P.E.S.
College of Engineering, Mandya – 571401
(An Autonomous Institution affiliated to Visvesvaraya Technological University, Belgavi)
Aided by the Govt. of Karnataka | Accredited by NBA & NAAC, New Delhi.
Department of Electronics and Communication Engineering
" RESUME SCREENING RATING AND RECOMMENDING SKILLS USING NLP"
Team Member Under the Guidance of
AMEENA KAUSAR 4PS19EC009 ShreeLakshmi C.S
ANANYA.M 4PS19EC013 Professor
ANISHA GEORGE 4PS19EC015 Department of Electronics and Communication Engg
DARSHAN K.R 4PS19EC034 P E S College Engineering, Mandya
Presentation
• Introduction And Objectives
Agenda • Objectives Achieved
• Literature Survey
• Design and implementation
• Individual Contribution Of the Project
• Development
• Requirements Analysis
• Results Obtained
• Reference
Introduction
Screening the resumes is mostly done using Natural Language Processing (NLP). Natural language refers to the way we humans
communicate with each other and is concerned with giving computers the ability to understand the text and spoken words.
NLP combines computational linguistics- rule based modeling of human language with statistical,machine
learning and deep learning models. Together combining these technologies helps computers process the
way human language works in the form of texts or voice data.
As the job market is growing in India, millions of new job seekers are joining the workforce every year, as per LinkedIn. If the
Companies hire in bulk there are company with an acceptable timeline.
Even here the Applications are in millions which is tedious task to go through them,hence these Companies use various machine
learning models(NLP) which will rank out the top resumes and fetches the best fit for the job role.
Objectives
1.Resume screening, rating and skills recommendation system automatically using natural language processing.
2. To improve the resume building quality.
3. To make resume selection easier by just uploading the resume.
4.To Achieve the Results in Voice manner.
.
Objectives Achieved
1. It helps to shortlist the number of resumes in less time with more efficiency.
2. It also suggests the required skills aswell as Courses from the applicants for the particular role in case if the skills
does not match with the job role and it also rates the resumes based on their skills, experience and certifications.
3. It helps With improvising the results by adding Voice Results.
Problem Statement
Pursuing human things through machines is the current hype in the information technology domain, especially
in Artificial Intelligence (AI). As a part of AI, machine learning had grabbed the vogue by spreading its
branches to various dimensions and degrees.
Deep learning is one of the latest areas, being applied in the massive processing-related fields. In the area of
NLP, various tools and technologies have been groomed and faded for their comparative pros and cons.
Candidate‘s job profiles can be screened by analyzing.
Literature Survey
Sl. No Author Title Objectives Drawbacks
1. Pradeep Kumar Roy Smart Evaluation For Job Text Classification using N- • Accuracy is Low.
Vacancy Application System grams. Used Machine Learning • Compared only the
to perform Classification with Outcomes with different
62.4% accuracy. algorithm.
2. Bhaliya , NLP based Extraction of Developed customed data set of • focused only on accepting
Nirali , relevent resume using machine 10343 resumes which was and rejecting resumes.
Jay Gandhi and learning acquired by the private resume
Yong Luo management company.
3. Sujit Amin Web Applications for Resume Web applications is divided • Candidates are not selected
screening into 3 divisions – based on skills.
1.Job applicant side • Designation is not
2.Server side classified.
3. Recruiter side
4. Riza tanaz Fareed , Resume Classification and Resume is passed through NLP • They didn’t have more
Rajath V , ranking Using KNN and Cosine pipeline where the words are features like rating and
Shraddha Devi similarity. extracted.Techniques like recommending.
Stop words,lemmatization to
correct words.
Design and Implementations
Block Schematic System
NATURAL
RESUME PDF PYTHON EXTRACTED
LANGUAGE
MODULE FEATURES
PROCESSING(NLP)
RESUME DESIGNATION SKILLS
RATING CLASSIFICATION RECOMMENDATI
ON
• UPLOAD RESUME IN PDF FORMAT
• PYTHON MODULE FETCHES THE UPLOADED RESUME
• TEXT DATA IS EXTRACTED FROM PDF AND FORWARDED TO NLP TECHNIQUE.
• THEN FEATURES ARE EXTRACTED AND BASED ON BAG OF SKILLS, SKILLS ARE RECOMENDED.
• RESUME IS RATED BASED ON SKILLS, CERTIFICATION AND RESUME CONTENT.
