2024 International Conference on Communication, Computer Sciences and Engineering (IC3SE)

Employee Attrition Analysis Using XGBoost

Isha Nitin Thapliyal Sheetal Solanki
Department of Computer Science Department of Computer Science Department of Computer Science
Graphic Era Hill University Graphic Era Hill University Graphic Era Hill University
Dehradun, India. Dehradun, India. Dehradun, India.
[email protected] [email protected] [email protected]

Neeraj Kumar Pandey Sangita Papola

Department of Computer Science and Department of Computer Science
Engineering Graphic Era Hill University
Graphic Era Deemed to be University Dehradun, India.
2024 International Conference on Communication, Computer Sciences and Engineering (IC3SE) | 979-8-3503-6684-6/24/$31.00 ©2024 IEEE | DOI: 10.1109/IC3SE62002.2024.10593326

Dehradun, India [email protected]

[email protected]

Abstract—In today’s time, employee attrition is a major Continual staff turnover can create uncertainties leading to
problem faced by organisations. The most precious resource less employees and financial drawbacks for the
for a firm is its workforce. They are the ones that bring values organization. Attrition not only affects the inside business
both quantitatively and qualitatively. A steady decline in the but also its image and reputation in the eyes of the outside
workforce due to resignation, death or retirement is referred
world. Usually, employee attrition over a period of time
to as attrition. It is a matter of concern when employees leave
their jobs of better opportunities, but this leads the company follows the above mentioned as basics and calculate a
to face its consequences. Employee turnover and attrition are numeric value in percentage referred as attrition rate.
equivalent. Leaders can avoid it by implementing better
retention techniques by mastering this concept. In this paper, To compute the attrition rate dividing the number of
we use dataset to analyse attrition and determine the root employees who quit their jobs by the total number of
cause of employee resignation. We did analysis of the training employees that were recruited at the beginning when the
dataset for effective data exploration. Machine Learning period started. For example, let us count employees to
algorithms, such as Random Forest Classifier, Logistic know how many of them were there at the initial year of a
Regression, XGBoost and ensemble stacking technique were
company, suppose 1,000. Keep a check on how many of the
used. On data exploration and enabling algorithm we
discovered that XGBoost classifier with an accuracy of 88. % people left the company through- out the year. Suppose
and precision 89% was the most appropriate algorithm for there were 250 employees who left the company
this dataset. We may also conclude that monthly income, age, intentionally or unintentionally. We also view at how many
distance from home can be a reason behind employees leaving new employees were hired within the year and conduct a
their job and finding better opportunities. final count for employees at the end of the year. For
example, let suppose a company hired 500 new employees
Index Terms—Attrition, Random Forest, Logistic Regression, in that particular year. This concludes that the final
XGBoost, Exploration. headcount of that company would be 1,500 for that year.
Now let’s look at the average number of employees for that
I. INTRODUCTION particular year. It comes out as (1000+1500)/2 = 1,250. As
we have the average number of employees with this we
Employee attrition is the term coined for the process of calculate the percentage of employees who left. The
gradual decline in size of organization’s workforce formula for it is Average number of employees/number of
overtime due to inescapable causes such as employee employees who left*100. Using the above formula attrition
resignation for personal or professional reasons or death. rate can be calculated as: (250/1250) *100. The percentage
The rate of employees leaving the organization is relatively comes out to be 20%.
higher than the rate of recruitment and such situations when
occur are beyond the reach of the employer. An employee It is very essential for an organization to minimize its
would choose to join or resign from a company considering employee attrition to minimum to standout in the
many factors. Some take into consideration variables like competition market. Therefore, it is important for
salary, department, working environment, distance from organisations to take necessary steps and opt proper
home, gender equity, etc while other takes into account strategies to improve their finances and productivity.
personal reasons like maternity, family, health issues. Google’s 2018 Annual Diversity Report dedicates a
Employee attrition is the unexpected or unpredictable separate section precisely discussing about employee
process either voluntarily or involuntarily. The basic attrition. It specified that black and Latin employee’s
purpose of this analysis is to get understanding of and attrition was way more than the attrition rate in company
handling employee attrition. Employee attrition is one of average by a wide margin. The bulk of workforce are white
the critical aspects for any company because it has an and Asian. Black employees contribute to just 2.5% of
enormous impact on the company’s financial and market Google’s U.S. employee while Asian people are more with
performance. It becomes a setback for companies when a very slight margin of at 3.6%. This clearly shows why
trained, expertise and important employees quit their jobs measuring and keeping a check on employee attrition of an
for better opportunities of growth in career. Higher attrition organization is important and necessary. This led Google to
rate acts as a setback for the company and negatively focus on their employees such that to provide justifiable
impacts its performance, market value and reputation.

