SRS REPORT ON

“RESUME ANALYZER”

SUBMITTED BY

SIA63- Subodh Unavane.

SIA66- Prashant Wadile.
SIA67- Limesh Warude.
SIA65- Prasanna Vishwas.

GUIDED BY

Prof.Kanchan Bhale
DEPARTMENT OF INFORMATION TECHNOLOGY

INTERNATIONAL INSTITUTE OF INFORMATION TECHNOLOGY,

PUNE - 411057
SAVITRIBAI PHULE PUNE UNIVERSITY

2024-2025

vii
 INTERNATIONAL INSTITUTE OF INFORMATION TECHNOLOGY
HINJEWADI, PUNE-57

DEPARTMENT OF INFORMATION TECHNOLOGY

Certificate
This is to certify that the SRS report entitled
“RESUME ANALYZER”

which is being submitted by

SIA63- Subodh Unavane.

SIA66- Prashant Wadile.
SIA67- Limesh Warude.
SIA65- Prasanna Vishwas

have partially completed the Project entitled “Resume Analyzer”, under my guidance in partial
fulfillment of the requirement for the Project Based Learning in S.E. Information Technology of
International Institute of Information Technology, Hinjewadi, by Savitribai Phule Pune
University for the academic year 2024 – 2025.

Prof.Kanchan Bhale Dr. Jyoti Surve

Guide Head of the Department

Internal Examiner: - Date:-

External Examiner: -

vii
 Acknowledgement
With great joy and thankfulness, we submit this project report entitled "Resume Analyzer" as part of the
Project-Based Learning (PBL) in Second Year Information Technology at International Institute of
Information Technology, Hinjawadi, Pune.

This project has been a very enriching and learning experience for us. We are very privileged to have
worked on this project under the able guidance of Prof. Kanchan Bhale. Her regular advice, motivation,
and constructive criticism have been of great help during the course of developing this project. Her
mentorship guided us to formulate and streamline our ideas into an operational system.

We would like to convey our heartfelt gratitude to Dr. Jyoti Surve, Information Technology Department
Head, whose guidance and support ensured a healthy learning environment and facilitated valuable
resources being made available to us throughout the duration of this project.

Our heartfelt gratitude also goes to Dr. Vaishali Patil, our esteemed Principal, for granting the
infrastructure, facilities, and academic environment necessary for carrying out the project smoothly and
efficiently.

In addition, we convey our sincere appreciation to all the faculty members of the Information
Technology Department, and management for their support and motivation during the project period.
This project has played an important role in our academic knowledge, personal development, and insight
into real-world applications of Natural Language Processing and Machine Learning in the field of
automated resume assessment systems

Students Name & Signature

Subodh Unavane.
Prashant Wadile.
Limesh Warude
Prasanna Vishwas.

ix
Contents

1 Introduction to Project Topic 1

1.1 Overview 1

1.2 Brief Description 1

1.3 Problem Definition 2

1.4 Apply Software Engineering Approach 3

2 Literature Survey 4
2.1 Research Paper Summary

2.1.1 Research Paper 1 5

2.2.2
2.3.3 Research Paper 2 6
2.4.4 Research Paper 3 7
Research Paper 4 8

3 Software Requirements Specifications

3.1 Introduction 9

3.1.1 Purpose 9

3.1.2 Intended Audience and Reading Suggestion 9

3.1.3 Project Scope 9

3.1.4 Design and Implementation Constraints 9

3.1.5 Assumptions and Dependencies 10

3.2 System Feature 10

3.2.1 Frontend 10

3.2.2 Backend 10

3.3 External Interface Requirements 10

3.3.1 User Interfaces 10

3.3.2 Hardware Interfaces 10

3.3.3 Software Interfaces 10

3.4 Functional Requirements 10

ix
 3.5 Non Functional Requirements 11

3.6 Analysis Model 11

3.6.1 Data Flow Diagram 11

3.6.2 ER Diagram 13

3.7 System Implementation Plan 13

3.7.1 Tools Technologies Used 13

4 System Design 15

4.1 System Architecture 15

4.2 UML Diagram 16

4.2.1 Use Case Diagram 16

4.2.2 Class Diagram 17

5 Technical Specification 20

5.1 Voice Video 20

5.1.1 Python Code 20

5.1.2 HTML Code 22

6 Result Discussion 28

7 Glossary 34

8 Conclusion 35

9 References 36

ix
 List of Figures

Sr. No Fig. No Figure Name Page No

1 3.6.1 Data Flow Diagram 12

2 3.6.2 Entity Relationship Diagram 13

3 4.1 System Architecture Diagram 15

4 4.2 Use Case Diagram 17

5 4.3 Class Diagram 19

List of Tables

Sr. No Table Name Page No.

