




































BANGLADESH JOURNAL OF MULTIDISCIPLINARY SCIENTIFIC RESEARCH 10(4) (2025), 1-10 

1 

        MULTIDISCIPLINARY SCIENTIFIC RESEARCH 
          BJMSR VOL 10 NO 4 (2025) P-ISSN 2687-850X E-ISSN 2687-8518 

         Available online at https://www.cribfb.com 

     Journal homepage: https://www.cribfb.com/journal/index.php/BJMSR 

                                                                                                                                                                                                    Published by CRIBFB, USA 
                                                                                                                                     

INVESTIGATING THE IMPACT OF AI ON THE WORKFORCE 

AND THE FUTURE OF WORK IN THE REGION: A MACHINE 

LEARNING PERSPECTIVE                                                                    
 

 Boumedyen Shannaq (a)1     Ahmed Alabri (b)   V. P. Sriram (c)   Oualid Ben Ali (d)   
 

(a) Associate Professor, Management of Information Systems Department, University of Buraimi, Al Buraimi Governorate, Sultanate of Oman; E-

mail: boumedyen@uob.edu.om 
(b) Finance and Administrative Affairs & Supporting Services, University of Buraimi, Al Buraimi 512, Sultanate of Oman; E-mail: 

drahmed.s@uob.edu.om 
(c) Associate Professor, Management of Information Systems Department, University of Buraimi, Al Buraimi Governorate, Sultanate of Oman; E-

mail: sriram.v@uob.edu.om 
(d) Associate Professor and Acting Head, Computer Science Department, Applied Science University, Kingdom of Bahrain; E-mail: 

oualid.ali@asu.edu.bh 
        

 
A R T I C L E I N F O 

 
 

Article History: 

 
Received: 19th February 2025 

Reviewed & Revised: 19th February 

to 10th June 2025 

Accepted: 12th June 2025 

Published: 20th June 2025 

 
Keywords: 

 

AI, Future Workforce, Clustering methods, 

Workforce Opportunities and Risks 

 
JEL Classification Codes: 

 

      O33, J24, J21, C55 

       

 

      Peer-Review Model:  
 

      External peer review was done through  

      double-blind method.        

 
A B S T R A C T      

 

This study examines the evolving impact of Artificial Intelligence (AI) on workforce dynamics within a 

regional context. The analysis of structured data, which includes job positions alongside industry 
categories, collected from Kaggle.com, AI implementation metrics, automation uncertainty assessments, 

skill requirements, compensation amounts, and work projection estimates, enables the application of 

machine learning approaches that generate insights for decision support. The purpose of this study is 

twofold: to analyze the impact of AI on work structures and identify automated jobs, as well as vulnerable 

sectors, to inform recommendations that help plan education systems and workforce development. Three 

clustering approaches, including K-Means and DBSCAN, along with Agglomerative Clustering, were 

implemented to categorize different jobs based on their AI acceptance levels, automation probability, 

and pay ranges. The performance analysis, as indicated by silhouette scores, revealed that 
Agglomerative Clustering generated meaningful clusters at a score of 0.289, while both K-Means and 

DBSCAN recorded scores of 0.262 and 0.093, respectively. The developed clusters enable researchers 

to identify vulnerable positions while proposing new career options and uncovering stable competencies, 

which include digital aptitude as well as emotional capability and troubleshooting abilities. The study 

directly provides answers to major research questions about how AI affects particular sectors while 

revealing portable skills across industries. Through the integration of cluster analytics and workforce 

analytics, this study provides policymakers, educational institutions, and workforce planners with 
strategic information, enabling a resilient labor market that is prepared for the future. 

 
 

© 2025 by the authors. Licensee CRIBFB, USA. This open-access article is distributed under the 
terms and conditions of the Creative Commons Attribution (CC BY) license 
(http://creativecommons.org/licenses/by/4.0).  

            

 

INTRODUCTION 

Artificial Intelligence (AI) is rapidly disrupting global labor conditions. At the same time, it reformulates workplace 

structures, reshapes necessary competencies, and develops innovative approaches to work (Öztaş & Arda, 2025; Lamees & 

Ramayah, 2025; Kaur et al., 2024). This research examines the evolving relationship between artificial intelligence 

technology and labor forces across the Arab Gulf nations, which are actively leveraging digital transformation to reduce 

their dependence on oil resources (Hazaa & Al Mubarak, 2024). The research employs machine learning techniques to 

conduct an extensive evaluation of the effects of AI adoption on job structures, as well as professional requirements and 

workplace risk factors. AI, together with automation technologies, has brought about significant changes throughout the 

global workforce organizations (Tenakwah & Watson, 2025). Artificial technology systems enable the automated 

performance of tasks that were previously restricted to human skills and cognition (Sigafoos et al., 2025). The changes in 

AI position various professions in different industries for elimination or thorough modification (Wang & Lu, 2025). By 

2030, over 40% of jobs are expected to be displaced by automation, artificial intelligence, virtual reality, and augmented 

reality. Less than 50% of students feel ready for future workforce demands. This study informs policymakers, educators, 

and strategists(Pandya et al., 2022). The effects of these findings become significant for Arab and Gulf economic systems 

                                                      
1Corresponding author: ORCID ID: 0000-0001-5867-3986 
© 2025 by the authors. Hosting by CRIBFB. Peer review is the responsibility of CRIBFB, USA.  

https://doi.org/10.46281/a7zcns96 

 
To cite this article: Shannaq, B., Alabri, A., Sriram, V. P. & Ali, O. B. (2025). INVESTIGATING THE IMPACT OF AI ON THE WORKFORCE AND 

THE FUTURE OF WORK IN THE REGION: A MACHINE LEARNING PERSPECTIVE. Bangladesh Journal of Multidisciplinary Scientific Research, 

