of algorithmic decision-making environments, which blend machine and bounded human rationality, strengthen the need for a holistic performance assessment of such systems. Indeed, this combination amplifies the risk of local rationality, necessitating a robust evaluation framework. We propose a novel simulation-based model to quantify algorithmic interventions within organisational contexts, combining causal modelling and data science algorithms. To test our framework’s viability, we present a case study based on a bike-share system focusing on inventory balancing through crowdsourced user actions. Utilising New York’s Citi Bike service data, we highlight the frequent misalignment between incentives and their necessity. Our model examines the interaction dynamics between user and service provider rule-driven responses and algorithms predicting flow rates. This examination demonstrates why understanding these dynamics is essential for devising effective incentive policies. The study showcases how sophisticated machine learning models, with the ability to forecast underlying market demands unconstrained by historical supply issues, can cause imbalances that induce user behaviour, potentially spoiling plans without timely interventions. Our approach allows problems to surface during the design phase, potentially avoiding costly deployment errors in the joint performance of human and AI decision-makers.

MACHINE LEARNING FOR ENHANCED CLASSROOM HOMOGENEITY IN PRIMARY EDUCATION

A homogeneous distribution of students in a class is accepted as a key factor for overall success in primary education. A class of students with similar attributes normally increases academic success. It is also a fact that general academic success might be lower in some classes where students have different intelligence and academic levels. In this study, a class distribution model is proposed by using some data science algorithms over a small number of students’ dataset. With unsupervised and semi supervised learning methods in machine learning and data mining, a group of students is equally distributed to classes, taking into account some criteria. This model divides a group of students into clusters by the considering students’ different qualitative and quantitative characteristics. A draft study is carried out by predicting the effectiveness and efficiency of the presented approaches. In addition, some process elements such as quantitative and qualitative characteristics of a student, data acquisition style, digitalization of attributes, and creating a future prediction are also included in this study. Satisfactory and promising experimental results are received using a set of algorithms over collected datasets for classroom scenarios. As expected, a clear and concrete evaluation between balanced and unbalanced class distributions cannot be performed since these two scenarios for the class distributions cannot be applicable at the same time.

An asymptotically exact estimate of the median noise eigenvalue of sample covariance matrices

The Dominant Mode Rejection beamformer [Abraham & Owsley (1990)] averages the noise subspace eigenvalues of the sample covariance matrix (SCM) to estimate the background noise power. This noise power estimate from snapshot deficient SCMs can be unreliable for large arrays in time-varying environments with limited snapshots. Median filtering offers robustness to outliers and subspace mismatch for these challenging problems. Recent numerical experiments [Campos Anchieta & Buck (2022)] identified a simple regression relating the median sample eigenvalue to the true background power for snapshot deficient SCMs. However, the complicated expression of the Marchenko-Pastur (MP) distribution for the noise eigenvalues of the SCM thwarted prior attempts to find a closed form estimator for the MP distribution median. This talk exploits a coordinate transform to obtain a closed-form median of the MP distribution that is exact for the leading term in the power series expansion of the distribution. The new median estimator is more accurate than the previous numerical regression when the ratio of sensors to snapshots is 2 or more. Given the central role of SCM eigenvalues in principal component analysis and constant false alarm rate detectors, the new median expression should find application in other data science algorithms for underwater acoustics. [Work supported by ONR Code 321US.]

Dirençli Punta Kaynağı Prosesinin KNN ve CART Makine Öğrenimi Teknikleri ile Değerlendirilmesi

Spot welding, a type of resistance welding, is a welding application widely used in the production area and it is a common method for joining metal sheets. The spot-welding process is widely used in many production areas, especially in the automotive industry, radiator, and wire mesh production. Spot welding in car production lines is mainly performed by robotic applications. Industry 4.0 and digital transformation trends have led to unprecedented data growth. Nowadays, the manufacturing industry benefits from the power of machine learning and data science algorithms to monitor production processes and make predictions for quality, maintenance, and production optimization. Applying machine learning algorithms reduces the duration and cost of experiments. This study aims to confirm whether the spot welding, applied by robotic arms, is within the ideal spot-welding norms, in real production area. The ideal parameter norms were evaluated by using KNN and CART machine learning algorithms. To use real production data, this study was executed in the body production assembly line, which is selected as the pilot area, at TOFAŞ factory. The data set used in this research consists of the welding parameters of the current year, 2023. By running machine learning algorithms on the dataset, the performance evaluation of each algorithm was examined and the most appropriate estimation method was determined. In the experiments, the best F1-Score value was obtained by the CART model with 93%.