1 Literature Survey 4
2 Phase Wise Implementation 13

xxii
Abstract
Resume Analyzer is a web application built with Flask that helps simplify the campus recruitment
process through the provision of an intelligent and well-structured platform for analyzing student
resumes. The system provides a platform for students to register, upload their resumes, and get
automated skill-based feedback and course suggestions, while Training and Placement Officers
(TPOs) can view analyzed data through a special dashboard.

Through the integration of natural language processing (NLP), user verification, and database
management, Resume Analyzer mechanizes the otherwise labor-intensive and manually intensive
process of resume screening. Not only does this minimize human error, but it also improves accuracy
and consistency in evaluating candidates' capabilities and preparedness for industry demands.

The system guarantees that resume assessments are performed in a standard format and are available
in real time to approved users. This centralized assessment platform enables TPOs to make informed
decisions, provide customized career advice, and effectively match students to prospective career
opportunities based on skill gaps.

With an easy-to-use interface, access-controlled role-based structure, and Python-based backend in a

modular framework, Resume Analyzer can be scaled up and utilized in educational institutions of any
size. Other features that may be added in the future include sophisticated AI-based scoring models,
integration with company recruitment portals, and analytical dashboards to monitor student
performance.

6
Chapter 1
Introduction to Project Topic
1.1 Overview

Resume screening is an important step in the recruitment and selection process in today's
competitive job market. Manual screening of resumes, however, is redundant with hundreds of
applications flooding in for every vacancy, proving to be time-consuming, cumbersome, and
subject to inconsistency or human prejudice. This is not unique to industries but is also the problem
of Training and Placement Officers (TPOs) in educational institutions in the context of campus
recruitments.

The Resume Analyzer project meets this challenge by having an automated, intelligent, and
scalable system analyzing student resumes via Natural Language Processing (NLP). The system is
implemented based on Flask (Python web framework) and takes advantage of PDF parsing, regular
expressions, and keyword-matching algorithms to pull structured information such as candidate
name, contact info, education, skills, and experience from uploaded resumes.

After parsing, the system rates against a domain-based dataset of keywords to generate scores and
highlight skill gaps. Based on this report, it sends course recommendations through which students
are able to improve their profiles in targeted sections of Web Development, Data Science, Android,
iOS, and UI/UX.

Resume Analyzer provides role-based access:

Students are able to register, login, upload resume, and see scores and recommendations.

TPOs have access to all student scores and data through a centralized dashboard, allowing them to
make data-driven placement decisions.

Overall, the system increases efficiency, decreases manual labor, and facilitates data-driven
decision-making in college placement processes..

1.2 Brief Description

Resume Analyzer is a web-based system that is aimed at helping Training and Placement Officers
(TPOs) screen student resumes effectively and efficiently. Developed using Flask, Python, and
Natural Language Processing (NLP), the system offers a platform where students can register, login,
and upload their resume in PDF format.

1
As soon as a resume is uploaded, the system itself identifies vital information like name, email
address, phone number, education background, skills, and work experience based on text extraction
and pattern-matching rules. The system compares the extracted data against a predetermined list of
domain-related keywords to measure the quality of the resume and determine a resume score.

According to the skill or domain gaps identified, the system also recommends appropriate online
courses to students so that they enhance their profiles and become industry-fit. At the same time,
TPOs are provided with a dashboard where they can see all student responses, scores, and suggested
improvements in one central place.

Resume Analyzer helps to ensure resume assessments are done uniformly, eliminating the necessity
for manual screening and minimizing the potential for human error. Resume Analyzer introduces
organization, precision, and instant accessibility into the placement process, rendering it an excellent
instrument for institutions of learning getting ready to deal with campus placement.

