Ijraset Journal For Research in Applied Science and Engineering Technology

Abstract

In this study, the researchers narrowed their focus to the application of algorithmic decision-making in ranking job applicants. Instead of comparing algorithms to human decision-makers, the study examined participants\' perceptions of different types of algorithms. The researchers varied the complexity and transparency of the algorithm to understand how these factors influenced participants\' perceptions. The study explored participants\' trust in the algorithm\'s decision-making abilities, fairness of the decisions, and emotional responses to the situation. Unlike previous work, the study emphasized the impact of algorithm design and presentation on perceptions. The findings are important for algorithm designers, especially employers subject to public scrutiny for their hiring practices.

Introduction

I. INTRODUCTION

AI, a term coined by John McCarthy in 1956, refers to the science and engineering of creating intelligent machines and computer programs. Hiring the right talent in today's competitive world is a challenging task for organizations. Recruiters strive to attract suitable candidates, provide a seamless recruitment experience, and foster a thriving workplace. Job seekers need insights into placement possibilities and salary expectations, approaching their search like product managers seeking the best market fit. The project aims to help candidates become smarter and understand the qualities and skills desired by the job market. Fairness in ranking algorithms for hiring is a crucial aspect, given the prevalence of Applicant Tracking Systems (ATS). New developments in AI and machine learning have the potential to improve human resource management, but assessing fairness and justice becomes difficult with automated procedures. AI HR tools can automate candidate sourcing and prescreening, saving time and reducing manual effort. Unconscious biases in HR decisions, such as gender, culture, and ethnicity, can be mitigated through AI-assisted chatbots and interview platforms. The project seeks to provide insights and solutions to enhance the hiring process, ensuring fairness for both candidates and organizations.

II. LITRATURE REVIEW

2020, Raghavan et al a three-phase experimental technique 52 CVs, content analysis, algorithm tweaking, and results execution, Kaggle website. The process of hiring employees is enhanced by RS based on content. After taking the organization's needs and necessary abilities into account, it assists in selecting and offering the best applicants. More candidate profiles from different industries need to be used to test the concept. The study also included a few parameters to screen resumes, however these criteria could be expanded.

2019, Cohen et al.  The study focused on how a chatbot may reduce both conscious and unconscious prejudice in hiring and was based on the Artificial Intelligence Chatbot Viewpoint. theoretical understanding that only uses case studies.

2021, Woods et al. Interviews with professionals in the hospitality industry were part of this study, which also used an HR information system with optical character recognition. The study focuses on how incorporating AI helps HR recruiters in the hospitality business while also avoiding the problems of eliminating prospects as a result of an ineffective keyword search. only concentrating on the hospitality sector, with no operational meaning of the term hospitality sector.

2020, Faliagka et al. They employed machine learning techniques and automated online recruitment. Test 100 candidates for technical jobs. developed a semantic matching, personality mining, and job application module-equipped automated e-recruitment system. only three technical roles were the focus. It is necessary to compare the applicability to another employment category. The platform selects users who have both a personal blog and a LinkedIn presence. With such predetermined parameters, strong profile prospects are therefore unlikely.

2019, Fernandez   et al In this study, automation, advanced analytics, and viewpoint were used. The importance of HR analytics is emphasised in early workforce detection and better recruitment tactics. advanced analytics were just briefly mentioned, with little emphasis placed on the application side of analytics.

2021, Gikopoulos   et al. The idea was built around automation. The results imply that in order to advance in the field of HRM for smarter recruitment, a human touch to automation is the key to blending along with Artificial Intelligence. Primary research can be used to support certain pieces of evidence, making them more reliable.

III. SOLUTION APPROACH

A web app or Android app is considered the most practical solution for enhancing the recruitment experience for both recruiters and candidates.

The web app will be developed using HTML/CSS, Bootstrap for the front-end, and Flask for the back-end. The Android app will be built using the Flutter framework along with Flask for the API.

Data science techniques, including exploratory data analysis, will be employed. Classification algorithms such as Linear Regression, Logistic Regression, Naïve Bayes, SVM, and Decision Tree Classifier will be used for model training. The API will be deployed using Heroku.