Autonomous screening of complex phase spaces using Bayesian optimization for SAXS measurements

The advent of modern, ultrafast X-ray experiments has enabled scientists to probe physical phenomena at an ever smaller scale. However, this has come at a cost of excessive data generation, to the point where current storage and hardware capabilities are routinely surpassed. As such, handling the data efficiently and selectively storing only the information of most relevance is crucial. In this paper, we propose to use Bayesian optimization as a method to alleviate this problem. We apply the method to locate global features in Small Angle X-ray Scattering spectra obtained from conducting experiments with supercritical CO2. By evaluating the algorithm on more than 250 data points, we show that the implementation is versatile, robust, and computationally efficient, often converging with just a few iterations and with a minimal error penalty. This paves the way for creating fully autonomous experiments, where data science algorithms such as the one presented herein operate hand-in-hand with the expert user to maximize scientific discovery and minimize the associated experimental cost.

Fabrication and characterization of cobalt- and copper-doped mesoporous borate bioactive glasses for potential applications in tissue engineering

Developing novel compositions of bioactive glasses (BGs) is key for accelerating tissue repair and regeneration. In this work, we developed a series of cobalt (Co)- and copper (Cu)-doped mesoporous bioactive glasses (MBGs) based on borate 13-93B3 composition using nitrate precursors. We took benefit from data science algorithms to predict and assess the physico-chemical and biological properties of the samples. The results showed that the presence of the dopants (Co and Cu) in the MBGs could change the glass transition temperature (Tg) (from 773 to 539 °C), the zeta potential (from -12 to -43 mV), and surface area (from 54 to 194 m2/g). However, the presence of 2.5 mol% of dopants in the composition led to just a slight decrease in their bioactivity. In vitro biocompatibility assays confirmed that all the glass samples were biocompatible. Furthermore, the doped MBGs exhibited potent antibacterial activity against both Gram-positive and Gram-negative bacteria. In addition, these glasses could induce the mobility of human umbilical vein endothelial cells (HUVECs) and enhance new blood vessel formation in ovo. According to the obtained data, it can be stated that this type of doped borate MBGs held great promise in tissue engineering applications.

Development of a Classification Model for Predicting Student Payment Behavior Using Artificial Intelligence and Data Science Techniques

Artificial intelligence today has become a valuable tool for decision-making, where universities have to adapt and optimize their processes, improving the quality of their services. In this context, the economic income from collections is vital for sustainability. There are several problems that can contribute to student delinquency, such as economic, financial, academic, family, and personal. For this reason, the study aimed to develop a classification model to predict the payment behavior of enrolled students. The methodology is a proactive, technological study of incremental innovation with a synchronous temporal scope. The study population consisted of 8,495 undergraduate students enrolled in the 2022 - II academic semester, containing information on academic performance, financial situation, and personal factors. The result is a classification model using the H2O.ai platform, discretization algorithms, data balancing, and the R language. Data science algorithms obtained the base from the institution's computer system. The data sets for training and testing correspond to 70% and 30%, obtaining the GBM Grid model whose performance metrics are AUC of 0.905, AUCPR of 0.926, and logLoss equivalent to 0.311; that is, the model efficiently complies with the classification of student debtors to provide them with early intervention service and help them complete their studies.

Quantification of motion during microvascular anastomosis simulation using machine learning hand detection.

Microanastomosis is one of the most technically demanding and important microsurgical skills for a neurosurgeon. A hand motion detector based on machine learning tracking technology was developed and implemented for performance assessment during microvascular anastomosis simulation. A microanastomosis motion detector was developed using a machine learning model capable of tracking 21 hand landmarks without physical sensors attached to a surgeon's hands. Anastomosis procedures were simulated using synthetic vessels, and hand motion was recorded with a microscope and external camera. Time series analysis was performed to quantify the economy, amplitude, and flow of motion using data science algorithms. Six operators with various levels of technical expertise (2 experts, 2 intermediates, and 2 novices) were compared. The detector recorded a mean (SD) of 27.6 (1.8) measurements per landmark per second with a 10% mean loss of tracking for both hands. During 600 seconds of simulation, the 4 nonexperts performed 26 bites in total, with a combined excess of motion of 14.3 (15.5) seconds per bite, whereas the 2 experts performed 33 bites (18 and 15 bites) with a mean (SD) combined excess of motion of 2.8 (2.3) seconds per bite for the dominant hand. In 180 seconds, the experts performed 13 bites, with mean (SD) latencies of 22.2 (4.4) and 23.4 (10.1) seconds, whereas the 2 intermediate operators performed a total of 9 bites with mean (SD) latencies of 31.5 (7.1) and 34.4 (22.1) seconds per bite. A hand motion detector based on machine learning technology allows the identification of gross and fine movements performed during microanastomosis. Economy, amplitude, and flow of motion were measured using time series data analysis. Technical expertise could be inferred from such quantitative performance analysis.

