Employee turnover has been identified as a key issue for organizations because of its adverse impact on work place productivity and long term growth strategies. To solve this problem, organizations use machine learning techniques to predict employee turnover. Accurate predictions enable organizations to take action for retention or succession planning of employees. However, the data for this modeling problem comes from HR Information Systems (HRIS); these are typically under-funded compared to the Information Systems of other domains in the organization which are directly related to its priorities. This leads to the prevalence of noise in the data that renders predictive models prone to over-fitting and hence inaccurate. This is the key challenge that is the focus of this paper, and one that has not been addressed historically. The novel contribution of this paper is to explore the application of Extreme Gradient Boosting (XGBoost) technique which is more robust because of its regularization formulation. Data from the HRIS of a global retailer is used to compare XGBoost against six historically used supervised classifiers and demonstrate its significantly higher accuracy for predicting employee turnover.

/pdf/prediction-of-employee-turnover-in-organizations-using-28ol5ljfmh.pdf

Prediction of Employee Turnover in Organizations using Machine Learning Algorithms

Supervised machine learning methods are described, demonstrated and assessed for the prediction of employee turnover within an organization. In this study, numerical experiments for real and simulated human resources datasets representing organizations of small-, medium- and large-sized employee populations are performed using (1) a decision tree method; (2) a random forest method; (3) a gradient boosting trees method; (4) an extreme gradient boosting method; (5) a logistic regression method; (6) support vector machines; (7) neural networks; (8) linear discriminant analysis; (9) a Naive Bayes method; and (10) a K-nearest neighbor method. Through a robust and comprehensive evaluation process, the performance of each of these supervised machine learning methods for predicting employee turnover is analyzed and established using statistical methods. Additionally, reliable guidelines are provided on the selection, use and interpretation of these methods for the analysis of human resources datasets of varying size and complexity.

Employee Turnover Prediction with Machine Learning: A Reliable Approach.

There are several areas in which organisations can adopt technologies that will support decision-making: artificial intelligence is one of the most innovative technologies that is widely used to assist organisations in business strategies, organisational aspects and people management. In recent years, attention has increasingly been paid to human resources (HR), since worker quality and skills represent a growth factor and a real competitive advantage for companies. After having been introduced to sales and marketing departments, artificial intelligence is also starting to guide employee-related decisions within HR management. The purpose is to support decisions that are based not on subjective aspects but on objective data analysis. The goal of this work is to analyse how objective factors influence employee attrition, in order to identify the main causes that contribute to a worker’s decision to leave a company, and to be able to predict whether a particular employee will leave the company. After the training, the obtained model for the prediction of employees’ attrition is tested on a real dataset provided by IBM analytics, which includes 35 features and about 1500 samples. Results are expressed in terms of classical metrics and the algorithm that produced the best results for the available dataset is the Gaussian Naive Bayes classifier. It reveals the best recall rate (0.54), since it measures the ability of a classifier to find all the positive instances and achieves an overall false negative rate equal to 4.5% of the total observations.

Predicting Employee Attrition Using Machine Learning Techniques

The growing interest in machine learning among business leaders and decision makers demands that researchers explore its use within business organisations. One of the major issues facing business leaders within companies is the loss of talented employees. This research studies employee attrition using machine learning models. Using a synthetic data created by IBM Watson, three main experiments were conducted to predict employee attrition. The first experiment involved training the original class-imbalanced dataset with the following machine learning models: support victor machine (SVM) with several kernel functions, random forest and K-nearest neighbour (KNN). The second experiment focused on using adaptive synthetic (ADASYN) approach to overcome class imbalance, then retraining on the new dataset using the abovementioned machine learning models. The third experiment involved using manual undersampling of the data to balance between classes. As a result, training an ADASYN-balanced dataset with KNN (K = 3) achieved the highest performance, with 0.93 F1-score. Finally, by using feature selection and random forest, F1-score of 0.909 was achieved using 12 features out of a total of 29 features.

Predicting Employee Attrition using Machine Learning

Employees are the valuable assets of any organization. But if they quit jobs unexpectedly, it may incur huge cost to any organization. Because new hiring will consume not only money and time but also the freshly hired employees take time to make the respective organization profitable. Hence in this paper we try to build a model which will predict employee churn rate based on HR analytics dataset obtained from Kaggle website. To show the relation between attributes, the correlation matrix and heatmap is generated. In the experimental part, the histogram is generated, which shows the contrast between left employees vs. salary, department, satisfaction level, etc. For prediction purpose, we use five different machine learning algorithms such as linear support vector machine, C 5.0 Decision Tree classifier, Random Forest, k-nearest neighbor and Naive Bayes classifier. This paper proposes the reasons which optimize the employee attrition in any organization.

