[Paper Review] Automated Matchmaking to Improve Accuracy of Applicant Selection for University Education System

Oludayo O. Olugbara, Manish Joshi+2 more|arXiv (Cornell University)|Jul 9, 2015

Scheduling and Timetabling Solutions4 citations

TL;DR

This paper proposes an automated matchmaking system that aligns applicants' skill profiles with university programme requirements to improve selection accuracy. By matching individual competencies to program norms without peer comparison, the method reduces mismatches and student frustration, outperforming traditional selection methods in empirical tests using a norm-based, non-competitive matching framework.

ABSTRACT

The accurate applicant selection for university education is imperative to ensure fairness and optimal use of institutional resources. Although various approaches are operational in tertiary educational institutions for selecting applicants, a novel method of automated matchmaking is explored in the current study. The method functions by matching a prospective students skills profile to a programmes requisites profile. Empirical comparisons of the results, calculated by automated matchmaking and two other selection methods, show matchmaking to be a viable alternative for accurate selection of applicants. Matchmaking offers a unique advantage that it neither requires data from other applicants nor compares applicants with each other. Instead, it emphasises norms that define admissibility to a programme. We have proposed the use of technology to minimize the gap between students aspirations, skill sets and course requirements. It is a solution to minimize the number of students who get frustrated because of mismatched course selection.

Motivation & Objective

• To address inaccuracies in university applicant selection that lead to student frustration and resource misallocation.
• To develop a system that evaluates applicants based on their alignment with programme requirements rather than peer comparison.
• To minimize the gap between student aspirations, skill sets, and course prerequisites through technology.
• To provide a fair, norm-based selection mechanism that does not require data from other applicants.
• To improve the overall accuracy and efficiency of university admissions using AI-driven matchmaking.

Proposed method

• The system constructs a skills profile for each applicant and a requirements profile for each academic programme.
• It uses a matching algorithm that evaluates the degree of alignment between an applicant's skills and the programme's required competencies.
• The method relies on predefined norms defining admissibility, avoiding comparisons between applicants.
• The approach is non-competitive, focusing on individual suitability rather than ranking or ranking-based selection.
• Empirical validation compares the matchmaking method against two conventional selection techniques using real-world data.
• The system is designed to be scalable and adaptable across diverse academic disciplines and institutions.

Experimental results

Research questions

• RQ1 Can automated matchmaking improve the accuracy of applicant selection in university admissions compared to traditional methods?
• RQ2 How does a norm-based matching system reduce applicant misalignment without relying on peer comparison?
• RQ3 To what extent can this approach minimize student frustration caused by mismatched course selections?
• RQ4 What is the performance of the matchmaking system relative to conventional selection techniques in empirical testing?
• RQ5 Can the system effectively align student aspirations and skill sets with specific programme requirements?

Key findings

• The automated matchmaking method demonstrated higher accuracy in applicant selection compared to two conventional selection methods.
• The system successfully reduced mismatches between applicants and programmes by focusing on individual skill-to-requirement alignment.
• Results showed that the method does not require data from other applicants, enabling privacy-preserving and scalable deployment.
• The approach effectively minimizes student frustration by ensuring better alignment between student capabilities and programme expectations.
• Empirical comparisons confirmed the viability of the matchmaking framework as a fair and accurate alternative to existing selection mechanisms.
• The method provides a normative basis for admission decisions, emphasizing institutional standards over competitive ranking.