IV. PLATFORM AND TOOLS USED

A. Frontend:

1. The frontend is created using Flutter, an open-source UI software development kit.
2. Cross-platform development is possible with Flutter for Google Fuchsia, Android, iOS, Linux, Mac, Windows, and the web.
3. Platform, layout, and foundation widgets are available together with a reactive framework thanks to the Flutter framework.
4. It supports Cupertino widgets for Apple's iOS Human interface principles as well as Material Design widgets for Google's design language.

B. Backend

1. Machine learning techniques are used in the backend to do prediction jobs.
2. Algorithms like Logistic Regression, Decision Tree Classifier, Support Vector Machines, Artificial Neural Networks, and Naive Bayes classifiers can be utilised for binary classification tasks like predicting job placement.
3. Linear Regression is appropriate for regression issues like pay prediction.
4. Machine learning algorithms are used to find patterns in the data and make predictions based on the learned patterns.

C. Exploratory Data Analysis (EDA) and Data Science

1. The dataset is processed and analysed using data science approaches.
2. Exploratory Data Analysis (EDA) is used to comprehend the data, spot trends, and obtain new perspectives.
3. To explore the data, EDA employs summary statistics and graphical representations.

D. Flask

1. The Python microweb framework Flask is used for backend development.
2. It has a simple design and allows for extensions to provide more application functionality.

E. Deploying Heroku Apps

1. The app is put on Heroku, a Platform-as-a-Service (PaaS), for cloud deployment.
2. The project files will be managed through a Git repository.
3. A new Heroku app is made using the Heroku CLI, which has been downloaded.
4. Pushing the local git repository to the remote Heroku app's repository deploys the application.
5. The Heroku CLI allows for the checking of logs.
6. For configuration, the sandbox's webhook URL is reset to the URL of the Heroku app.

VI. COMMERCIAL UTILITY OF PROJECT

The project is ideal for testing on a commercial scale. This is due to the project's suitability for training on any organization's dataset. It has a number of benefits that make it ready for the market. These benefits consist of:

1. Establishing an intermediary layer between the manual auditors and the dataset and substantially    reducing manual effort.
2. Effectiveness in predicting outcomes and potential application of many coupled algorithms.
3. Shortening the process's duration and accelerating its speed.
4. Less reliance on the availability of subject-matter specialists.
5. User interface with ease of use.

Both business-to-business and business-to-customer interactions are made easier by the initiative. It's relatively easy to develop on demand for both iOS and website deployment.

Conclusion

Target clients can access specially trained data and analytics tools, which would permit better decisions, enhanced predictions, and acquire insights about the various elements affecting a candidate\'s placement, through the application of Machine Learning (ML) and Data Science. An application powered by data science would assist in digitising a process that is currently carried out manually and is, thus, obviously far more time-consuming and demands enormous quantities of work. As a result of our project endeavour, candidates would be able to: • Gain comprehension and an idea about what requirements are necessary to get successfully in an organisation thanks to our ML-based recruiter-candidate interaction tool. • Acquire knowledge and comprehension of the elements necessary for success in an organisation. • Learn more about the impact these factors have on placement. • Gain knowledge of the likely compensation they can anticipate given a particular set of marks, work experience, and other criteria. The study demonstrated the impact of AI technologies only on the Recruitment and Selection process. This study can go further by focusing on other HRM procedures like performance management or training and development. 141 samples of HR professionals from Chennai-based IT organisations were used in the current study. This study can be expanded by including larger sample sizes from other major cities, such as Bangalore, Delhi, and others. The recruiters revealed that utilizing AI technologies in the recruitment process can speed up the recruitment process and cost effective. Applying AI technologies in the recruitment process can enhance the recruitment process\'s quality with a high accuracy level and reduced human bias. The application of AI in the recruitment process can make it possible for the recruiter to get the right candidate with the right skill set for the right job with ease. Overall, the implementation of AI technologies in the recruitment process can reduce the workload for recruiters with enhanced candidate experience. It is suggested that the companies (recruiters) learn to join hands with AI technologies; they can train AI technologies to be extensions of their teams and not replace them.