1770
979-83503-6684-6/24/$31.00 ©2024 IEEE
Authorized licensed use limited to: Tri Sagirani. Downloaded on January 22,2025 at 03:24:45 UTC from IEEE Xplore. Restrictions apply.
2024 International Conference on Communication, Computer Sciences and Engineering (IC3SE)

future. Uber is another example which highlights the use of Extreme Gradient Boosting (XGBoost), a more
importance of keeping a check on employee attrition. While reliable approach. The primary objective of the research is
Uber has continuously denied having an above average to demonstrate that XGBoost forecasts employee attrition
attrition rate but there are not enough evidence to prove it with a high degree of accuracy when compared to different
suggest otherwise. When Uber filed for an IPO in 2019, its models. Zangeneh et al. [2] presents a methodology for
public disclosure documents highlighted a number of predicting attrition and uses real dataset from IBM HR as a
problems of workspace culture. It included a decrease in case study. As there are many features in the databases, the”
employee incentives, poor employer’s image which can be max out” feature selection strategy is used during the pre-
a major reason for employees who reportedly left the processing step. The results show that this technique
company in the past few years. From the previously improves the F1-score performance. Yang et al. [3]
mentioned two case studies, it is evident that Google and analysed IBM Employee Attrition dataset to determine the
Uber possess different perspective on attrition. On the one factors affecting employee’s decision to quit their
hand Uber views attrition as an inevitable consequence of respective jobs. Using a correlation matrix, firstly
its policies at workplace, Google on the other hand opts for unimportant or un- wanted attributes were eliminated from
a more honest and candid approach, Google openly admits the dataset and selected the most important features. Then,
that employee attrition is one issue that needs consideration a Random Forest approach was used to split the population
and addressing. In their 2019 Retention Report, Work into two clusters utilizing monthly income, age and the
Institute discovered that preventable attrition surpasses count of companies worked for. This was done after
unpredictable attrition in frequency. 6 out of 100 employees dividing the population through K-means clustering and
left their jobs due to retirement, a change of career or other conducting quantitative analysis using binary logistic
such reason which can also be expressed as 22%. The report regression, it was determined that the proportion of
anticipated that the attrition rate would be so high by the employees who frequently travelled was 2.4 times greater
end of this year that 35% of employees would leave their than those who travelled infrequently. Additionally, there
present jobs to work in a different company. For the first was observed to be a higher propensity for employees in the
time in 2018, it was noticed that the number of vacancies human resources department to depart. Setiawan et al. [4]
crossed the number of unemployed workers. This helped aimed to use logistic regression to evaluate employee
companies drew a conclusion that job satisfaction was not attrition. R was applied for data preparation, integration,
enough to retain the employees. Work-life development exploratory analysis, logistic regression, model evaluation,
came out to be the number one reason for employees and visualization. Five datasets containing identical, unique
leaving jobs, followed by work- life balance and manager employee IDs were created from the data. To understand
behaviour. Also, attrition cannot always be considered the dataset, the functions summary () and str () were
negatively. It can also turn out the other way round for utilized. The desired value of a variable was transformed
businesses where poor performing, or weak employees from yes/no to levels 0/1 for the purposes of model
leave the company, and this empty space can be filled with selection and training. The dataset divided into seventy-
new and promising talent. It is noticed that in tech thirty percentage in which (30%) for the test set and (70%)
companies usually males have a higher number of the training set. The prediction model’s sensitivity,
employments which can cause inequality. Here, attrition evaluation accuracy, and test accuracy were 75%, 73%, and
may assist in the termination of workers who aren’t a right 75%, respectively. It was shown that an employee’s
fit for the organization. It helps to establish an engaged probability of leaving a company rises noticeably with the
workforce in the organisation as the same employees number of years and companies they had worked for. By
having same ideologies won’t be operating the company for accessing the job satisfaction, working environment, and
a longer time. Employee attrition is confidential data of the workload with communications between supervisors and
company and can’t be disclosed to the world, so a dataset staff, the business can enhance its HR division and reduce
is randomly selected from Kaggle, and it is used to analyse staff churn. Nesree et al. [5] analyses records of employees
the data and predict the attrition rate of a company. We used who worked in reputed institutions in Nigeria and left it
some methodologies of machine learning by stacking them between 1978 and 2006, were used as a dataset. Mostly,
for classification of data and prediction of result. job-related and demographic data were used to categorize
workers into established attrition classifications. See5 for
II. LITERATURE SURVEY Windows and the Waikato Environment for Knowledge
Analysis (WEKA) were used to create decision tree
Several machine learning research on employee attrition models. The predictive model [18][19] was then built using
have been conducted. Taking ideas from the completed the results of the decision tree model and the produced
work we apply combinations of some of these models and rulesets. With this technique, predictions about probable
techniques on our selected dataset and analysis it to get cases of employee attrition were made. A solitary decision
proper results. Below are some of the works: tree of 15 by 3 sub-trees was produced, with a
misclassification rate of 25.2 per cent. Salary was used 100
Punnoose et al. [1] suggested Global retailer’s HRIS percent of the time, duration of service was used 49 percent
database, BLS (Bureau of Labour Statistics) data and of the time, and rank was used 16 of the time. These
trained and tested supervised algorithms like Random findings suggest that employee salary and length of service
Forest, Logistic Regression, XGBoost, KNN, SVM Naïve were the main drivers of whether or not they chose to
Bayes on ROC-AUC metric on the database. Suggested the continue with the organization.