The Road of Post-Industrialization Transformation in Developing Countries Based on Weighted Markov and Grey Correlation Theory, Taking the Change of Industrial Structure in Heilongjiang Province of China as a Case Study

In the context of economic globalization, the comparative advantages of developing countries and developed countries tend to be more complex. After the economic crisis in 2008, the challenges of developing countries are intensified, and the game is more cyclical; Especially in the critical period when the “Fourth Industrial Revolution” is pending, many countries have increased their efforts in policy formulation and efficiency improvement, focusing on industrial transformation and upgrading, and closely combining structural reform with the industrialization process. Therefore, it is particularly important to analyze, change and forecast the industrial structure of post-industrial regions in developing countries based on data science algorithms, and to reshape the understanding of the adjustment of the world economic order and the evolution of the international trade division system. Based on the proportion of gross regional product and output value of three industries in Heilongjiang Province of China from 2001 to 2020, the minimum deviation model of industrial structure in Heilongjiang Province is constructed through Markov theory. The Lingo software is used to obtain the transition probability matrix of the industrial structure state, and combined with the autocorrelation coefficient of each order and the transfer weight, the change trend of the proportion of the output value of the three industries and the change of the contribution rate of the three industries (to GDP) in Heilongjiang Province in the next 10 years are obtained; At the same time, through the grey correlation index and spss software, this paper analyzes the correlation changes between the three industrial adjustments and economic development in Heilongjiang Province in the past 20 years, discusses the new growth points of economic development in Heilongjiang Province and puts forward corresponding suggestions for the adjustment of different industrial structures in Heilongjiang Province, and finally extends the general rules of the development of post-industrialization in the world. This article believes that it is necessary to adjust the structure of the first, second, and third industries reasonably based on different historical and natural endowments, contemporary backgrounds, and other practical factors, in accordance with local conditions, circumstances, and each with its own emphasis. At the same time, it also requires the support and inclination of government policies; Adapting industrial structure to local economic development while actively leading productivity and local economic development.

Cattle Disease Prediction Using Artificial Intelligence

Abstract: In today’s world identifying cattle disease and providing proper treatments is a challenging task in the current medical sector. As it is difficult to identify thecattle disease in real time, we require a method topredict cattle disease and related patterns. There are so many research works on this topic. Most of the researchworks just presented the idea of cattle disease prediction. There are many works where implementation is done and many papers predicts cattle disease using efficient data science algorithms. Research works where implementation is done uses PYTHON language or R language as programming language for cattle disease prediction. As PYTHON language and R language supports all ready libraries to process training datasets and to predict cattle disease. Many papers use training datasets from www.kaggle.com, www.dataworld.com etc.Research works uses efficient algorithms for prediction, algorithms such as Naive Bayes algorithm, KNNclassifier, SVM classifier, Decision Tree classifier, Random Forest algorithm etc. Most of the papers got very good results of using these algorithms. So many works on this cattle disease and pattern prediction is done using data science techniques.

Algorithms of statistical anomalies clearing for data science applications

The paper considers the nature of input data used by Data Science algorithms of modern-day application domains. It then proposes three algorithms designed to remove statistical anomalies from datasets as a part of the Data Science pipeline. The main advantages of given algorithms are their relative simplicity and a small number of configurable parameters. Parameters are determined by machine learning with respect to the properties of input data. These algorithms are flexible and have no strict dependency on the nature and origin of data. The efficiency of the proposed approaches is verified with a modeling experiment conducted using algorithms implemented in Python. The results are illustrated with plots built using raw and processed datasets. The algorithms application is analyzed, and results are compared.

How can artificial intelligence and data science algorithms predict life expectancy - An empirical investigation spanning 193 countries

Improving life expectancy, which is falling and declining globally, is getting harder, especially given the limited economic resources in most parts of the world. This study seeks to identify the most significant factors that positively and negatively affect life expectancy, separating them from less significant factors. This way, specific areas can be identified to channel scarce economic resources towards increasing life expectancy. Using data science techniques on a dataset comprising economic, immunological, health, personal, and social attributes, we have been able to achieve this. Furthermore, four machine learning tree regression algorithms were employed using the identified attributes to develop and evaluate life expectancy prediction models. The extremely randomized tree model performed best using evaluation matrices: MAE, RMSE, R2, and CV score. This research can help governments, especially in low-income, developing countries, make better decisions and investments, as well as help digital health experts develop technologies that could improve life expectancy.

