Resume Recommendation System

Through AI
 In Guidance
Of
Project Guide : Shivam Chaturvedi Sir

HOD : Dr. JP Dixit Sir

 AGENDA
• PROBLEM STATEMENT • HARDWARE REQUIREMENTS
• INTRODUCTION • SOFTWARE REQUIREMENT
• OBJECTIVE • UML DIAGRAMS
• PROPOSED SYSTEM • SCREENSHOTS
• FEASIBILITY STUDY • SOFTWARE TESTING
• PROJECT SCHEDULE • CODING TIPS
• TECHNOLOGIES USED • FUTURE SCOPE
• CONCLUSION
 PROBLEM STATEMENT
Today the major problem being faced across the industry is how to acquire the right talent, using minimal resources
over the internet and in minimal time. There are three major challenges that are required to be overcome, to bring
efficiencies to the complete process.

• Separating the right candidates from the pack

• Making sense of candidate CVs
• Knowing that candidates can do the job before you hire them
Problem Definition
 INTRODUCTION
Technology today has made it possible to
send a resume within the tap of a button.
Hundreds of resumes are being received
for a particular job posting. This makes the
job of an HR department especially
difficult as it is impossible to peruse
through each of the resumes and shortlist
the candidates. Moreover, each resume has
a different format. There is a need to
extract the relevant information from the
resume and store it in the database, so that
sifting through the data becomes easier. In
order to make the job of a recruiter easier,
we propose an automated system that does
most of the preliminary filtering and
presents the data in a standard format.
 Objective
 The major objective of our system is to take the resume recommend system to other level
and makes it more flexible for both the entity.
 1) Candidates, who has been hired.
 2) Client company, who is hiring the candidates.
 PROPOSED SYSTEM
 The system proposed downloads resumes using a web crawler. The downloaded resumes are unstructured data set in .pdf format. These
are converted to .Json format using information extractor format. The converted resumes are normalized to bring all resumes on the
same platform. The resumes are further clustered based on skills, education, work experience. The system proposed is made to help
recruiters get the best candidates for a particular job profile. To facilitate this dynamic clustering is performed and resumes are scored
based on unit scoring method. The Figure 1 shows the module diagram of the proposed system.

Resume Information Data Clustering

Downloader Extractor Normaliser Unit

Job
Specification
Recruiter Candidate
Scoring and Database
Shortlisted Candidate Match Candidate
Ranking unit
FEASIBILITY STUDY
 Technical Feasibility : Our project is
technically feasible because we are
using python in our project
 Financial Feasibility : we are going to
use opensource libraries. In this project
we don’t have to invest money on it .
So, the project is economically feasible.
 Legal Feasibility : We are not using any
copyright documents or model so, the
proposed project confirms legal and
ethical requirements.
PROJECT SCHEDULE (GANTT CHART)
 TECHNOLOGIES USED

Python is used for

backend coding.

The front end part is

being developed using
Flask.

NLTK is used as a
library.

For the database, we are

using SQLAlchemy

Scikit-learn library is
used for extracting the
features of resume and
 HARDWARE
REQUIREMENTS
 Server Side:
Processor: Intel i3 or above
RAM: 4 GB or above
Display: Monitor
Others: Keyboard,
Mouse
 Client Side:
RAM: 512 MB or above
Processor: Intel Pentium Dual Core or
above
Display: Monitor
Others:
 SOFTWARE REQUIREMENTS
 Server Side:  Client Side:
• Library: NLTK • Operating System: Windows
• Programming Language: Python 7 or above

• Frame work: flask-sqlalchemy • Input: pdfminer.six

• IDE: Visual Studio Code

• Operating System: Windows 7 or
above
DFD LEVEL 0
UML DIAGRAMS

USE CASE DIAGRAM

UML DIAGRAM

Class Diagram
ER Diagram
Home Page For Admin
Home Page for User
Sign In Page
Code Blocks
SOFTWARE TESTING
1. Feasibility Testing
• Technical feasibility
• Financial feasibility
• Schedule feasibility
• Legal feasibility

2. Unit Testing
It focuses on smallest unit of
software design. In this we test
an individual unit or group of
inter related units.
 CODING TIPS

• Take short breaks

• Dry run before coding

• Comments are a blessing

• Errors are important

• Analyze others’ code

Does our Idea has
any future Scope?
FUTURE SCOPE
Further advancements that can be done
with our proposed approach is:
1. Scoring can be done based on weights
given to each parameters. Higher
weights can be given to more relevant
parameters. The relevancy of the
parameters can be measured using
past recruitment trends.

2. Personality analysis can be done of the

shortlisted candidates using social media
information provided in the resumes.
This analysis will help to judge whether
the candidate’s personality as per his/her
social life matches the job requirements.
 CONCLUSION
• We have proposed this system to make it easier for
the recruiter to select candidates. It also presents the
information in a standardized format. The raw data
we acquired through the résumés is normalized,
clustered and scored to display the top N candidates.
We have also incorporated the recruiters demands
while scoring the resume, thus making it recruiter
specific.
 Team Members

• Bilal Mirza (1836310025)

• Mohd Umair Shahid (1836310048)
• Mohd Saifuddin (1836310051)
• Saurabh Sonkar (1836310084)
• Shashank Mishra (1836310089)
Thank You