10(4), 1-10. https://doi.org/10.46281/a7zcns96 

http://creativecommons.org/licenses/by/4.0/)
http://creativecommons.org/licenses/by/4.0/)
https://www.openaccess.nl/en
https://doi.org/10.46281/a7zcns96
https://orcid.org/0000-0001-5867-3986
https://orcid.org/0009-0009-9743-211X
https://orcid.org/0000-0002-7105-0713
https://orcid.org/0009-0001-1433-017X


Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

2 

(Naim et al., 2025). The youth-oriented population growth combined with government leadership in digital transformation 

and national AI investments form the key features of these regions according to Oman's Vision 2040 (Awashreh, 2025), 

Saudi Arabia's Vision 2030 (Suleiman & Ming, 2025) and the United Arab Emirates' National AI Strategy (Hassouni & 

Mellor, 2025). The positive outlook faces resistance from the workforce deficit between current job abilities and future AI 

economy requirements (Nadimpalli et al., 2025). AI adoption at work requires urgent analysis of its time-dependent effects 

on employment markets alongside strategy development to create flexible and resilient job forces (Nawaz & Li, 2025)  

The field of global literature presents multiple perspectives on the impact of AI on work. However, it highlights a 

lack of studies focused on the Arab labor market supported by actual data. Most contemporary employment guidelines and 

academic syllabi do not adequately address the rapid changes enabled by AI technology. The lack of empirical evidence on 

this subject has led to the development of fragmented strategies, which fail to provide practical solutions for mitigating the 

increasing risks of automation and evolving skill requirements. Policymakers, along with workforce planners, struggle to 

make informed strategic decisions for workforce development because they lack sufficient information. 

The proposed research examines how AI adoption varies across both industrial sectors and occupational titles in 

the region, with a particular focus on high-tech industries, service sectors, educational institutions, and manufacturing 

facilities to determine differential penetration levels among various economic groups. By assigning automation-likelihood 

scores and conducting field examinations, job categories are stratified into distinct levels of AI automation risk, allowing us 

to identify those abilities—such as complex problem-solving, emotional intelligence, and digital fluency—that remain 

resilient despite technological advancements. Moreover, the study explores machine learning–based methods for 

recommending reskilling or upskilling pathways to workers, thereby informing policy decisions on workforce development. 

Finally, by applying clustering algorithms (including K-Means, DBSCAN, and Hierarchical Clustering), the research 

categorizes the workforce into risk profiles. It proposes feasible career transition options for individuals most vulnerable to 

automation. 

The study employs machine-learning techniques to analyze structured data, integrating information on industry 

sectors, job categories, AI implementation levels, automation likelihood predictions, and competency requirements to 

achieve several key objectives. First, it aims to examine how the adoption of AI reshapes the regional employment structure. 

Second, it aims to identify which job profiles and industries are more vulnerable to automation. Third, it intends to generate 

data-driven recommendations that can inform educational strategies and workforce planning initiatives. Ultimately, by 

identifying transferable skill sets that remain valuable across various sectors, the research fosters greater resilience in the 

labor market. 

The paper follows this structure: Section 2 investigates the literature on the impact of AI on the workforce, and 

Section 3 describes the research methodology. Section 4 presents the experiment and results, followed by Section 5, which 

discusses the findings and examines future study directions. 

 

LITERATURE REVIEW 

Artificial intelligence, machine learning, and digitalization are transforming workplaces, particularly in knowledge-

intensive sectors. Scholars debate these disruptions, yet their impact on institutions, organizations, and individual’s remains 

underexplored (Hamdan, 2025). A relevant investigation examines AI’s role in the Future of Work, discussing key changes, 

consequences, and research directions while summarizing contributions from this special issue in management studies 

(Sarala et al., 2025). Worldwide labor markets are experiencing dramatic changes due to Artificial Intelligence technology, 

which is leading industries through transformation while altering the identities of all job functions (Rawashdeh, 2025). 

Modern economies face new opportunities alongside complex challenges due to the rise of AI-based automation, predictive 

analysis, and the adoption of machine learning systems (Bahoo et al., 2025). The initial section of this review examines the 

worldwide impact of AI on employment practices. Still, it concentrates the subsequent sections on the Middle East Gulf 

regions through a strategic evaluation of their principal advancements, alongside workplace obstacles and forthcoming 

market prospects. 

Several academic studies have documented the disruptive effects that AI technology has on work environments. 

The research conducted by Frey and Osborne (2017) indicates that machine automation threatens 47 percent of all positions 

across the United States (The Future of Work: Embracing AI's Job Creation Potential, 2025). Forecasts that AI and 

automation will eliminate 85 million positions during the next five years but will establish 97 million brand-new 

employment roles. 

AI-powered automation technologies have created the most significant disruption in standard routine operations 

within industrial manufacturing sectors, as well as in retail operations and finance services (Tariq, 2025). Knowledge-based 

sectors, including healthcare, software development, and education, currently require additional experts who possess 

knowledge of AI applications (Kumar, 2025). Business operations benefit from predictive analytics, which encompasses 

workforce management and personalized customer service, leading to higher productivity (Leveraging Predictive Analytics 

to Optimize Business Performance and Drive Operational Excellence, 2025). 

AI raises ongoing concerns about security challenges in employment, alongside widening income disparities and 

ongoing requirements for skill adaptation across the workforce (Burton, 2024). Various economies are funding nationwide 

upskilling initiatives to protect workers from automated processes and foster effective working relationships between 

humans and AI systems. 

The region of the Middle East currently experiences rapid digital change that results in the swift growth of AI 

within its various sectors; AI will generate $320 billion in economic value across the region during the following decade, 

as predicted by The Potential Impact of Artificial Intelligence in the Middle East - PwC Middle East (2025). The United 

Arab Emirates, together with Saudi Arabia, leads AI integration projects by focusing on AI-driven approaches for economic 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

3 

diversification. 

Healthcare benefits from AI diagnostics technology and telemedicine services that increase access to healthcare 

within the industry (Unanah & Mbanugo, 2025). Intellectual institutions integrate AI-based education systems into their 

traditional curricula (Abbasi et al., 2025). The finance sector achieves operational efficiencies and detects fraudulent 

activities through the implementation of AI technology (Dayalan & Sundaramurthy, 2024). 