1.3 Problem Definition

In universities, Training and Placement Officers (TPOs) screen hundreds of student resumes during
campus recruitment drives. Historically, it is done manually, which is not only inefficient and time-
consuming but also prone to human error, inconsistency, and bias. Screening resumes of different
formats and structures increases the complexity, and it becomes challenging to quickly and
objectively identify suitable candidates.

Also, students are generally unaware of the lacunae in their skillsets and do not receive individualized
mentorship to enhance their resumes according to industry standards. No uniform technique is
available to examine the quality of a resume or propose focused courses that can make the students
more employable.

Thus, there is a requirement for a computerized, smart, and scalable mechanism capable of:

 Parsing resumes received in PDF format,

 Extracting structured data reliably,
 Assess the resume on domain-specific skillsets,
 Recommend customized learning paths, and
 Offer a central dashboard for the TPOs to view student data and scores.

Resume Analyzer solves this issue by offering an web-based solution with Natural Language
Processing (NLP) that assists both the students and the TPOs in efficiently streamlining the
recruitment preparation process.

2
1.4 Applying Software Engineering Approach

To achieve systematic development, maintainability, and understandability, the Resume Analyzer

project utilizes the Waterfall Model of Software Development — a conventional linear and sequential
software development process. This model segments the project life cycle into separate phases, where
the output of a phase is used as the input for the subsequent phase.

The development phases implemented in this project are as follows:

1. Requirement Analysis:

During this stage, the project objectives and non-functional and functional requirements were clearly
outlined. The system was designed to support the upload of resumes, NLP parsing, resume scoring,
and course suggestion for students and a TPO dashboard.

2. System Design:

The system architecture, ER diagrams, and data flow were prepared based on requirements. Critical
modules like login/registration, upload file, resume parser, and dashboards were organized for
execution.

3. Implementation

The system was implemented with Python (Flask framework) as the backend and simple HTML/CSS
as the frontend. External libraries such as pdfminer, re, and spaCy were used for parsing and NLP
operations. Code was modularized to improve readability and testability.

4. Testing:

The application was tested with resumes of various formats and industries to validate correct
extraction of candidate information and correct functionality of the resume scoring mechanism.

5. Deployment and Maintenance:

The end system was also implemented locally and tested locally, and it has been implemented for
scalability purposes, ready for the future deployment into a college intranet or cloud hosting service.

Clean coding practices and programming principles, good version control via Git, as well as
comprehensive documentation, have been employed in the course of development. In this way, the

3
system would be highly dependable, manageable, and easily expanded with future added features like
analysis, live updation, or integration into foreign platforms..

4
Chapter 2
Our Resume Analyzer using NLP addresses the limitations of existing solutions by combining
rule-based and intelligent NLP techniques for accurate and structured data extraction. The project
is designed to be lightweight, cost-effective, and adaptable for educational or recruitment purposes,
ensuring both flexibility and performance..

Literature Survey Table:

SR Methodology / Limitations Improvements in Our

Paper / Tool / Technique Used Project
No
System

1 Keyword matching, Misses context, prone to Uses NLP with spaCy

Basic Applicant rule-based parsing false positives, for context-aware
Tracking cannot handle varied extraction
Systems (ATS) formats
2 Text extraction from Only extracts plain text, Used in combination
PyPDF2 (Library) PDF no intelligent parsing with NLP for
intelligent field
recognition
3 Named Entity Requires training or Customized spaCy
spaCy (Library) Recognition (NER), tuning, may not pipeline to fit
Part-of-Speech directly support resume fields like
tagging resume sections education, skills
4 Machine learning for Needs large dataset for Our system works with
Research Paper on classification of training, heavy rule-based + NLP
Resume Parsing resume sections computational hybrid method for
(IEEE/Elsevier) resources low-resource use
5 Regex-based or shallowCannot generalize acrossModular design allows
GitHub Open NLP techniques formats, hard to easy updates and
Source Resume maintain format
Parsers independence
6 AI-powered parsing andPaid, not open-source, Open-source and
Commercial ranking limited customization customizable for
Resume Parsers academic and
(e.g., HireVue, institutional use
Zoho)