Evaluation of machine learning models for employee churn prediction

Employee turnover is a serious concern in knowledge based organizations. When employees leave an organization, theycarry with them invaluable tacit knowledge which is often the source of competitive advantage for the business. In order foran organization to continually have a higher competitive advantage over its competition, it should make it a duty to minimizeemployee attrition. This study identifies employee related attributes that contribute to the prediction of employees’ attritionin organizations. Three hundred and nine (309) complete records of employees of one of the Higher Institutions in Nigeriawho worked in and left the institution between 1978 and 2006 were used for the study. The demographic and job relatedrecords of the employee were the main data which were used to classify the employee into some predefined attrition classes.Waikato Environment for Knowledge Analysis (WEKA) and See5 for Windows were used to generate decision tree modelsand rule-sets. The results of the decision tree models and rule-sets generated were then used for developing a a predictivemodel that was used to predict new cases of employee attrition. A framework for a software tool that can implement therules generated in this study was also proposed.Keywords: Employee Attrition, Decision Tree Analysis, Data Mining

/pdf/analyzing-employee-attrition-using-decision-tree-algorithms-1niewjof1x.pdf

Analyzing employee attrition using decision tree algorithms

There have been major improvements in the trauma system and injury prevention in Al-Ain City. We aimed to study the impact of these changes on the incidence, pattern, injury severity, and outcome of hospitalized motorcycle-related injured patients in Al-Ain City, United Arab Emirates.This is a retrospective analysis of two separate periods of prospectively collected data which were retrieved from Al-Ain Hospital Trauma Registry (March 2003 to March 2006 compared with January 2014 to December 2017). All motorcycle-injured patients who were admitted to Al-Ain Hospital for more than 24 h or died in the Emergency Department or after hospitalization were studied.The incidence of motorcycle injuries dropped by 37.1% over the studied period. The location of injury was significantly different between the two periods (p = 0.02, Fisher's exact test), with fewer injuries occurring at streets/highways in the second period (69.1% compared with 85.3%). The anatomical injury severity of the head significantly increased over time (p = 0.03), while GCS on arrival significantly improved (p < 0.0001), indicating improvements in both prehospital and in-hospital trauma care. The mortality of the patients significantly decreased (0% compared with 6%, p = 0.002, Fisher's exact test).The incidence of motorcycle injuries in our city dropped by almost 40% over the last 15 years. There was a significant reduction in the mortality of hospitalized motorcycle-injured patients despite increased anatomical severity of the head injuries. This is attributed to improvements in the trauma care system, including injury prevention, and both prehospital and in-hospital trauma care. 

/pdf/reduction-of-motorcycle-related-deaths-over-15-years-in-a-qx6su3af.pdf

Reduction of motorcycle-related deaths over 15 years in a developing country

In-Hospital Cardiac Arrest (IHCA) is a significant burden on healthcare worldwide. Outcomes of IHCA are worse in developing countries compared with developed ones. We aimed to study the epidemiology and factors determining outcomes in adult IHCA in a high income developing country.We abstracted prospectively collected data of adult patients admitted to our institution over a three-year period who suffered a cardiac arrest. We analysed patient demographics, arrest characteristics, including response time, initial rhythm and code duration. Pre-arrest vital signs, primary diagnoses, discharge and functional status, were obtained from the patients' electronic medical records.A total of 447 patients were studied. The IHCA rate was 8.6/1000 hospital admissions. Forty percent (40%) achieved ROSC with an overall survival to discharge rate of 10.8%, of which 59% had a good functional outcome, with a cerebral performance category score of 1 or 2. Fifty-four percent (54%) of patients had IHCA attributed to causes other than cardiac or respiratory. Admission Glasgow Coma Scale (GCS), shockable rhythm and short code duration were significantly associated with survival (p < 0.001).A combination of patient and system-related factors, such as the underlying cause of cardiac arrest and a lack of DNAR policy, may explain the reduced survival rate in our setting compared with developed countries. 

https://doi.org/10.1016/j.resplu.2022.100220

The epidemiology and outcomes of adult in-hospital cardiac arrest in a high-income developing country

Correspondence to Dr Kehinde Olawale Ogunyemi; ogunyemikehinde89@ gmail. com © Author(s) (or their employer(s)) 2022. Reuse permitted under CC BYNC. No commercial reuse. See rights and permissions. Published by BMJ. Forecasting is an important aspect of decisionmaking in health and other social aspects of human life. Forecasting can simply be defined as the process of making probabilities about a realworld event using existing data built in a mathematical model. This understanding underpins why forecasting is sometimes used interchangeably with the word ‘modelling’. Forecasting capacity describes a system that comprises of surveillance database, experts and relevant technologies, and it remains an indispensable workforce development need for promoting datadriven decisionmaking in health, as frequently advocated by the WHO. Generally, the usefulness and usability of forecasting is one that is not unknown or unbeneficial to most people across the world, particularly in nonemergency situations, from its use in daily weather reporting through global projections on economy, and diseases burden. Similarly, from epidemiological perspective, evidence from various forecasting models was observed to have played a major role to improve emergency response in past disease outbreaks (eg, Ebola) and towards the COVID19 pandemic in the areas not limited to SARSCoV2 patterns determination, containment and mitigation measures implementation, risk communication, resource management and vaccine development. Despite the demonstrated availability of forecasting capacity and its associated benefits on health protection at the global level, unfortunately, the ownership and usage of forecasting knowledge capita in public health emergency management remain significantly limited at the regional level as laid bare by the COVID19 pandemic, with Africa being the most disproportionately impacted region, despite having a record high figure of over 100 public health emergencies annually compared with other regions of the world. This disparity in forecasting capacity is reflected in current forecasting evidence on COVID19 pandemic, where most of the studies were either conducted in the developed regions or for Africa by foreign experts such as in a study by Frost and colleagues. In addition, while anecdotal evidence shows that forecasting capacity exists in some settings in Africa such as academia, governments, nongovernmental organisations (NGOs), this capacity is largely underresourced, uncoordinated and shortterm probably due to weak surveillance systems and lack of national emergency forecasting centres or forecasting units within the existing national public health institutes. Other reasons for the illdeveloped forecasting capacity in Africa could as well be attributed to the lack of political will and weak partnerships between the government and the academia, where most of the forecasting SUMMARY BOX

/pdf/developing-forecasting-capacity-for-public-health-emergency-1u9azafy.pdf

Developing forecasting capacity for public health emergency management in Africa using syndemics approach: lessons from the COVID-19 pandemic

Abstract Aim To study the epidemiology and pattern of trauma-related deaths of hospitalized patients in Abu Dhabi Emirate, United Arab Emirates, in order to improve trauma management and injury prevention. Methods The Abu Dhabi Trauma Registry prospectively collects data of all hospitalized trauma patients from seven major trauma centres in Abu Dhabi Emirate. We studied all patients who died on arrival or after admission to these hospitals from January 2014 to December 2019. Results There were 453 deaths constituting 13.5% of all trauma deaths in the Abu Dhabi Emirate. The median (IQR) age of the patients was 33 (25–45) years, and 82% were males. 85% of the deaths occurred in the emergency department (ED) and the intensive care unit (ICU). Motor vehicle collision (63.8%) was the leading cause of death. 45.5% of the patients had head injury. Two of the seven hospitals admitted around 50% of all patients but accounted for only 25.8% of the total deaths ( p &lt; 0.001). Those who died in the ward (7%) were significantly older, median (IQR) age: of 65.5 (31.75–82.25) years, ( p &lt; 0.001), 34.4% of them were females ( p = 0.09). The median (IQR) GCS of those who died in the ward was 15 (5.75–15) compared with 3 (3–3) for those who died in ED and ICU ( P &lt; 0.001). Conclusions Death from trauma predominantly affects young males with motor traffic collision as the leading cause. Over 85% of in-hospital deaths occur in the ICU and ED, mainly from head injuries. Injury prevention of traffic collisions through enforcement of law and improved hospital care in the ED and ICU will reduce trauma death. 

D Alao

Papers

Analyzing employee attrition using decision tree algorithms

Reduction of motorcycle-related deaths over 15 years in a developing country

The epidemiology and outcomes of adult in-hospital cardiac arrest in a high-income developing country

Developing forecasting capacity for public health emergency management in Africa using syndemics approach: lessons from the COVID-19 pandemic

Trauma deaths of hospitalized patients in Abu Dhabi Emirate: a retrospective descriptive study