Staff in the oil and gas industries benefit from AI technologies that utilize predictive maintenance methods 

alongside energy efficiency models (Rojas et al., 2025). 

Despite the clear advantages that artificial intelligence offers to the Middle Eastern market, several obstacles 

impede effective AI adoption. First, there is a pronounced shortage of workers with the requisite AI competencies, which 

undermines organizational efforts to deploy and maintain advanced systems (R. & Archana, 2025). Moreover, efforts to 

establish standardized AI governance frameworks have been hindered by the absence of cohesive regulatory structures, 

resulting in delayed implementation and uncertainty regarding compliance (Lund et al., 2025). As a consequence, unskilled 

employees face a heightened risk of displacement since automation increasingly targets routine tasks without sufficient 

retraining pathways in place (Broady et al., 2025). Compounding these issues is a general lack of public awareness regarding 

how AI is reshaping occupational roles and workforce dynamics, which further slows the transition to a digitally skilled 

labor market (Gayathri & Mangaiyarkarasi, 2024; Karam et al., 2025). 

The Gulf Cooperation Council (GCC) member countries—Saudi Arabia, the UAE, Qatar, Bahrain, Oman, and 

Kuwait—are proactively integrating AI into their economies, exemplified by Saudi Arabia’s Vision 2030 and the UAE’s 

National AI Strategy 2031, both of which position AI as a key growth engine (Azoury & Karam, 2025). Recent studies 

highlight several promising developments: government investments are catalyzing an increase in AI research centers and 

funding bodies; AI-driven thoughtful city planning is enhancing infrastructure management, improving passenger mobility, 

safeguarding citizens, and promoting environmental sustainability; predictive analytics in the oil and gas sector is 

streamlining resource management; and AI applications in hospitality are enabling personalized services for tourists (Visvizi 

et al., 2025; Kshetri & Sharma, 2025). However, these opportunities are tempered by workforce challenges, including a 

heavy dependence on expatriate talent that limits the growth of local AI expertise, a need for universities to bolster AI 

curricula and programs, and the imperative for labor regulations to evolve in response to AI-induced workplace 

transformations (Abdalla, 2025; Akinwale et al., 2025). 

Artificial Intelligence is transforming worldwide and regional employment markets, introducing new challenges 

but also substantial opportunities. AI adoption rates are increasing throughout Middle Eastern and Gulf countries. However, 

obstacles to workforce readiness can be addressed through specialized policies that combine educational changes with 

collaboration from the business sector. An AI-driven strategic transformation of the workforce, combined with practical 

implementation methods, will create job market sustainability over the next few decades. Table 1 presents a comparative 

analysis of global trends, Middle East Trends, and Gulf Region Trends, considering AI Progress, Challenges, and 

Opportunities. 

 

Table 1. Comparative Analysis 

 
Aspect Global Trends Middle East Trends Gulf Region Trends 

AI Progress Advanced adoption in developed 

economies; AI-driven job creation and 

loss are balanced. 

AI adoption is accelerating across key 

sectors, including finance, healthcare, 

and education. 

Government-led AI strategies, heavy 

investment in smart cities, oil, and 

tourism. 

Challenges Skill gaps, ethical concerns, job 

displacement, and regulatory issues. 

Lack of AI talent, limited AI regulatory 

frameworks, and workforce 

displacement. 

Reliance on expatriates, slow educational 

integration of AI, and outdated labor 

policies. 

Opportunities AI in automation, data science, and 

emerging tech industries. 

AI-powered fintech, energy, and 

innovative education initiatives. 

Vision 2030 and AI-led economic 

diversification programs. 

 

MATERIALS AND METHODS 

A detailed information collection effort consolidated data regarding professional positions, business fields, AI 

implementation statistics, occupational specifications, and robotic process automation probability levels. The clustering 

models comprised K-Means and DBSCAN, along with the Agglomerative Hierarchical Clustering method, to extract data 

clusters and identify susceptible positions with suitable reskilling transitions. Organizational recommendations directed to 

stakeholders depend on findings derived from this research. 

The research analysis utilized structured data points, including industry sectors, job positions, AI deployment 

scales, automation risk measurements, and necessary competencies for handling tasks. The research employed three machine 

learning (ML) clustering approaches — K-Means, DBSCAN, and Hierarchical Clustering — to categorize job roles into 

distinct risk categories. The analysis utilized risk profiles to match emerging skill requirements, enabling the production of 

recommendations regarding career transitions. 

 

 

 

 

 

 

 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

4 

Administrative support personnel, along with basic accounting staff and data entry workers, face the highest risk 

of displacement (Tharmalingam & Pereira, 2025). The proposed research methodology in this work is presented in Figure 

1. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1. The proposed Research Methodology 

 

Data Collection 

The study relied on data from Kaggle.com, a platform known for its extensive collection of datasets available for use in 

academic and professional studies. For this purpose, the "AI Adoption & Automation Risk (San Francisco, CA)" dataset 

was utilized to investigate how AI adoption, automation risk, and labor market trends interact with one another. 

The dataset offers a realistic portrayal of jobs in San Francisco, examining how AI is impacting various job sectors. 

Among other things, it offers job titles, industrial sectors, AI involvement and results, risk of automation, needed skills, 

salary details, and forecasts for growth in positions. The dataset was constructed to closely resemble practical situations. 

Because it gives both categories and numbers for key variables, statistical analysis using structural equation modeling 

becomes possible. It is particularly suitable for research examining the effects of AI on the job market, the skills required 

for jobs, and career preparation in countries that heavily rely on AI. The data is highly accurate; the findings from this data 

need to be verified against other datasets related to the region of study to confirm their applicability to practical decision-

making and education. A sample of the dataset is presented in Table 2. 