2.1Research Paper Summary 1

5
Prof. C.S. Wagh et al. (2023).
Resume Ranking System Using Machine Learning & NLP.
International Journal of Scientific Research in Engineering and Management (IJSREM).
🔗 https://ijsrem.com/download/resume-ranking-system-using-machine-learning-nlp/

Summary:
The article solves the problem of effectively screening and ranking a large number of resumes during the
hiring process. It suggests an automated system that employs Natural Language Processing (NLP)
and machine learning (ML) to make the process efficient. The system seeks to identify important
information in resumes, match candidates' qualifications with job specifications, and rank resumes
according to their appropriateness. This method attempts to cut down time invested by recruiters on
manual filtering and increase the probability of selecting the top candidate.

Key Points of the Research Paper:-

• The mechanism incorporates NLP and ML for classifying resumes and ranking applicants.

• MaLSTM model (Siamese networks + LSTM with Manhattan distance) is utilized for ranking
candidates according to similarity between the resume and the job description.

• The system pulls information such as skills, experience, and education from resumes.

• The data that is pulled is saved in a structured format (JSON).

• The system architecture consists of a user interface, a processing block, and a database.

Conclusion from the Paper:

The study concludes that resume ranking via machine learning proves more efficient compared to
conventional methods using human intuition. The proposed approach will aid in constructing future
headhunting solutions through candidate ranking and re-ranking as per hiring behavior.

6
2.2 Research Paper Summary 2
Kamal Chandwani et al. (2024).
Smart Resume Analyzer using NLP.
International Journal of Advanced Research in Engineering, Science and Management (IJARESM),
12(5).
🔗 https://www.ijaresm.com/smart-resume-analyzer-using-nlp

Summary
This research study discusses the application of artificial intelligence (AI) for enhancing resume analysis
and refining the recruitment process. The researchers built a resume analyzer with the help of natural
language processing (NLP) and machine learning algorithms. The objective is to design a system
that can effectively extract prominent information from resumes and determine candidate suitability
correctly. The research investigates how various AI methods handle the analysis of resumes across
various industries and job levels. It also looks into how the formatting and content of resumes affect
candidate selection. The research findings seek to offer insights that will enable recruiters to hire
more effectively and advise job seekers on how to produce effective resumes.

Key Points from the Research Paper:

• The article discusses the issues of conventional manual resume screening, such as time
inefficiencies, biases, and inconsistencies.

• Resume analyzers powered by AI provide a solution by streamlining resume evaluation,

enhancing candidate selection accuracy, and minimizing time-to-hire.

• The resume analyzer built utilizes NLP to extract key information such as skills, experience,
education, and certifications.

• Machine learning models are applied to predict candidate suitability.

• The system is made user-friendly, with an intuitive interface and adjustable features to
support recruiters in making decisions.

Conclusion from the Paper:

The study concludes that AI can transform resume analysis and recruitment. By using sophisticated NLP
methods and developing custom analyzers, one can have deeper insights into potential candidates.
But it's important to shun typical errors such as too much dependence on keywords and the omission
7
 of soft skills. Ongoing refinement, user input, and remedying biases are necessary to build a more
equitable and streamlined recruitment process.

2.3 Research Paper Summary 3

Md Tanzim Reza, Md. Sakib Zaman (2022).

Analyzing CV/resume using Natural Language Processing and Machine Learning.
Bachelor Thesis, BRAC University.
🔗https://www.researchgate.net/publication/365299910_Analyzing_CVresume_using_natural_langua
ge_processing_and_machine_learning

Summary:-
This paper proposes a model for the extraction of significant details from semi-structured CVs/resumes
and prioritizing them in terms of a given employer's needs.
Three key phases comprise the process: segmenting the CV/resume based on subject, retrieving
structured data from unstructured format, and ranking this data on the basis of a decision tree
algorithm.
The CV/resume is rendered as HTML for enabling the extraction of structured data. The system employs
the ID3 decision tree algorithm to classify and weight the extracted information, with the positively
weighted data being employed to train the system for future application.

