An Enhanced Freelancer Management System with Machine Learning-based Hiring

  • A Fatwimah Humairaa Mahomodally School of Innovative Technologies and Engineering, University of Technology Mauritius, La Tour Koenig, Pointe-aux-Sables, Mauritius https://orcid.org/0000-0003-0583-818X
  • Geerish Suddul School of Innovative Technologies and Engineering, University of Technology Mauritius, La Tour Koenig, Pointe-aux-Sables, Mauritius https://orcid.org/0000-0003-4927-4902
Keywords: Freelancer Management System, Freelance entrepreneur, Freelancer, Intelligent hiring, Machine Learning, Natural Language Processing

Abstract

Existing Freelancer Management Systems are not being adequately efficient, inconveniencing to a certain degree the freelance workforce, which comprises around 1.1 billion freelancers globally. This paper thereby aims to resolve the impediments of similar existing systems. Pertaining to the methodology, qualitative analysis was adopted. Interviews, participant observation, interface analysis, workshop documents, research papers, books and articles were used to draw data about similar applications. A web application was implemented to fulfil the objectives by using WAMP as a local development server, Visual Studio Code as a source code editor, and HTML, PHP, Python, SQL, JavaScript and CSS as programming languages along with Ajax for requests-handling functionalities, and already available APIs, and jQuery and Python libraries. The contributions brought forth are providing a shortlist of the best-qualified freelancers for each project via Machine Learning technique, generating an automated invoice and payment as soon as an entrepreneur supplies a monetary figure when approving the deliverable of a project, and enabling freelancers to sign contracts electronically to comply with business terms on one centralised repository, unlike existing systems which do not support these 3 features together on the same platform. The multivariate regression model used for intelligent hiring performs satisfactorily by yielding a R2 of around 0.9993.

Published
2022-01-01
Statistics
Abstract views: 822 times
PDF downloads: 220 times
Section
Articles