Matching and Skills Suggester on Job Market Finland

Skills Suggester

Skills Suggester is Job Market Finland's AI service that helps the user to select suitable keywords for their job search profile, for example. The keywords are professions and skills according to the European ESCO classification. You can find the ESCO glossary and general additional information on the European Commission's website. 

https://esco.ec.europa.eu

Keywords affect the matching calculation, so the user gets better workplace recommendations when they select enough matching keywords.

The operating principle of the Skills Suggester is relatively simple. The text entered by the user is compared with the descriptions of different professions. The aim is to find professions that best match the content of the text. For example, if the text mentions “forest” and “trees”, according to the suggester feature, suitable professions could include “forestry worker”, “forestry machine operator” or “forest management advisor”. Similarly, the suggester feature looks for skills that match the content of the text, which in this example could be “inspecting trees”, “protecting forests” or “planting trees”.

Skills and professions are closely related, so related skills are suggested for the professions found and related professions are suggested for the skills found. The professions and skills selected by the user also have an impact on the recommendations. Therefore, it is a good idea to update the suggestions after selecting some keywords.

As its name suggests, the Skills Suggester makes suggestions for suitable keywords, but the user (e.g. job applicant or employment authority) ultimately decides which keywords to choose.

Matching calculation

Matching calculation, or matching for short, is an AI-based functionality of Job Market Finland that aims to match job applicants to job postings. In other words, it looks for both suitable job advertisements for job applicants and suitable job applicants for employers’ vacancies.

The matching consists of several search algorithms, which provide a higher score the better the job applicant and job posting match. The scores generated by these algorithms are added up and, finally, the job postings or job applicants with the highest scores are presented to the user. Some of the algorithms are very simple, and matching keywords (e.g. “locomotive driver” or “monitor construction site”) score high. Matching language skills or driving licences also provide additional points. Some algorithms, on the other hand, use more complex methods to analyse written text and, for example, search texts for common words or terms belonging to the same subject area. For example, the words “lawyer”, “law” and “court” belong to the same subject area and thus link texts even if they do not use the same words.

Naturally, matching a suitable workplace with a job applicant is not always simple. The job may have special requirements that the matching calculation cannot take into account, the job applicant may not have described their skills in the words used by the employer or the most essential keywords may have been omitted. Therefore, it is good to understand that matching calculations do not only offer one correct job or job applicant, but a set of opportunities from which the user can choose the ones that suit them best.

Updated: