International Research Journal of Engineering and Technology (IRJET) Volume: 09 Issue: 04 | Apr 2022
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
A Review Paper on Resume scanning using python Mr. Panil jain¹,Mr. Nandkishor kamble2, Ms. Risha Nadar3, Mr. Shaikh Luqman4, 1234student,
Department of Electronics and Tele-comm Engineering, Xavier Institute of Engineering, India
---------------------------------------------------------------------***--------------------------------------------------------------------
Abstract Traditional hiring practises are becoming ineffective as internet recruitment continues to rise in popularity. This is because employment portals receive a large number of unstructured resumes – in a variety of styles and forms – from applicants with varying levels of knowledge and specialisation. As a result, structured data extraction from applicant resumes is required not just to assist automated candidate screening, but also to efficiently route individuals to their appropriate occupational groups. This helps businesses manage and organise resumes and sift out irrelevant candidates with less effort. Finding acceptable candidates for an open position can be difficult, especially when there are a lot of them. It can stymie team progress in terms of getting the right person in the right place at the right time. An automated “Resume Classification and Matching” system could greatly simplify the timeconsuming procedure of fair screening and shortlisting, as well as speed up candidate selection and decision-making. This system could handle a large number of resumes by first classifying them into the appropriate categories using various classifiers. Once that is done, top candidates could be ranked based on the job description using Content-based Recommendation, cosine similarity, and k-NN to find the CVs that are closest to the job description. Key Words: Docopt, Hiring Pattern, Human Resources, PyPDF2, Python, Resume
1.INTRODUCTION Within Human Resources, talent
acquisition is a critical, difficult, and time-consuming activity.The sheer size of the Indian market is mindboggling. Not only are one million people entering the labour market every month, but there is also a lot of churn. According to LinkedIn, India has the largest percentage of workers who are “actively looking for a new job.” Clearly, this is a very liquid and large market, but it also contains a lot of irritating inefficiencies. The lack of a consistent organisation and format is the most difficult aspect. For a resume, which makes shortlisting potential profiles for required tasks extremely time-consuming and laborious. To determine the relevance and applicability of a profile for a specific job, effective resume screening necessitates domain expertise. Short-listing presents a difficulty for the human resource department because there are so many various job roles available nowadays, as well as the customary big number of applications submitted. This is
© 2021, IRJET
| Impact Factor value: 7.529
|
exacerbated by the HR department’s lack of broad skill sets and subject knowledge, both of which are necessary for efficient HR management. India’s market is massive. Not only do one million people enter the labour market every month, but there is also a lot of turnover. The sector is currently confronted with three significant challenges: • Separating the right individuals from the pack – With millions of people looking for work in India, screening CVs and finding the appropriate fit is nearly difficult. This makes the entire hiring process slow and inefficient, wasting time and money for businesses. • Making sense of candidate CVs – The fact that CVs on the market are not standard poses a second issue. Practically every resume on the market has a different structure and format. HR must manually review the CVs in order to determine the best fit for the job description. This is time consuming and prone to inaccuracy, as a suitable candidate for the job may be overlooked in the process. • Confirming that candidates are capable of performing the job before hiring them – The third and most difficult problem is matching the CV to the job description in order to determine whether the candidate is qualified for the position for which she is being hired. In this research, we provide an automated Machine Learning-based methodology to address the aforementioned concerns in the resume short-listing process. The model takes the features taken from the candidate’s CV as input and categorises them, then maps the categorised resume to the required job description and recommends the best candidate’s profile to HR. The following are our major contributions: 1. We created a resume recommendation system that is automatic. 2. Classification techniques based on machine learning and similarity functions are applied to find most relevant resume. 2. Literature Review Over 50000 internet recruitment sites exist, all of which need job applicants to submit their resumes through their websites. Classification techniques for filtering resumes are not even used on some of these websites. The firm recruiter’s task is to manually go over all of the candidate resumes. Selecting the most capable candidates for the succeeding phases of the hiring process is a challenging assignment for recruiters. Meanwhile, some job boards have applied the clever idea of automatically scoring or classifying resumes submitted by candidates for a certain
ISO 9001:2008 Certified Journal
| Page 3849