 

Table 2. Sample of the Dataset 

 
Job_Title Industry AI 

Adoption 

Level 

AI 

Adoption 

Score 

Automation 

Risk 

Automation 

Risk Score 

Required_ 

Skills 

Salary 

(USD) 

Job Growth 

Projection 

Job 

Growth 

Score 

Cybersecurity 

Analyst 

Retail High 3 High 3 Cybersecurity 75862.86 Growth 3 

AI Researcher Energy Low 1 High 3 Machine 

Learning 

71211.88 Growth 3 

AI Researcher Retail Medium 2 High 3 Data Analysis 71374.65 Stable 2 

AI Researcher Manufacturing Medium 2 High 3 Marketing 99743.29 Decline 1 

AI Researcher Manufacturing Low 1 High 3 JavaScript 104107.3 Decline 1 

UX Designer Manufacturing High 3 Medium 2 Project 

Management 

101649 Growth 3 

AI Researcher Transportation High 3 Medium 2 Python 73151.99 Growth 3 

Software 

Engineer 

Finance High 3 Medium 2 UX/UI Design 56076.4 Growth 3 

 

Table 3 describes the Equivalencies of each Job Growth Score. 

 

Table 3. Equivalencies 

 
AI Adoption Level Score 

High 3 

Medium 2 

Low 1 

Automation Risk Score 

High 3 

Medium 2 

Low 1 

Job Growth Projection Score 

Applied Machine 

Learning Models 

K-Means and 

DBSCAN and 

Hierarchical 

Clustering 

Data Collection 

Source: 

Kaggle.com 

Data Preprocessing 

Categorical and ordinal 

encoding, Normalization 

Results and Analysis 

 

Visualization (2D plot) showing 

job groupings, Cluster 

interpretations 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

5 

Growth 3 

Stable 2 

Decline 1 

 

Job roles that require human-centered design, along with critical thinking, creativity, and complex problem-solving 

skills, demonstrate strong resilience. The skills centered on AI literacy, combined with data analytics and digital 

communication, remain highly resistant to industry changes in the future.  

 

Data Preprocessing 

Prior to applying clustering algorithms such as K-Means, DBSCAN, and Agglomerative Hierarchical Clustering, the data 

underwent a series of preprocessing steps to ensure consistency and reliability. First, any records containing missing or null 

values were either removed or, when feasible, imputed using appropriate techniques: numerical features with missing entries 

were replaced by their mean values, while categorical features were filled using the mode. Next, categorical variables—

such as industry, AI adoption level, and job growth projection—were transformed using label encoding or one-hot encoding 

as dictated by each Algorithm's requirements. To prevent features with larger numeric ranges from dominating distance 

calculations, continuous variables (for example, salary, AI adoption scores, and automation risk scores) were scaled using 

either Min-Max Normalization or a Standard Scaler to restrict their values to a [0, 1] range. In cases where the high 

dimensionality of the dataset threatened clustering performance, Principal Component Analysis (PCA) was applied to reduce 

the feature space while retaining essential variance. Finally, because DBSCAN is sensitive to extreme values—treating them 

as potential outliers—any numerical features exhibiting extreme dispersion were adjusted beforehand to limit their influence 

on centroid separation and distance-based point assignments. 

Because of these steps, the data was tidied up, adjusted, and could be used for clustering without any supervision. 

Table 4 demonstrates the proposed clustering models and equations. 

 

Table 4. Clustering models and equations 

 
Clustering models  Equations Symbols Reference 

K-Means Clustering  

 

 

 

 k: Number of clusters 

 Ci: Set of data points 

in cluster i 

 x: A data point 

 μi: Centroid of cluster 
iii 

 ∥x−μi∥2: Squared 

Euclidean distance 

between a point and 
its cluster centroid 

(Jin & Han, 2011) 

DBSCAN(Density-Based 

Spatial Clustering of 

Applications with Noise) 

 

 
 

 

 D: dataset 

 dist(p,q): typically 

Euclidean distance 
between points p and 

q 

 ε: radius for 

neighborhood 
 

(Sander et al., 1998) 

Agglomerative Hierarchical 

Clustering 

 

 

 

 D(A, B): Distance 
between clusters A 

and B 

 ∣A∣,∣B∣: Number of 
points in clusters 

AAA and BBB 

 μA,μB: Centroids of 
clusters A and B 

 

(Zepeda-Mendoza 

& Resendis-
Antonio, 2013) 

 

Machine Learning Algorithm and Experiments 

This work proposed an Algorithm for Job Clustering Based on AI Adoption, Automation Risk, and Salary. The proposed 

Algorithm utilizes unsupervised AI learning methods to cluster job positions by analyzing the AI Adoption Score, along 

with the Automation Risk Score and Salary data. Then, it evaluates the outcomes achieved using K-Means, DBSCAN, and 

Agglomerative Clustering methods. The proposed Algorithm consists of 10 steps and is presented in Figure 2. The 

Algorithm was implemented using Python code and ran in a Google Colab environment. 

The proposed Algorithm employs a multi-stage clustering approach to analyze job data centered on AI and 

automation risk. It is essential to gather data first, then select the most critical features and standardize their measurements. 

All three methods—K-Means, DBSCAN, and Agglomerative Hierarchical Clustering—are used together to organize similar 

jobs based on the features picked. Every clustering model utilizes the Silhouette Score to evaluate the effectiveness of the 

created groups. DBSCAN is given parameters eps=1.0 and min_samples= =3, which allows it to find noise and clusters with 

flexible shapes. To facilitate visualization, the data is reduced from 3D to 2D using Principal Component Analysis (PCA). 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

6 

Each model's clusters are colored to make them easier to understand. The linkage method by Ward gives you a dendrogram 

to display how clusters are formed. In the final part, models and outputs are compared to highlight tasks that are most likely 

to be automated, providing helpful suggestions for reskilling individuals. It combines statistical accuracy with valuable 

insights into the workforce. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2. The proposed Algorithm 

 

RESULTS 

The results show a comparison of the three algorithms — K-Means, DBSCAN, and Agglomerative Hierarchical Clustering 

— using their Silhouette Scores. Based on their respective Silhouette Scores, they help objectively determine how well 

clusters are divided and grouped by each Algorithm. As presented in Table 5, Agglomerative Clustering scored highest on 

the Silhouette Score (0.289), with K-Means closely behind (0.262), suggesting that the clusters formed by these methods 

are well-defined. DBSCAN performed the worst (0.093), meaning it detected a greater spread or possible noise within the 

dataset. The analysis reveals that hierarchical clustering is effective in identifying the primary trends in the dataset regarding 

AI use and the potential for each job to be replaced by machines. Based on the findings, an appropriate clustering approach 

is chosen for further planning and policy recommendations, which aids in identifying unsafe jobs and developing strategies 

for retraining workers to adapt to AI-based work. 