References

[1] S. Bhardwaj, \"Data analytics in small and medium enterprises (SME): A systematic review and future research directions,\" Information Resources Management Journal, vol. 35, pp. 1-18. http://doi.org/10.4018/IRMJ.291691, 2022. [2] J. Fraij and V. Laszlo, \"A literature review: Artificial intelligence impact on the recruitment process,\" International Journal of Engineering and Management Sciences, vol. 6, pp. 108-121. doi:10.21791/IJEMS.2021.1.10, 2021. [3] K. G. Laudon and K. C. Laudon, Management Information Systems: Managing the Digital Firm (16th Edition). Harlow, England: Pearson, 2020. [4] S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach (4th Edition). Upper Saddle River: Prentice Hall, 2020. [5] P. Henman, \"Improving public services using artificial intelligence: possibilities, pitfalls, governance,\" Asia Pacific Journal of Public Administration, vol. 42, pp. 209-221. doi:10.1080/23276665.2020.1816188, 2020. [6] P. R. Daugherty, H. J. Wilson, and R. Chowdhury, \"Using artificial intelligence to promote diversity,\" MIT Sloan Management Review, vol. 60, pp. 59-63, 2019. [7] H. Ming-Hui and R. T. Rust, \"Artificial Intelligence in service,\" Journal of Service Research, vol. 21, pp. 155-172, 2018. [8] A. K. Upadhyay and K. Khandelwal, \"Applying artificial intelligence: Implications for recruitment,\" Strategic HR Review, vol. 17, pp. 255-258, 2018. [9] L. Cao, \"Data science: challenges and directions,\" CACM, vol. 60, pp. 59-68, 2017. [10] H. Ekbia, M. Mattioli, I. Kouper, G. Arave, A. Ghazinejad, T. Bowman, V. R. Suri, A. Tsou, S. Weingart, and C. R. Sugimoto, \"Big data, bigger dilemmas: A critical review.,\" Journal of the Association for Information Science & Technology, vol. 66, pp. 1523-1545, 2015. [11] U. Jovanovi, A. Stimec , and D. Vladusi, \"Big-data analytics: A critical review and some future directions,\" International Journal of Business Intelligence and Data Mining, vol. 10, pp. 337-355, 2015. [12] J. Fan, F. Han, and H. Liu, \"Challenges of Big Data Analysis,\" National Science Review Journal, vol. 1, pp. 293-314, 2014. [13] Othamar Gama Filho (2018) Artificial Intelligence Tech Is Evolving The Recruiting Process. Available at: https://chiefexecutive.net/artificial-intelligence-tech-evolving-recruiting-process/ (Accessed: 17 September 2020). [14] Paul Attfield (2018) How artificial intelligence is transforming the recruitment process - The Globe and Mail. Available at: https://www.theglobeandmail.com/business/article-how-artificialintelligence-is-transforming-the-recruitment-process/ (Accessed: 16 September 2020). [15] Rebecca Greenfield and Riley Griffin (2018) Artificial Intelligence Is Coming for Hiring, and It Might Not Be That Bad - Bloomberg. Available at: https://www.bloomberg.com/news/articles/2018-08-08/artificial-intelligence-is-coming-for hiring-and-it-might-not-be-that-bad (Accessed: 17 September 2020). [16] Savar, A. (2017) 3 Ways That A.I. Is Transforming HR and Recruiting. doi: 10.1080/14746689.2017.1407551. [17] Vijay Sundaram (2018) From Sourcing to Hire: Artificial Intelligence in Recruiting - HR Daily Advisor. Available at: https://hrdailyadvisor.blr.com/2018/10/02/from-sourcing-to-hire artificial-intelligence-in-recruiting/ (Accessed: 17 September 2020).

Copyright

Copyright © 2023 Surya Partap Singh, Amitesh Srivastava, Suryansh Dwivedi, Mr. Anil Kumar Pandey. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.