Abstract:

Students, lifelong learners and prospective employees gain new knowledge, either in an academic environment or via open courses and other learning options to enhance their skills and knowledge in their field of interest as well as improve their professional profile aiming to acquire a job position in the case of students, or further advance their careers, in terms of a better salary or position and enhanced reputation in the case of employees and lifelong learners. However, Higher Education Institutions (HEIs) oftentimes offer complicated curricula that in most cases are based on strictly academic criteria, being disconnected from the actual requirements of the job market. On top of that, they are not updated very often resulting in the content of many courses being obsolete, thus, making it difficult for students to decide which courses they should select and more importantly, also facing uncertainty on whether the skills and the knowledge they will acquire will help them enhance their position in the labour market.

This publication presents the QualiChain Recommender that is being developed under the context of the EU funded project QualiChain that aims to reengineer and disrupt the way that academic and other qualifications are archived, managed and shared and also provide the computational intelligence that is needed by both academia and the labour market to analyse data and take informed decisions. The QualiChain Recommender is an analytics and decision support component that aims to identify the skills that are in high demand in the labour market, as well as find and recommend courses, that teach these skills, to users based on their personal interests. To be more precise, the QualiChain Recommender analyses a user’s Curriculum Vitae (CV) and recommends skills that are in high demand and the respective courses that fit well with the user’s profile and have not been taught to the user in the past. Except the user’s profile, the QualiChain Recommender uses also curriculum data from the schools whose courses are recommended, as well as labour market data. As a result, the users not only dive deeper in the fields that they are more interested in, but also, give more emphasis to skills that are highly demanded by the labour market..

Of course, providing accurate personalised recommendations is not an easy task. Instead, it requires large volumes of data regarding the job market along with exhaustive analyses, the university curriculum data at a convenient form that focuses on the skills of the courses, and a mechanism for parsing the users’ CV and extracting their skills and other information about their profile. From a technical point of view, several text analytics, text matching and data mining techniques have been utilised to extract insights from job postings, the user’s profile, and a university curriculum. In this publication, the analysis and methodology that have been followed along with the results of the QualiChain Recommender are presented in detail.

Keywords:

Recommender Systems, Labour Market, Data Analytics, Data Mining, Text Analytics, Text Matching, Clustering.