Development, validation and psychometric evaluation of the Chinese version of the biopsychosocial impact scale in orofacial pain patients.

The objective of this study was to develop the Chinese version of the biopsychosocial impact scale (BPIm-S) to assess functional limitation and psychosocial distress in orofacial pain (OFP) patients in mainland China, and investigate the factor structure, reliability and validity, measurement invariance, as well as scores differences across genders, age and educational status among OFP patients. The BPIm-S was developed and evaluated in four stages: (1) concept selection and item generation; (2) a pilot study assessing face and content validity; (3) the factors structure, reliability, convergent validity, and measurement invariance; and (4) concurrent validity and clinical responsiveness. Exploratory (EFA) and confirmatory factor analyses (CFA) were performed on data gathered from 406 OFP patients to assess construct validity. Composite Reliability (CR) and the Average Variance Extracted (AVE) were used to assess internal convergent validity. CR, internal consistency, and split-half reliability were also performed to determine the reliability. Multigroup CFA (MGCFA) was used to assess measurement invariance across genders, age and educational status. Mann-Whitney test compared scores across different genders, age and educational status. Participants completed the BPIm-S, visual analog scale (VAS), brief pain inventory facial (BPI-F), General Anxiety Disorder-7 (GAD-7) and Patient Health Questionnaire-9 (PHQ-9), and spearman's correlation coefficient was used to evaluate the concurrent validity and item-total correlations. A total of 12 patients with OFP completed the BPIm-S twice to test clinical responsiveness. To conduct the CFA and measurement invariance analysis, Mplus 8.4 was used. IBM SPSS Statistics 21 software and SPSSAU, a web-based data science algorithm platform tool, were used for all additional studies. For the preliminary version, 17 items were chosen. A total of four items were removed following the pilot research. The remaining 13 items of the BPIm-S comprised an overall summary scale. Excellent reliability (Item-to-total correlations ranged from 0.763 to 0.912) and strong internal consistency (Cronbach's α = 0.970, functional limitation, 0.962, and psychosocial distress, 0.977) were discovered. CFA also validated the structural validity of the 13-item scale. EFA was performed and a two-factor structure was investigated. In addition, MGCFA corroborated the measurement invariance of the BPIm-S across gender, age, and educational status. Patients over the age of 30, those with a medium level of education, and those with a low level of education showed substantially greater levels of functional limitation and psychological distress (Wilcoxon test, p < 0.001). Both concurrent validity and clinical responsiveness were assessed to be of good quality. The BPIm-S demonstrated good psychometric qualities and is a reliable tool that can now be used by clinicians to evaluate functional limitation and psychosocial distress among OFP patient.

A Modeling Approach for Measuring the Performance of a Human-AI Collaborative Process

Despite the unabated growth of algorithmic decision-making in organizations, there is a growing consensus that numerous situations will continue to require humans in the loop. However, the blending of a formal machine and bounded human rationality also amplifies the risk of what is known as local rationality. Therefore, it is crucial, especially in a data-abundant environment that characterizes algorithmic decision-making, to devise means to assess performance holistically. In this paper, we propose a simulation-based model to address the current lack of research on quantifying algorithmic interventions in a broader organizational context. Our approach allows the combining of causal modeling and data science algorithms to represent decision settings involving a mix of machine and human rationality to measure performance. As a testbed, we consider the case of a fictitious company trying to improve its forecasting process with the help of a machine learning approach. The example demonstrates that a myopic assessment obscures problems that only a broader framing reveals. It highlights the value of a systems view since the effects of the interplay between human and algorithmic decisions can be largely unintuitive. Such a simulation-based approach can be an effective tool in efforts to delineate roles for humans and algorithms in hybrid contexts.

Data science knowledge integration: Affordances of a computational cognitive apprenticeship on student conceptual understanding

AbstractThis study implements a computational cognitive apprenticeship framework for knowledge integration of Data Science (DS) concepts delivered via computational notebooks. This study also explores students' conceptual understanding of the unsupervised Machine Learning algorithm of K‐means after being exposed to this method. The learning of DS methods and techniques has become paramount for the new generations of undergraduate engineering students. However, little is known about effective strategies to support student learning of DS and machine learning (ML) algorithms. The research questions are: How do students conceptualize their understanding of an unsupervised ML method after engaging with interactive visualizations designed using the computational cognitive apprenticeship approach? How do the affordances of the interactive visualizations support or hinder student knowledge integration of an unsupervised machine learning method? Design‐based research allowed for the iterative design, implementation, and validation of the pedagogy in the context of a working classroom. For this, data collection methods often take the form of student artifacts. We performed a qualitative content analysis of students' written responses and reflections elicited during the learning process. Results suggest that the computational cognitive apprenticeship promoted knowledge integration. After interacting with the computational notebooks, most students had accurate conceptions of the goal and the nature of the method and identified factors affecting the output of the algorithm. Students found it useful to have a concrete representation of the method, which supported its conceptual understanding and showcased the acquisition of strategic knowledge for its appropriate execution. However, we also identified important misconceptions students held about the algorithm.