 

Table 5. Silhouette Score Comparison of Clustering Algorithms 

 
Clustering Method Silhouette Score 

K-Means 0.262 

DBSCAN 0.093 

Agglomerative 0.289 

 

 

 

 

 

 

Step 6: Apply Agglomerative Clustering 

Step 2: Feature Selection Step 1: Data Collection 

Step 3: Data Normalization 

Start 

Step 4: Apply K-Means Clustering Step 5: Apply DBSCAN Clustering 

  Initialize K-Means with n_clusters=3. 

  Fit the model to the normalized data. 

  Assign cluster labels to each job. 

  Compute the Silhouette Score for evaluating cluster 

quality. 

 

 Initialize DBSCAN with eps=1.0 and 
min_samples=3. 

 Fit the model and assign labels to each job. 

 If DBSCAN finds more than one cluster, compute 

the Silhouette Score; otherwise, assign -1 
(indicating poor clustering). 

 

  Apply Agglomerative Hierarchical Clustering with 
n_clusters=3. 

  Fit the model to the normalized data. 

  Assign cluster labels. 

  Compute the Silhouette Score for the results. 

 

Step 7: Dimensionality Reduction  Visualization 

  Use Principal Component Analysis (PCA) to reduce 

the dataset from 3D to 2D. 

  Plot scatter graphs for each clustering method using 

PCA-transformed data. 

  Use different colors to represent different clusters for: 

 K-Means 

 DBSCAN 

 Agglomerative Clustering 
 

Step 9: Performance Comparison 

Step 10: Output Results 

End 

Step 8: Generate Dendrogram 

  Apply Ward's linkage method to generate a 

dendrogram. 

  The dendrogram visualizes the hierarchical structure of 

clusters formed by Agglomerative Clustering. 

 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

7 

Figure 3 demonstrates the comparison of clustering methods, and Figure 4 presents the Agglomerative dendrogram. 

 

 
 

Figure 3. Comparison of Clustering Methods 

 

 
Figure 4. Agglomerative dendrogram 

 

Explanation of Clustering Results and Silhouette Score Comparison 

The Silhouette Score is used to evaluate the effectiveness of the clustering results by measuring the similarity between each 

data point and its cluster compared to points in other clusters. Scores range from –1 to 1, where values near 1 indicate that 

clusters are well separated and internally cohesive, reflecting excellent delineation among groups. Scores around zero 

suggest substantial overlap between clusters, implying that the clustering structure may not be optimal. Negative values 

indicate that some samples are likely misassigned, as they are closer to points in another cluster than to those within their 

group. 

 

The three clustering methods yield the following Silhouette Scores for evaluation: 

 

K-Means Clustering (Silhouette Score = 0.262118) 

K-Means is a centroid-based clustering algorithm that partitions data into K clusters by minimizing the variance within each 

cluster. In this analysis, the resulting Silhouette Score of 0.26 indicates that, while clusters are reasonably well defined, there 

is still some overlap among them. Because K-Means group’s observations based on similarity, specific job roles or skill 

profiles may not fit neatly into any one cluster, limiting the Algorithm's ability to assign all roles unambiguously. 

Additionally, K-Means assumes clusters are roughly spherical—a constraint that may not accurately reflect the actual 

structure of workforce segmentation. 

 

DBSCAN achieved a Silhouette Score measurement of 0.092709 

DBSCAN's clustering approach relies on identifying dense regions of data to distinguish core points, border points, and 

noise points; however, in this analysis, the Algorithm's performance was inadequate, as reflected by a low Silhouette Score 

of 0.09. The underlying dataset contains numerous low-density areas, which hinder the formation of meaningful clusters 

because noise points dominate the structure. As a result, DBSCAN struggled to generate workforce-related groups that align 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

8 

with AI adoption patterns. In essence, the lack of clear density-based separation within the workforce and AI-adoption data 

rendered DBSCAN ineffective for producing relevant clusters in this context. 

 

Agglomerative Hierarchical Clustering (Silhouette Score = 0.288834) 

Hierarchical clustering constructs a dendrogram by iteratively merging groups that exhibit the highest similarity. This study 

achieved a Silhouette Score of 0.29—surpassing both DBSCAN and K-Means—making it the most effective method for 

the given dataset. The resulting hierarchy reveals that job roles, their associated skill categories, and AI adoption levels 

naturally align within an intrinsic structural framework. Consequently, hierarchical clustering enables researchers to more 

accurately capture gradual transitions between job categories and predict AI adoption risks, outperforming alternative 

approaches such as K-Means in this context. 

 

DISCUSSIONS  

The best results emerged from Agglomerative Hierarchical Clustering, with a score of 0.288, indicating that workforce 

segmentation benefits from tree-based clustering that develops hierarchical structures by job role and AI integration level. 

K-Means clustering successfully segmented the workforce data into reasonably well-defined groups (0.262), although it did 

not yield flawless results. The value of 0.092 for DBSCAN indicates that there is no substantial density-based cluster 

formation between AI adoption and workforce composition. The hierarchical clustering methodology stands out as the most 

effective method for analyzing the impact of AI on the workforce, as it aligns with existing job roles and industry divisions. 