1771

Authorized licensed use limited to: Tri Sagirani. Downloaded on January 22,2025 at 03:24:45 UTC from IEEE Xplore. Restrictions apply.
2024 International Conference on Communication, Computer Sciences and Engineering (IC3SE)

Kamath et al. [6] primary objective of the research was to

examine the reasons behind the unreasonable departure of
skilled and experienced employees and also forecast which
employee would resign. The dataset was selected from
Kaggle. The study was done using R and Rattle data mining
platform. The dataset was divided into training, testing and
validation. Decision trees, random forest, support Vector
Machine and linear regression techniques were used to
build the model. The model’s performance was evaluated
through error ma- trix and pseudo R square estimation of
error rate. It was predicted that Random Forest was the
most suitable model for employee attrition analysis in
comparison to other models. “Job satisfaction” was the
dominant factor that influenced the attrition rate the most.
Raza et al. [7] examined the organizational components that Figure 2. Heatmap
influence employee attrition and used machine learning to From fig.6 monthly income and job level are highly
foresee attrition. Additionally, Employee Exploratory Data correlated, also job level and total working year.
Analysis (EEDA) was used to determine the contributing
causes of employee attrition. Support Vector Machines Recursive feature eliminator (RFE) is used for data pre-
(SVM), Logistic Regression (LR), Decision Tree Classifier processing and feature selection. Feature selection can be
(DTC), and Extra Tree Classifier (ETC) are the four defined as a technique that chooses a subset of features that
machine learning approaches compared in this work for are most relevant for the problem statement from the
analysis. The proposed Extra Trees Classifier (ETC) dataset. By using feature selection, unnecessary and
approach achieved a 93% accuracy score in employee irrelevant features are excluded, and machine learning
attrition pre- diction. It was concluded that age, work level, algorithms run more efficiently as they have less space and
hourly rate, and monthly income were the primary factors time complexity. If this technique is avoided, our prediction
contributing to employee attrition. may result in wrong or worse results, and we may redo it.

III. PROPOSED METHODOLOGY For feature selection algorithm [8], RFE is a wrapper-type
algorithm known to be popular, easy to use and effective in
selecting relevant features from the dataset. The process of
data exploration and visualization plays a significant role in
data science. Data analysts use exploration process to see
or identify specific patterns in the dataset. It helps the
information to be represented easily and interactive visual
elements making the data being understood better.

This step helps users to understand their data statistically

and visually through graphical means. Graphs and charts
are used for making complex structures simpler and finding
Figure 1. Architecture of the proposed approach relations within data as the ability to see different visual
patterns and colours helps to interpret data more accurately.
The dataset is a set of data represented in the form of table Breaking down the data into simpler structures will be
to make it easily visible and understandable. In dataset each helpful for professional sectors. We used various graphs for
column describes a particular feature, and each row gives plotting. Having attrition as a common feature, we plotted
information about a member corresponding to that feature it with the number of employees, department, age of the
this way it constitutes to data management. Being specific employee, job satisfaction etc.
about this project, dataset was selected from Kaggle.