Data Science: The Way How to Use Data

Data Science, a new discovery illustration, is potentially one of the most significant advances of the early 21st century. Derived from scientific discovery, it is being applied to every human effort for which there is sufficient data. Significant and remarkable successes have been achieved; even greater claims have been made. Along with benefits, challenge and risks are abound. The science underlying data science has yet to emerge. This claim is based on observing the centuries-long developments of its predecessor paradigms - empirical, theoretical, and Jim Gray's Fourth Paradigm of Scientific Discovery (Hey, Tansley&amp; Tolle, 2009) (aka eScience, data-intensive, computational, procedural). This paper is mainly focuses on essential questions for data science- what is data science, importance of data science in our everyday life, data wrangling, data science algorithms.

Traditional African Dances Preservation Using Deep Learning Techniques

Human action recognition continues to evolve and improve through deep learning techniques. There have been studies with some success in the field of action recognition, but only a few of them have focused on traditional dance. This is because dance actions, especially in traditional African dance, are long and involve fast movements. This research proposes a novel framework that applies data science algorithms to the field of cultural preservation by applying various deep learning techniques to identify, classify, and model traditional African dances from videos. Traditional dances are an important part of African culture and heritage. Digital preservation of these dances in their multitude and form is a challenging problem. The dance dataset was constituted from freely available YouTube videos. Four traditional African dances were used for the dance classification process: Adowa, Swange, Bata, and Sinte dance. Five Convolutional Neural Network (CNN) models were used for the classification and achieved an accuracy between 93% and 98%. Additionally, human pose estimation algorithms were applied to Sinte dance. A model of Sinte dance that can be exported to other environments was obtained.

Data Science Toolkit: An all-in-one python library to help researchers and practitioners in implementing data science-related algorithms with less effort

<h2>Abstract</h2> Data Science Toolkit (DST) is a python library built as a wrapper layer on top of several libraries to increase the abstraction level of the code, making its users more efficient and productive. The current version is widely used in our ongoing research activities that focus on optimizing agricultural management practices using artificial intelligence. DST adopts an object-oriented approach in implementing data science algorithms and is therefore composed of multiple classes such as the DataFrame class that adds additional functionalities to the standard pandas dataframe and the Model class that facilitates the building, training, and evaluation of machine learning models.

Development of a data-driven scientific methodology: From articles to chemometric data products

Information and data science algorithms were combined to predict the outcome of an experiment in chemical engineering. Using the Scientific Method workflow, we started the journey with the formulation of a specific question. At the research stage, the common process of querying and reading articles on scientific databases was substituted by a systematic review with a built-in recursive data mining method. This procedure identifies a specific community of knowledge with the key concepts and experiments that are necessary to address the formulated question. A small subset of relevant articles from a very specific topic among thousands of papers was identified while assuring the loss of the least amount of information through the process. The secondary dataset was bigger than a common individual study. The process revealed the main ideas currently under study and identified optimal synthesis conditions to produce a chemical substance.Once the research step was finished, the experimental information was compiled and prepared for meta-analysis using a supervised learning algorithm. This is a hypothesis generation stage whereby the secondary dataset was transformed into experimental knowledge about a particular chemical reaction. Finally, the predicted sets of optimal conditions to produce the desired chemical compound were validated in the laboratory.

Limits to the Productivity in Biobased Territorial SMEs

In Latin América there are large exporting companies intensive in natural resources which do not connect with the SMEs in the territories. The aim of this study is to provide elements that explain the limits to the productivity of small and medium enterprises located in bio-based territories, providing inputs for a public policy that enhances local development. The case of sawmills located in the Province of Arauco, Bío-Bío Region, Chile, is studied. The results of a survey applied to 42 companies are analyzed, in the context of an initiative financed by the Regional Government of Biobío, which represent 84% of the total sawmills in the Province of Arauco. Different data science algorithms are used to generate clusters and decision trees. Based on the results, a heterogeneity within the group of companies is verified from the perception of the main problems, the processes that are internalized, the number of sectors it sells to and the cost of raw materials. Public policies recommendations are presented aimed at building support networks, improving quality standards through certifications, attracting, and retaining qualified human capital and promoting collaborative design among companies.