Agglomerative Hierarchical Clustering emerges as the preferred approach for segmenting the workforce based on 

AI adoption risk, as it consistently outperforms alternative methods. To refine cluster validity, decision-makers decision-

makers should pair Silhouette Analysis with the Elbow Method when selecting the optimal number of clusters (K) and 

experiment with multiple distance metrics—such as Euclidean, Manhattan, and Cosine—to improve group delineation. In 

contrast, DBSCAN should be avoided for workforce segmentation since its strength in anomaly detection does not translate 

into clear, actionable clusters for AI-driven job profiles. Nonetheless, the proposed framework remains sufficiently flexible 

to identify outlier occupations that require dedicated AI transition strategies, even outside the primary clustering process. 

Enhancements to the underlying dataset—particularly by incorporating finer-grained features like automation risk estimates, 

digital skill classifications, and industry-specific AI readiness indicators—will facilitate a more accurate grouping of similar 

roles. Finally, integrating time-series analytics to capture evolving AI adoption patterns and workforce shifts over defined 

intervals will enable ongoing monitoring and adjustment of training programs and policy interventions. 

The Hierarchical Clustering insights tool should proactively identify employees at high risk of displacement due 

to AI advancements and recommend alternative job opportunities that align with their existing competencies. The system 

should leverage AI-driven analyses to generate individualized skill development proposals, tailoring recommendations to 

each worker's assigned cluster profile. 

Government agencies and universities should utilize these analytical findings to design workforce development 

initiatives that align with current AI-driven labor market dynamics. At the same time, educational institutions must create 

adaptive AI curricula that specifically address the skill gaps within job clusters most susceptible to automation; by 

integrating robust clustering models with evidence-based best practices, workforce planning platforms can deliver data-

driven strategies that facilitate career mobility and resilience throughout the ongoing AI transformation. 

 

CONCLUSIONS  

The results of this research provide actionable insights for workforce planners, policymakers, and educational institutions. 

The research results enable evidence-based decisions for workforce development by identifying at-risk professions and 

predicting skills with longevity. Advanced analytics with machine learning (ML) models demonstrate their potential for 

guiding adaptive educational approaches, workforce reskilling programs, and national workforce strategies that will emerge 

in an AI-centered future. This analysis reveals that employment positions across logistics operations, manufacturing, and 

administrative support functions are the most prone to automation. However, creative and high-level leadership positions, 

along with analytical work, show greater sustainable potential. Research has confirmed that cross-functional competencies 

related to flexibility, alongside data processing and technological know-how, operate as defensive mechanisms against 

technical change. Targeted interventions, led by clustering outcomes, show how selected workforce blocks can receive 

upskilling programs. Insights generated from this research enable policymakers, educational institutions, and labor force 

strategists in the region to create a landmark that connects talent development to economic demands driven by AI 

technology. This research employs machine learning (ML) based methods and regional economic data to generate crucial 

insights into modern work systems for global academic discussions. 

 

 
Author Contributions: Conceptualization, B.S.; Methodology, B.S.; Software, B.S.; Validation, B.S., A.A. and V.P.S; Formal Analysis, B.S.; 

Investigation, A.A.; Resources, A.A. and V.P.S; Data Curation, B.S. and V.P.S; Writing – Original Draft Preparation, B.S.; Writing – Review & Editing, 
A.A. and V.P.S; Visualization, B.S.; Supervision, A.A.; Project Administration, A.A.; Funding Acquisition, A.A. and V.P.S. Authors have read and agreed 

to the published version of the manuscript. 

Institutional Review Board Statement: Ethical review and approval were waived for this study, due to that the research does not deal with vulnerable 
groups or sensitive issues. 

Funding: This research was funded by the University of Buraimi, Oman. The authors acknowledge the university’s support in facilitating this study. 

Acknowledgments: The authors would like to express their sincere gratitude to the University of Buraimi (UoB) for its generous support and funding of 
this research. This work was conducted as part of the internal project titled "Analyzing the Impact of AI on Academic Job Requirements and Forecasting 

Future Job Replacements," which has been officially accepted for the UoB Internal Research Grant for the Academic Year 2024–2025. The university's 

commitment to advancing academic research and innovation played a pivotal role in enabling the successful completion of this study. 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

9 

Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. 

Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available 

due to restrictions. 

Conflicts of Interest: The authors declare no conflict of interest.                                                                                                                                                                                                                                   

 

REFERENCES 
Abbasi, B. N., Wu, Y., & Luo, Z. (2025). Exploring the impact of artificial intelligence on curriculum development in global 

higher education institutions. Education and Information Technologies, 30, 547–581. https://doi.org/10.1007/s10639-

024-13113-z 

Abdalla, A. A. (2025). Navigating AI Adoption Challenges in HRM: Insights from the UAE. In M. U. Tariq & R. P. Sergio (Eds.), 

Practice, Progress, and Proficiency in Sustainability (pp. 215–232). IGI Global. https://doi.org/10.4018/979-8-3373-

1917-9.ch010 

Akinwale, O. E., Kuye, O. L., & Doddanavar, I. (2025). Scourge of replacing contemporary work environment with artificial 

intelligence (AI-dark-side): The role of capacity development in quality of work-life and organisational performance. 

Journal of Systems and Information Technology, 27(1), 116–145. https://doi.org/10.1108/JSIT-08-2024-0297 

Awashreh, R. (2025). Omani Public Governance in the Age of Artificial Intelligence: In K. Wongmahesak, J. Ahmad, & A. 

Mustanir (Eds.), Advances in Computational Intelligence and Robotics (pp. 137–166). IGI Global. 

https://doi.org/10.4018/979-8-3693-9286-7.ch007 

Azoury, N., & Karam, J. (2025). Innovation and AI Adoption: Middle East’s Path to Economic Growth. In N. Azoury & G. 