Key Points from the Research Paper:

• The system divides CVs/resumes into topic-based sections.

• It translates CVs/resumes into HTML in order to extract structured data.

• The ID3 decision tree algorithm is employed for classification and weighting of data.

• The system is trained on positively weighted CV/resume data.

• The performance of the system is tested using other classifier algorithms such as logistic
regression.

Conclusion from the Paper:

8
According to the research, the proposed model is capable of extracting and ranking information from
CVs/resumes efficiently. Using NLP and machine learning, the system seeks to automate and make
the resume screening process more efficient.

2.4 Research Paper Summary 4

Pratham Rasal et al. (2023).

Resume Parser Analysis Using Machine Learning and Natural Language Processing.
International Journal for Research in Applied Science and Engineering Technology (IJRASET).
🔗 https://www.ijraset.com/research-paper/resume-parser-analysis-using-machine-learning

Summary:-

This research paper discusses the issue of high-volume resume processing during recruitment. It
suggests a resume parser based on machine learning and natural language processing (NLP) to
process resumes and identify key information. The objective is to make the extraction of important
information from resumes, which may be in different formats, automatic to assist HR departments
and recruiters in finding appropriate candidates more effectively.

Key Points of the Research Paper:

• The difficulty encountered by HR departments in sifting through large volumes of resumes of

diverse structures manually is described in the paper.

• The solution involves extracting data from resumes using NLP techniques.

• The system seeks to transform unstructured resume data into a structured one for simpler
analysis.

• The analysis is conducted using machine learning algorithms, Support Vector Machines, and
Random Forest.

Conclusion of the Paper:

The research concludes that the suggested method is able to simplify the hiring process through the
automation of resume evaluation. The automation seeks to free time for both recruiters and
applicants, enhance candidate choice, and enhance general recruitment efficiency.

9
Chapter 3
Software Requirements Specification
3.1 Introduction
This chapter outlines the software requirements of the Resume Analyser system. It defines the
objectives, scope, and constraints that guide the design and development process. The goal is to
ensure that all functional and non-functional requirements are clearly identified and addressed to
create a reliable, efficient, and scalable application.

3.1.1 Purpose
The primary purpose of the Resume Analyser system is to automate the resume screening process by
using Natural Language Processing (NLP) techniques. It enables extraction of relevant candidate
information such as name, contact details, education, skills, and experience from uploaded PDF
files. The system streamlines the recruitment process by offering structured data that assists
recruiters or placement officers in faster and more accurate shortlisting of candidates..

3.1.2Intended Audience and Reading Suggestion

• Developers – for coding reference.

• Testers – to design test cases.
• TPOs/Recruiters – for understanding the system flow.
• Project Guides – for system review and evaluation.

3.1.3Project Scope

• Upload resumes in PDF format.

• Extract details: name, contact, education, skills, experience.
• Display or store extracted information.
• Allow TPO to view parsed resume data
• Recommend relevant courses based on resume.

3.1.4Design and Implementation Constraints

10
 • Only PDF resumes are supported.
• Uses Python libraries: spaCy, PyPDF, re
• Requires basic structure in resume content
• Works with Python 3.10.6 environment.
• No support for image-based resumes in current version.

3.1.5Assumptiond and Dependencies

 Resumes are readable (non-scanned).

 SpaCy model (en_core_web_sm) is installed.
 Resume contains keywords like "Skills", "Education", etc.
 Flask and necessary Python libraries are properly set up.

3.2 System Features

3.2.1System Feature 1 – Frontend

• Login/Registration for students and TPO.

• Resume upload interface.
• Dashboard to view extracted information.

3.2.2 System Feature 2 – Backend

• Resume evaluation and course recommendation logic.

• PDF parsing using PyPDF2.

3.3 External Interface Requirements

3.3.1User Interfaces

• Web-based interface (HTML + Flask).

• Upload button for resume.
• Dashboard displaying extracted data.

3.3.2Hardware Interfaces

• Standard PC/Laptop with internet.

• Minimum 4 GB RAM and basic processor.

3.3.3Software Interfaces

• Python 3.10.6
11
 • Libraries: PyPDF2, spaCy, pandas
• Web: Flask Framework

3.4 Functional Requirements

• User Registration & Login.

• Resume Upload in PDF format.
• Extract and structure resume details:
• Name, Email, Phone, Skills, Education, Experience

3.5 Non-Functional Requirements

3.5.1Software Quality Attributes

The Resume Analyser system is designed with the following key software quality attributes in mind:
• Reliability: The system must consistently extract accurate information from various resume formats
without crashing or malfunctioning.
• Maintainability: The modular structure of the code (separate functions for text extraction, parsing,
and display) allows for easy updates, debugging, and future enhancements.
• Scalability: Designed to support future improvements like batch resume processing or AI-based
analysis without major changes to the architecture.

3.6 Analysis Model

3.6.1 Data Flow Diagram

The Data Flow Diagram (Level 0) represents the overall flow of data within the Resume Analyser
system

• Input: The user uploads a resume in PDF format.

• Text Extractor Module: This module uses PyPDF to extract raw text from the PDF.
• NLP Parser Module: spaCy is used to perform named entity recognition (NER) and extract
details like name, email, phone number, skills, education, and experience.
• Data Structuring: The extracted information is organized into a structured format (dictionary or
table).
• Output: The structured data is displayed to the user or stored in a database for future use.

Data Flow Diagram:-The Data Flow Diagram of our Resume Analyzer illustrates the step-by-step
process where a student uploads a resume, the backend extracts text, applies NLP to identify key
fields like skills and education, and finally generates structured data with a resume score and course
recommendations visible to the TPO.

12
Figure 3.1: Data Flow Diagram

13
 3.6.2 Entity Relationship Diagram

The ER Diagram models the logical relationships between data entities in the system:

 Candidate: Stores basic information like name, email, and contact number.
 Education: Stores educational qualifications such as degree, institution, and year.
 Skills: Stores a list of identified technical and soft skills.
 Experience: Stores past job roles, company names, and duration.
 Relationships:
 A Candidate can have one-to-many relationships with Education, Skills, and Experience entities.
Entity Relationship Diagram:- The ER diagram of our Resume Analyzer system shows how entities
like Student, Resume, Skills, Education, and Domain are linked—where each student uploads one
resume, which is parsed into related skills and education fields, and matched to relevant domains
using the backend keyword dataset.

Figure 3.2: Entity Relationship Diagram

3.7. System Implementation Plan

3.7.1 Phase Wise Implementation

Week Task Description
1-2 Requirement gathering, problem definition, and setup of development environment (Python,
spaCy, PyPDF).
3-4 Development of Resume Parser Module to extract raw text from PDF resumes using
PyPDF.
5 Implementation of NLP techniques using spaCy for identifying and extracting entities
(name, skills, experience, etc.).
6 Testing and validation of system performance, accuracy evaluation etc.
14
3.7.2 Tools Technologies Used

 Programming Language: Python 3.10.6.

 Libraries: spaCy (for NLP and NER), PyPDF (for text extraction), re (for regex-based pattern
matching), pandas (for data structuring).
 Software: Jupyter Notebook / Python IDE.
 Environment: Windows/Linux OS, local machine setup with Python and libraries installer

15
Chapter 4

4.System Design
The Resume Analyser system is structured to process resumes efficiently using a layered architecture.
The system is divided into modules to ensure each task is handled independently and can be easily
maintained or upgraded in the future.

4.1 System Architecture

The system architecture of the Resume Analyser defines the flow of data and how different modules
interact with one another. It follows a modular architecture, ensuring separation of concerns and
making the system scalable and maintainable.

System Architecture Diagram:-The system architecture of the Resume Analyzer represents the interaction
between the user interface, Flask-based backend, NLP engine, and the database to enable automated resume
parsing and analysis

Figure 4.1: System Architecture Diagram

16
 1. User Upload Interface:
 The user uploads a resume in PDF format through a command-line interface or web-based
dashboard.
2. PDF Text Extraction Layer:
 The uploaded PDF is processed using PyPDF2, which reads and extracts the raw textual content.
3. NLP Processing Layer:
 The extracted text is passed to spaCy, which performs Named Entity Recognition (NER) to detect
fields like:
 Name
 Email
 Phone Number
 Education
 Skills
 Experience
4. Regex Extraction:
 In addition to NLP, regular expressions are applied for pattern-based extraction of contact
information such as email IDs and phone numbers.
5. Data Structuring Module:
 Parsed data is cleaned and formatted into a structured dictionary or JSON format. Optionally, it can
be stored in a CSV or database for future use.
6. Output Display Layer:
 The extracted and structured data is displayed back to the user or exported, based on the
implementation type.

