What is Artificial Intelligence for the HR sector
It is a system for searching and indexing the professions contained in the curriculum vitae, in the job vacancies (specifically, within the job description) and in the internal applicant database.
Artificial Intelligence does not replace keyword research, also known as full text index, but it enhances it through a specific semantic engine for talent management. Recruiters will have a recruiting matching system at their disposal, which is able to carry out the candidates’ pre-screening activity independently.
CV writing languages and job offers are very different from each other, only a semantic matching engine can narrow the gap between the two.
A semantic engine for the HR world
There are many semantic engines on the market but most of them have a purely linguistic connotation: they are very good at reading common texts but not at analyzing vertical documents, such as job vacancies and CVs, by nature.
Our technology was born and implemented to get the most out of reading HR documents, with a particular focus on the analysis of skills in the field of talent acquisition and talent management.
Advantages of AI Matching Recruiting
The semantic engine is able to perform a bidirectional matching between resumes and job descriptions and an automatic ranking of profiles that are compatible with the searches. In this way the recruiter can start the selection process with a pool of candidates already ranked by skills; the profiles are then read by a parsing system that identifies and extracts professional skills in the form of tags.
Our AI matching engine is multilingual and cross-lingual; it is therefore able to do the matching even when the CV and the job description are written in different languages.
Artificial Intelligence applied to recruiting processes allows the candidate to streamline one of the longest, and often most complex activities: the insertion of work experiences. It will not be necessary to fill in any format but simply to attach the CV. A multilingual CV parsing will identify the candidate’s job title and extract the skills acquired from the profile automatically. However, the candidate has the possibility to make changes to the result offered by the semantic reading.
After completing the subscription form, each candidate will also be able to see the level of compatibility of their profile with job vacancies.
Once CV and job description have been obtained, the “job semantic matching software” comes into play, by doing four actions in a few seconds:
1. Parsing and data extraction
The text contained in the curriculum is extracted, be it a reading file or an image, as a .jpeg. The system is able to analyze images, thanks to an integrated image processing software based on neural networks, which parses the CV and classifies the text content into sections.
2. Data normalization
The data are normalized in order to improve understanding and a grammar software conducts a logical analysis on the extracted content to identify the semantic meaning of each sentence in the text.
3. Semantic analysis
Our calculation algorithm searches and indexes professional skills, skill works experience, in the analyzed text, that is, the tasks that uniquely identify a profession.
The semantic system is based on an Elasticsearch database.
4. Indexing and matching
Professional skills are analyzed and evaluated by calculation algorithms, in order to be accepted or rejected on the basis of professional proximity parameters.
A similar operation is carried out on job offers, whose analysis returns the matching of CVs that are suitable for the selected talent pool.
The AI generates the percentages according to which the candidates are ranked.
AI matching blog
AI matching FAQ
Once users have posted a job ad, our semantic engine, is able to detect the related duties and tasks and to rank the applicants based on the professional skills included in their resumes. SkillSkan highlights the skills detected in the CVs that match with the job position, even though the texts were written in a different way. The software is able to analyze sentences logically as well.
Yes, it is possible to search the candidate database at any time. By performing a search by professional title, it is possible to use the semantic search engine to find the candidates who correspond to a specific job title, and therefore to the relative tasks, for a bidirectional matching between job offers and candidates. These searches can be saved, which allows having always updated folders with the new candidates corresponding to the selected parameters. The search can be performed on all candidates within the database, on the candidates registered to your ads or on those for which an evaluation has been added.
AI matching software is an easy-to-use tool on both the company and candidate sides. Candidates can actually apply by filling out a form: they can easily enter their personal data and attach their CV. Our semantic system will define the compatibility of the profile with the job offer.
Within the system we have a module dedicated to statistics that allows you to view reports relating to ads posted and current searches or check the activities entered by recruiters. Furthermore, statistics on the registered candidates are available for each job, which allow for example to understand the origin of the applications. You can even consult the details related to your database (total number of candidates, assessed candidates, country/domicile, etc.) at any time.
The AI matching solution system works with a semantic engine capable of reading and classifying CVs and job offers, by automatically pre-screening the applicants’ profiles.
The ranking is not limited to the candidates who applied to the job, but also includes those present in the CV database. In fact, after reading the job description, the matching engine scans the CVs present in the database and proposes the most suitable profiles to the recruiter, by creating a first talent pool of passive candidates.
Elasticsearch is a scalable full text search engine that allows the collection and display of data relating to the recruitment of candidates, by identifying different behavior targets.
The system enables an active and proactive approach to recruiting, by providing the recruiter with behavioural patterns and other information needed to maximize results and find the right people, with appropriate tools and as quickly as possible.
Semantic technology defines and connects information by developing languages to express rich and self-descriptive interrelationships of data in a form that machines can process and store. In this way, the machine is not only able to process long strings of characters and contextualize them, but also allows you to store, manage and retrieve information based on meaning and logical relationships, and relate them to each other.