The dataset comprises 1470 records and 35 fields, including

definite and numeric features.

The heatmap of the dataset, which represents the

coefficients, gives a visualization of the strength of
correlation among variables.

Figure 3. random forest feature importance

1772

Authorized licensed use limited to: Tri Sagirani. Downloaded on January 22,2025 at 03:24:45 UTC from IEEE Xplore. Restrictions apply.
2024 International Conference on Communication, Computer Sciences and Engineering (IC3SE)

From fig.3, the top three features are monthly income, age
and distance from home are dominant, and they indicate
whether the employee has a tendency to leave. While job
role, marital status indicate employees are less likely to go.
Monthly income is the primary reason that employees
choose to leave their firm. People getting low salaries will
depart the firm more likely and the employees getting high
salary package will stay in the organization. No one wanted
a low salary package particularly the fresher or young
employees who just started their careers. In search of a
higher salary package, they find other firms that offer more
money than earlier. Additionally, recent graduates are more
inclined to explore various job opportunities and eventually
discover a fitting career path as they transition from
universities.
Figure 4. Random Forest.
Distance from home the other main reason is the distance
of company(office) and home as commuting for longer time For binary classification tasks in machine learning, logistic
leads to wastage of time and energy which can be regression is a basic and popular approach. The likelihood
physically and mentally draining further can lead to job of a binary result dependent on one or more predictor
dissatisfaction. Due to the adverse effects of long variables is modeled using this linear model, despite its
commutes on mental health, physical activity, and stress name. Due to its simplicity, interpret-ability, and efficiency
levels, workers are prioritizing their health and seeking for in modelling binary outcomes, logistic regression is widely
opportunities closer to home. Overall, the distance between utilized in a variety of disciplines, including marketing (for
home and the workplace can significantly influence the customer churn prediction), finance (for credit scoring),
determination of an employee to depart from a company, as and healthcare (for illness risk prediction). The linear
it directly impacts their quality of life, job satisfaction, and relationship between the log-odds of the outcome and the
overall happiness. predictor variables, which logistic regression depends on,
might not always stay true in real-world circumstances.
Age adds to attrition because of many factors such as health
considerations, retirement, career switches and the need to Extreme Gradient Boosting, or XGBoost for short, is a
explore new domains and opportunities. sophisticated and very effective use of the gradient boosting
technique. It is widely used in machine learning
We divide the dataset into training and testing dataset competitions and real-world applications due to its
consisting of 1176 and 294(80-20%) records respectively exceptional performance and scalability. Feature
with the same 24 features. engineering, structured data, and tabular data are just a few
of the machine learning applications that choose XGBoost
IV. MODEL BUILDING due to its state-of-the-art performance, scalability, and
adaptability. By providing insights into feature relevance,
The report offers recommendations and insights to assist XGBoost enables users to choose which features are most
firms in lowering their rate of employee turnover. Certain pertinent for usage in prediction.
machine learning classifiers were put to use to identify
significant features and conduct data analysis. Classifier is
an algorithm that automatically distinguish or classifies
data ac- cording to classes. Types of classifiers that are used
in machine learning are random forest classifier, logistic
regression, naive Bayes, decision tree, support vector
machine (SVM) and many more. Here, we have used
Random Forest classifier, Logistic Regression, XG-boost
and ensemble stacking technique to find which suits best on
our dataset. Machine learning tasks including both
regression and classification, Random Forest is a flexible
and effective ensemble learning technique. It constructs
multiple decision trees during training, uses a random
subset of features at each node and a variable part of the Figure 5. Logistic Regression.
data to train each tree. For classification, where the mode
of predictions is chosen, or for regression tasks, where the
mean of predictions is derived, an averaging process
determines the final prediction in a Random Forest. By
combining predictions from several trees, the ensemble
method reduces over-fitting and boosts adaptability.

1773

Authorized licensed use limited to: Tri Sagirani. Downloaded on January 22,2025 at 03:24:45 UTC from IEEE Xplore. Restrictions apply.
2024 International Conference on Communication, Computer Sciences and Engineering (IC3SE)

employees to seek better opportunities elsewhere. Apart

from this using different model we found comparisons that
males tend to depart more as compared to female, people
who travel frequently leaves more because travelling leads
to disruption of work life balance, fatigue and stress and
those who work overtime are more likely to leave.
Additionally, the Sales Department witnessed the highest
turnover rate among employees.