Yahchouchi (Eds.), AI in the Middle East for Growth and Business (pp. 7–25). Springer Nature Switzerland. 

https://doi.org/10.1007/978-3-031-75589-7_2 

Bahoo, S., Goodell, J. W., Rhattat, R., & Shahid, S. (2025). Artificial Intelligence in Economics Research: What Have We 

Learned? What Do We Need to Learn? Journal of Economic Surveys, joes.12694. https://doi.org/10.1111/joes.12694 

Broady, K. E., Booth-Bell, D., Barr, A., & Meeks, A. (2025). Automation, artificial intelligence, and job displacement in the U.S., 

2019–22. Labor History, 1–17. https://doi.org/10.1080/0023656X.2025.2477153 

Burton, S. L. (2024). Cybersecure Futures: Bridging AI, Employment, and the Digital Divide. In D. N. Burrell (Ed.), Advances 

in Human Resources Management and Organizational Development (pp. 155–204). IGI Global. 

https://doi.org/10.4018/979-8-3693-8562-3.ch006 

Dayalan, P., & Sundaramurthy, B. (2024). Exploring the Implementation and Challenges of AI-Based Fraud Detection Systems 

in Financial Institutions: A Review. In B. Sundaramurthy & P. Dayalan (Eds.), Advances in Business Information 

Systems and Analytics (pp. 25–38). IGI Global. https://doi.org/10.4018/979-8-3693-4187-2.ch002 

Frey, C. B., & Osborne, M. A. (2017). The future of employment: How susceptible are jobs to computerisation? Technological 

Forecasting and Social Change, 114, 254–280. https://doi.org/10.1016/j.techfore.2016.08.019 

Gayathri, J., & Mangaiyarkarasi, S. (2024). The Impact of AI Integration on Workforce Dynamics: Legal Challenges and Labour 

Law Implications in the Era of Business Technology Transformation. In B. Sundaramurthy & P. Dayalan (Eds.), 

Advances in Business Information Systems and Analytics (pp. 277–300). IGI Global. https://doi.org/10.4018/979-8-

3693-4187-2.ch013 

Hamdan, R. K. (Ed.). (2025). Sustainable Data Management: Navigating Big Data, Communication Technology, and Business 

Digital Leadership. Volume 1 (Vol. 171). Springer Nature. 

Hassouni, A., & Mellor, N. (2025). AI in the United Arab Emirates’ Media Sector: Balancing Efficiency and Cultural Integrity. 

Journalism and Media, 6(1), 31. https://doi.org/10.3390/journalmedia6010031 

Hazaa, B., & Al Mubarak, M. (2024). Digital Transformation in the Oil and Gas Industry in the GCC: Exploring Opportunities 

and Addressing Challenges. In M. Al Mubarak & A. Hamdan (Eds.), Innovative and Intelligent Digital Technologies; 

Towards an Increased Efficiency (Vol. 569, pp. 641–662). Springer Nature Switzerland. https://doi.org/10.1007/978-3-

031-71649-2_54 

Jin, X., & Han, J. (2011). K-Means Clustering. In: Sammut, C., Webb, G.I. (eds) Encyclopedia of Machine Learning. Springer, 

Boston, MA. https://doi.org/10.1007/978-0-387-30164-8_425 

Karam, J., El Hayek, Z., & Youssef, L. (2025). AI and Cultural Adaptation: Leveraging MENA Values in Business. In N. Azoury 

& G. Yahchouchi (Eds.), AI in the Middle East for Growth and Business (pp. 61–73). Springer Nature Switzerland. 

https://doi.org/10.1007/978-3-031-75589-7_5 

Kaur, H., Reddy, K. K., Reddy, M. K., & Hanafiah, M. M. (2024). Collaborative Approaches to Navigating Complex Challenges 

and Adapting to a Dynamically Changing World: In B. K. Mishra (Ed.), Advances in Mechatronics and Mechanical 

Engineering (pp. 209–234). IGI Global. https://doi.org/10.4018/979-8-3693-7076-6.ch010 

Kshetri, N., & Sharma, R. S. (2025). Editorial: Development, diffusion, and impact of generative AI in the Gulf Cooperation 

Council economies. Journal of Global Information Technology Management, 28(1), 1–5. 

https://doi.org/10.1080/1097198X.2025.2454097 

Kumar, P. (2025). Artificial intelligence (AI)-augmented knowledge management capability and clinical performance: 

Implications for marketing strategies in healthcare sector. Journal of Knowledge Management, 29(2), 415–441. 

https://doi.org/10.1108/JKM-01-2024-0111 

Lamees, A.-Z., & Ramayah, T. (2025). How artificial intelligence-based supply chain analytics enable supply chain agility and 

innovation? An intellectual capital perspective. Supply Chain Management: An International Journal, 30(2), 233–249. 

https://doi.org/10.1108/SCM-09-2024-0558 

Leveraging Predictive Analytics to Optimize Business Performance and Drive Operational Excellence. (2025). International 

Journal of Computer Applications Technology and Research. https://doi.org/10.7753/IJCATR1402.1005 

Lund, B., Orhan, Z., Mannuru, N. R., Bevara, R. V. K., Porter, B., Vinaih, M. K., & Bhaskara, P. (2025). Standards, frameworks, 

and legislation for artificial intelligence (AI) transparency. AI and Ethics. https://doi.org/10.1007/s43681-025-00661-4 

Nadimpalli, C. S. C., Muttamsetty, L. S., Pamu, R. N., & Teku, K. (2025). The Future of Employment With the Advent of AI: 



Shannaq et al., Bangladesh Journal of Multidisciplinary Scientific Research 10(4) (2025), 1-10

 

10 

Boon or Bane? In A. Günar (Ed.), Advances in Finance, Accounting, and Economics (pp. 267–296). IGI Global. 

https://doi.org/10.4018/979-8-3693-7036-0.ch012 

Naim, N., AlDebasi, A., & Price, D. (Eds.). (2025). Innovation and development of knowledge societies: Artificial intelligence 

and knowledge-based socioeconomic growth. Gulf Research Meeting, New York. Routledge. 

