What is resume parsing?

For a number of years now specialist firms have been providing software which is able to analyze text and convert it into structured information. One of the integral methods of a parsing tool is the so-called semantic analysis, which intelligently analyzes and detects the contents of a document.

There are numerous applications for the semantic analysis. From a recruitment perspective the software-driven analysis plays an important role in analyzing and converting resume documents and related files. The specialist term “cv parsing” originates from “cv”, meaning “curriculum vitae” and “parsing”, which refers to the dissecting / analyzing of the documents.

A simple example of the functionality of cv parsing

In a resume for example, both name and address need to be detected. The parser operates using a set of rules, which instruct that the name and address are typically found at either the top or at the bottom of a resume document. With the aid of a forename database, the parser then searches the designated areas for a valid forename. Simultaneously the title as well as the gender of the candidate are detected and attributed to the name. If a forename is detected then the parser assumes that a surname will be located either in front or behind it. Thus the document is parsed step by step, and based on the rules set, detects not only text but also images:

The advantages of using a resume parser

For a start the resume parser can be implemented for a job portal on a homepage. This offers the potential applicants the chance to upload their application documents without the hassle of having to fill out any complicated forms. Applicants find this particularly convenient, as they usually have their application documents to hand (Word, PDF etc.).  By reducing the hurdles and thus simplifying the procedure, the probability that visitors to the job portal actually send an application increases significantly.

For incoming applications the resume parser also ingeniously complements applicant management systems. The applicant’s email and associated attachments are sent to rexx recruitment, where relevant fields such as name, title and photo are extracted by the parser.

And not forgetting the paper applications, which can be simply scanned in and processed by the resume parser.

Regardless how applications enter the system, a resume parser ensures  that manual data entry is cut to a minimum – beneficial for both applicant and recruiter.

But why extract structured information from application documents in the first place?

On the one hand to present all candidates clearly and identically via an applicant management system, hence the “digital personnel file”. And on the other hand of course for the electronic mass-processing of e.g. the sending of personalized acknowledgments of receipt, for which the name and gender of the candidate are required.

What capabilities does rexx Enterprise Recruitment offer?

Rexx ER7 supports two different resume parsers. The resume parser “CVlizer” is a third-party partner product. During numerous benchmarking tests rexx systems identified this product as the most efficient and reliable resume parsing software, although the other products tested also delivered good results.

The second parser, which can be utilized in association with rexx Recruitment, is an in-house development: the so-called rexx built-in-parser. With high levels of security the parser detects the following data: surname, gender, title, date of birth, address, photo (via face-recognition), email, telephone and mobile number.

The following data formats are supported: pdf, doc, docxx, txt, rtf, png, jpg, gif. An OCR routine ensures that text documents in graphic format such as png or jpg can be detected and extracted (e.g. scans of qualifications or job references). Thanks to the so-called pattern recognition that the rexx built-in parser operates with, all principal European languages are supported.

The external CVlizer by comparison, detects even more text content than the rexx built-in parser, e.g. a candidate’s language skills or country of birth. On request additional data such as previous career positions, employers, school qualifications and professional knowledge can also be taken into account and analyzed.