Figure 6. XGBoost. VI. FUTURE SCOPE

Results can be improved and made more accurate by Future studies [22] could enhance the analysis by taking
combining several methods of machine learning. This is the into account factors that are positively correlated with
concept Ensemble learning. By using multiple models on worker intentions to leave, such as inequality, insufficient
behalf of single model, the performance and prediction both peer support, unfavourable working conditions [23], and
are improved. Its advantage is that it gives majority vote, limited career chances. This paper would help in analysis
Bagging Randomness Injection, Feature selection, Error- of employee attrition of an organization [8] and letting it
correcting output coding. In ensemble classification we know the reason for the same.
used stacking technique. Stacking consists of model
training for combination of predictions of multiple different VII. CONCLUSION
algorithms. Models of Logistic Regression are usually used
as combiners. Fig 5 presents a combiner algorithm makes The results obtained by the model indicate that our findings
the final prediction after combining all the other algorithms align with observations from the real world [10] and
as additional input or using cross-validated prediction. This previous studies conducted by other academics. When
gives better predictive results. Both supervised and attrition occurs within a company the workload among
unsupervised learning have been successfully existing members of the team increases despite a pay raise.
accomplished with ensemble learning. Regression, The workload of the existing team members increases when
classification, and distance learning are examples of attrition occurs within a company without any pay raise.
supervised learning; density estimation is an example of The HR specialist even deals with this workload. Due to the
unsupervised learning. It has also proved to be useful for role being eliminated due to attrition, there may not be a
error rate estimation of bagging. chance for promotion in the workplace. We can conclude
V. RESULTS that it is an issue which is to be taken under consideration.
Job satisfaction, daily rate, employee number and age were
After data exploration and applying model algorithms, we some of the factors that led to attrition. It was also observed
equated them to know the best-suited algorithm with the that sales department had the highest attrition rate.
highest accuracy for our dataset. The confusion matrix of Knowing these aspects will enable the organization to take
the algorithms came out to be as follows. the required actions to lower attrition rate and prevent a
significant number of employees from leaving.
This list consists of random forest classifier, XG-boost,
Logistic Regression, and ensemble classifiers. Data REFERENCES
processing is done using RFE (recursive feature
[1] Rohit Punnoose and Pankaj Ajit.”Prediction of employee
elimination). We made some graphs between various Turnover in organization using Machine Learning Algorithms”.In:
features like age, job satisfaction with attrition rate to get a International Journal of Advanced Research in Artificial
rough idea about feature selection. After data exploration, Intelligence (2016).
we use models to find most suitable algorithm with highest [2] Mangal, V., Dhamija, S. Analysing Theoretical Models for
Predicting Employee Attrition: A Comparative Study in the
accuracy and precision for our dataset. FMCG Sector.
[3] Yang, Shenghuan, and Md Tariqul Islam. ” IBM employee attrition
analysis.” arXiv preprint arXiv:2012.01286 (2020).
[4] Setiawan, I. A., et al.” HR analytics: Employee attrition analysis
using logistic regression.” IOP Conference Series: Materials
Science and Engineering. Vol. 830. No. 3. IOP Publishing, 2020.
[5] El-Rayes, Nesreen, et al. “Predicting employee Attrition using Tree-
based Models.” International Journal of Orgaizational Analysis
(2020).
[6] Kamath, Dr RS, Dr SS Jamsandekar, and Dr PG Naik.” Machine
Figure 7. Confusion Matrix of XG-Boost learningapproach for employee attrition analysis.” Int. J. Trend Sci.
Res. Dev., vol. Special Is, no. Special Issue-FIIIIPM2019 (2019):
The analysis revealed that Xg-boost proved as the most 62-67.
suitable algorithm for predicting employee attrition, [7] Raza, Ali, et al.” Predicting employee attrition using machine
learning approaches.” Applied Sciences 12.13 (2022): 6424.
boasting an accuracy of 88% and a precision of 89%. The employee-attrition- and-factors.
graphical representation of feature importance indicated [8] Vyas, S., Bhargava, D. and Khan, S., 2023. Healthcare 4.0: A
that Monthly Income exerted the greatest influence, systematic review and its impact over conventional healthcare
followed by distance from home and age, prompting system. Artificial Intelligence for Health 4.0: Challenges and
Applications, pp.1-17.