Nawaz, W., & Li, Z. (2025). Complexity to Resilience: Machine Learning Models for Enhancing Supply Chains and Resilience 

in the Middle Eastern Trade Corridor Nations. Systems, 13(3), 209. https://doi.org/10.3390/systems13030209 

Öztaş, Y. E., & Arda, B. (2025). Re-evaluating creative labor in the age of artificial intelligence: A qualitative case study of 

creative workers’ perspectives on technological transformation in creative industries. AI & SOCIETY. 

https://doi.org/10.1007/s00146-025-02180-6 

Pandya, B., Patterson, L., & Ruhi, U. (2022). The readiness of workforce for the world of work in 2030: Perceptions of university 

students. International Journal of Business Performance Management, 23(1/2), 54. 

https://doi.org/10.1504/IJBPM.2022.119555 

R., K. S., & Archana, T. (2025). Artificial Intelligence (AI) and Digital Competencies in the Public Sector: In E. B. Sevinç Çubuk 

(Ed.), Advances in Public Policy and Administration (pp. 95–120). IGI Global. https://doi.org/10.4018/979-8-3693-

6547-2.ch005 

Rawashdeh, A. (2025). The consequences of artificial intelligence: An investigation into the impact of AI on job displacement in 

accounting. Journal of Science and Technology Policy Management, 16(3), 506–535. https://doi.org/10.1108/JSTPM-

02-2023-0030 

Rojas, L., Peña, Á., & Garcia, J. (2025). AI-Driven Predictive Maintenance in Mining: A Systematic Literature Review on Fault 

Detection, Digital Twins, and Intelligent Asset Management. Applied Sciences, 15(6), 3337. 

https://doi.org/10.3390/app15063337 

Sarala, R. M., Post, C., Doh, J., & Muzio, D. (2025). Advancing Research on the Future of Work in the Age of Artificial 

Intelligence (AI). Journal of Management Studies, joms.13195. https://doi.org/10.1111/joms.13195 

Sander, J., Ester, M., Kriegel, HP. et al. Density-Based Clustering in Spatial Databases: The Algorithm GDBSCAN and Its 

Applications. Data Mining and Knowledge Discovery 2, 169–194 (1998). https://doi.org/10.1023/A:1009745219419 

Sigafoos, J., Lancioni, G. E., & O’Reilly, M. F. (2025). Assistive Technology Solutions for Enabling People with Intellectual and 

Motor or Sensory-Motor Disabilities to Access Communication Events. In J. Sigafoos, G. E. Lancioni, & M. F. O’Reilly, 

Communication Strategies for People with Severe Disabilities (pp. 143–162). Springer Nature Switzerland. 

https://doi.org/10.1007/978-3-031-83807-1_9 

Suleiman, A. K., & Ming, L. C. (2025). Transforming healthcare: Saudi Arabia’s vision 2030 healthcare model. Journal of 

Pharmaceutical Policy and Practice, 18(1), 2449051. https://doi.org/10.1080/20523211.2024.2449051 

Tariq, M. U. (2025). AI-Powered Business Process Automation: Revolutionizing Firm Management. In A. M. S. Derbali (2025), 

Advances in Finance, Accounting, and Economics (pp. 199–224). IGI Global. https://doi.org/10.4018/979-8-3693-7357-

6.ch010 

Tenakwah, E. S., & Watson, C. (2025). Embracing the AI/automation age: Preparing your workforce for humans and machines 

working together. Strategy & Leadership, 53(1), 32–48. https://doi.org/10.1108/SL-05-2024-0040 

Tharmalingam, L., & Pereira, M. (2025). AI Adoption & Automation Risk. Retrieved from 

https://www.kaggle.com/datasets/martnpereira/ai-adoption-and-automation-risk-san-francisco-ca 

The Future of Work: Embracing AI’s Job Creation Potential. (2025). Retrieved April 4, 2025, from 

https://www.forbes.com/councils/forbestechcouncil/2024/03/12/the-future-of-work-embracing-ais-job-creation-

potential/ 

The potential impact of Artificial Intelligence in the Middle East—PwC Middle East. (2025). Retrieved from 

https://www.pwc.com/m1/en/publications/potential-impact-artificial-intelligence-middle-east.html 

Unanah, O. V., & Mbanugo, O. J. (2025). Telemedicine and mobile health imaging technologies: Business models for expanding 

U.S. Healthcare Access. International Journal of Science and Research Archive, 14(2), 470–489. 

https://doi.org/10.30574/ijsra.2025.14.2.0398 

Visvizi, A., Troisi, O., & Grimaldi, M. (Eds.). (2025). ICT, the business sector and smart cities. Routledge. 

Wang, K.-H., & Lu, W.-C. (2025). AI-induced job impact: Complementary or substitution? Empirical insights and sustainable 

technology considerations. Sustainable Technology and Entrepreneurship, 4(1), 100085. 

https://doi.org/10.1016/j.stae.2024.100085 

Zepeda-Mendoza, M. L., & Resendis-Antonio, O. (2013). Hierarchical Agglomerative Clustering. In: Dubitzky, W., 

Wolkenhauer, O., Cho, KH., Yokota, H. (eds) Encyclopedia of Systems Biology. Springer, New York, NY. 

https://doi.org/10.1007/978-1-4419-9863-7_1371 

 
 

Publisher’s Note: CRIBFB stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. 

 

 

© 2025 by the authors. Licensee CRIBFB, USA. This open-access article is distributed under the terms and conditions of the Creative Commons 

Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0). 

Bangladesh Journal of Multidisciplinary Scientific Research (P-ISSN 2687-850X   E-ISSN 2687-8518) by CRIBFB is licensed under a Creative 

Commons Attribution 4.0 International License. 

 

http://creativecommons.org/licenses/by/4.0).
http://creativecommons.org/licenses/by/4.0/
http://creativecommons.org/licenses/by/4.0/
http://creativecommons.org/licenses/by/4.0/