Individual contribution of the project
importing libraries and display design –Ameena
Interface development –Darshan
Voiceover Code –
What is NLTK?
NLTK is an open-source toolkit for natural language processing.
This toolkit is one of the most powerful NLP libraries which contains packages to make machines
understand human languages and respond in an appropriate manner.
Using NLTK we can perform operations such as data cleaning, visualization, and vectorization that
will help us in classifying our text.
How does it work?
We can preprocess text data, convert it into a bag of words model, and perform sentiment analysis
using Vader's sentiment analyzer.
We can perform various tasks using NLTK such as discussed below:
• SEGMENTATION • LEMMETIZATION
• TOKENIZING • SPEECH TAGGING
• STOP WORDS • NAMED ENTITY TAGGING
• STEMMING • CLASSIFICATION
SEGMENTATION:-
BREAKING THE SENTENCES IS CALLED SEGMENTATION.
Eg: cricket was invented in England. cricket is a rich game now.
The above sentence is segmented and then the result would be
•cricket was invented in England
•cricket is a rich game now
TOKENIZING:-
THE SEGMENTED SENTENCE IS THEN TOKENIZED
Eg: cricket •Was •Invented •In •England (and so on..)
STOP WORDS:-
AFTER TOKENIZATION OF WORDS STOP WORDS ARE REMOVED
THE STOP WORDS LIKE -> IS THE ARE etc.
STEMMING:-
MAKING THE MACHING UNDERSTAND WORDS LIKE:-
SKIP + ING
SKIP + S
SKIP + ED AS SIMPLY SKIP
Lemmatization:-
Lemmatization usually refers to doing things properly with the use of a vocabulary of words, normally aiming to
remove inflectional endings only and to return the base or dictionary form of a word
THIS PROCESS TRIES TO UNDERSTAND THE BASE WORDS LIKE MOOD GENDER etc.Also simply called as lemma.
SPEECH TAGGING:-
IDENTIFYING THE PARTS OF SPEECH FOR WORDS LIKE NOUN, VERB etc
NAMED ENTITY TAGGING:-
UNDERSTANDING THE MACHING ABOUT PERSON , NAME , LOCATION, ORGANIZATION etc
CLASSIFICATION:-
NOW BASED ON THESE PROCESS THE DATA IS SIMPLY CLASSIFIED ON THE REQUIREMENT
Requirements Analysis
PROGRAMMING LANGUAGE
• Python3.7
PACKAGES
• Pandas-Library used for working with data sets.
• Base64-Used to convert bytes that have binary to ASCII characters.
• Numpy-Used to working with arrays
• PyResparser-Extract information from resumes
• PdfMiner-Tool for extracting information from PDF documents
• Plotly-Provides online graphing, analytics and statistics.
Results:-
FRESHER LEVEL:
Experience not added.
Internships not added.
Current skills are less compared to expected skills.
Hobbies and interests are not added.
Achievements and certifications are not added.
Objective and summary is added.
Education details added.
RESUME SCORE - 44
INTERMEDIATE LEVEL:
Experience is intermediate.
Internships are added.
Skills are more compared to fresher level.
Projects are added.
Achievements and certification not added.
Hobbies and interests not found.
Objectives and summary is added
RESUME SCORE - 66
EXPERIENCED LEVEL:
Experience is added.
Internships are added.
Skills are more compared to fresher and intermediate level.
Achievements and certifications are added.
Projects are added.
Education details are added.
Objectives/summary is added.
RESUME SCORE - 91
Utility of Project
• Resume Screening is the process of identifying a subset of resumes that are most relevant to a given job opening.
• The goal of resume shortlisting is to minimize the time and effort required to find qualified candidates by using machine learning algorithms
to automatically identify relevant resumes.
• By using machine learning and NLP to screen resumes, you can more efficiently identify candidates who have the skills and qualifications you
are looking for.
• Machine learning / NLP could also be used to shortlist the most qualified candidates by extracting key words and phrases from resumes. ML /
NLP can be used to identify key skills and experience mentioned in a resume. These skills and experiences can then be used to group these
elements together and compare them against a set of predetermined criteria.
• This can help to quickly identify applicants who are a good match for a particular role.
• machine learning models (classification) can be used to find / classify resumes with probability score which can be indicative of success in a
particular job. For example, if a job requires customer service skills, ML classification models can be used to identify resumes that mention
customer service experience.
THANK YOU