1774

Authorized licensed use limited to: Tri Sagirani. Downloaded on January 22,2025 at 03:24:45 UTC from IEEE Xplore. Restrictions apply.
2024 International Conference on Communication, Computer Sciences and Engineering (IC3SE)

[9] Reference for recursive feature elimination

https://machinelearningmastery.com/rfe-feature-selection-in-
python/
[10] Thapliyal, N., Dimri, P. Task scheduling using fuzzy logic with
best-fit-decreasing for cloud computing environment. Cluster
Comput (2024). https://doi.org/10.1007/s10586-024-04378-7
[11] Qutub, Aseel, et al.” Prediction of employee attrition using machine
learning and ensemble methods.” Int. J. Mach. Learn. Comput
11.2(2021): 110-114.
[12] Jain, Rachna, and Anand Nayyar.” Predicting employee attrition
using xgboost machine learning approach.” 2018 international
conference on system modeling advancement in research trends
(smart). IEEE, 2018.
[13] Link for dataset
https://www.kaggle.com/datasets/thedevastator/employee- attrition-
and-factors
[14] Najafi-Zangeneh, Saeed, et al.” An improved machine learning-
based employees attrition prediction framework with emphasis on
feature selection.” Mathematics 9.11 (2021): 1226.
[15] James, M.J. and Faisal, U., 2013. Empirical study on addressing
high employee attrition in BPO industry focusing on employee
salary and other factors in Karnataka and Kerala states of India.
Research Journalof Management Sciences.
[16] Pandey, N. and Kaur, G., 2011. Factors influencing employee
attritionin Indian ITeS call centres. International Journal of Indian
Culture and Business Management, 4(4), pp.419-435.
[17] Alshiddy, M.S. and Aljaber, B.N., 2023. Employee Attrition
Prediction using Nested Ensemble Learning Techniques.
International Journal of Advanced Computer Science and
Applications, 14(7).
[18] Al-Alawi, A.I. and Ghanem, Y.A., 2024, January. Predicting
Employee Attrition Using Machine Learning: A Systematic
Literature Review. In 2024 ASU International Conference in
Emerging Technologies for Sustainability and Intelligent Systems
(ICETSIS) (pp. 526-530). IEEE.
[19] Gim, Simon, and Eun Tack Im. ”A Study on Predicting Employee
Attri-tion Using Machine Learning.” In IEEE/ACIS International
Conference on Big Data, Cloud Computing, and Data Science
Engineering, pp. 55-69. Cham: Springer International Publishing,
2022.
[20] Agarwal, V, S Taware, S A Yadav, D Gangodkar, A L N Rao, and V
K Srivastav, ‘Customer - Churn Prediction Using Machine
Learning’, in Proceedings of International Conference on
Technological Advancements in Computational Sciences, ICTACS
2022, 2022, pp. 893–99.
[21] Gangadharan, S. & Dharani, M. & Thapliyal, Nitin & Yamsani,
Nagendar & Singh, Jagendra & Singh, Prabhishek. (2023).
Comparative Analysis of Deep Learning-Based Brain Tumor
Prediction Models Using MRI Scan. 1-6.
10.1109/CISCT57197.2023.10351227. D. M. El Din, "Online
paper review analysis," Int. J. Adv. Comput. Sci. Appl., vol. 6, no.
9, 2015.
[22] N. Thapliyal, M. Aeri, A. Kumar, V. Kukreja and R. Sharma,
"Combining Spatial and Temporal Analysis: A CNN-LSTM Hybrid
Model for Maize Disease Classification," 2024 IEEE International
Conference on Computing, Power and Communication
Technologies (IC2PCT), Greater Noida, India, 2024, pp. 1529-
1533, doi: 10.1109/IC2PCT60090.2024.10486456.
[23] Gupta, N, P K Juneja, S Sharma, and U Garg, ‘Future Aspect of 5G-
IoT Architecture in Smart Healthcare System’, in Proceedings - 5th
International Conference on Intelligent Computing and Control
Systems, ICICCS 2021, 2021, pp. 406–11.
[24] Sheth, Kalgi, Jaynil Patel, and Jaiprakash Verma.” Machine
Learning- Based Investigation of Employee Attrition Prediction
and Analysis.” In Emerging Technology Trends in Electronics,
Communication and Networking: Select Proceedings of the Fourth
International Conference, ET2ECN 2021, pp. 221-238. Singapore:
Springer Nature Singapore,2022.

1775

Authorized licensed use limited to: Tri Sagirani. Downloaded on January 22,2025 at 03:24:45 UTC from IEEE Xplore. Restrictions apply.