This layered structure improves readability, reusability, and debugging during development.

4.2 UML Diagrams

4.2.1 Use Case Diagram

The Use Case Diagram showcases the interaction between the user (actor) and the Resume Analyser
system. It helps in identifying all the services (use cases) offered by the system from the user's
perspective.

Actors:

 User: Uploads resume and views results

 System: Processes and extracts data

Use Cases:

 Upload Resume

17
  Register
 View Resume Score
 Login
 View All Resumes

UML Diagram:-The UML diagram illustrates the main classes of the Resume Analyzer system—such as
ResumeParser, TextExtractor, EntityRecognizer, and User—and their relationships, showing how data flows
and objects interact within the application

Figure 4.2: Use Case Diagram

4.2.2 Class Diagram

The Class Diagram defines the main classes in the system and the relationships between them. It reflects
the object-oriented structure of the project.

Classes and Responsibilities:

18
🔹 1. User

 user_id: int
 name: str
 email: str
 password: str
 role: str (e.g., "Student", "TPO")

🔹 2. Student (inherits from User)

 resume_file: str (path to uploaded resume)

 score: int (resume score after analysis)
 recommended_courses: list

🔹 3. TPO (inherits from User)

 (inherits all attributes from User)

 access_level: str (e.g., "TPO")

🔹 4. ResumeParser

 resume_text: str
 extracted_skills: list
 extracted_education: list
 extracted_experience: list
 extracted_contact: dict (email, phone)

🔹 5. NLPAnalyzer

 keywords_dataset: dict
 matched_keywords: list
 score: int

🔹 6. CourseRecommender

 domain_keywords: dict
 missing_skills: list
 suggested_courses: list

🔹 7. DatabaseManager

 connection_string: str
 student_records: list
 resume_data: dict
19
  course_mapping: dict

Relationships:

 ResumeUploader uses TextExtractor

 TextExtractor sends data to EntityRecognizer
 EntityRecognizer passes structured data to ResultFormatter

These classes allow for encapsulated logic and cleaner integration of new modules in the future.

Class Diagram:- The class diagram outlines the core components of the system, including classes like
ResumeParser, TextExtractor, EntityRecognizer, and ResultFormatter, highlighting their attributes
and interactions within the object-oriented structure of the application.

Figure 4.3: Class Diagram

20
Chapter 5
Technical Specification
It includes frontend (HTML, CSS, JavaScript), backend (Flask, Python), database (MySQL), and
system requirements like authentication and automated report generation to ensure seamless
functionality.

5.1 voice video

..\voice video\voice - Made with Clipchamp.mp4

5.1.1Python Code (Snippet)

The core functionality of the Resume Analyser system is implemented in Python. Various modules
are created to handle tasks such as PDF reading, text preprocessing, information extraction using
NLP (spaCy), and formatting the output. The code follows a modular approach for clarity and easy
maintenance.

from flask import Flask, render_template, request, jsonify, send_file, session,

redirect, url_for
from werkzeug.utils import secure_filename
import nltk
nltk.download('stopwords')
from nltk.corpus import stopwords
stop_words = set(stopwords.words('english'))

import spacy
nlp = spacy.load('en_core_web_sm')
import re
import PyPDF2
import pdfplumber
import requests
import pandas as pd
import base64, random
import time, datetime
import io, os
from pyresparser import ResumeParser
from pdfminer3.layout import LAParams, LTTextBox
from pdfminer3.pdfpage import PDFPage
from pdfminer3.pdfinterp import PDFResourceManager
from pdfminer3.pdfinterp import PDFPageInterpreter
from pdfminer3.converter import TextConverter
from PIL import Image
import pymysql
21
from Courses import (
ds_course, web_course, android_course, ios_course, uiux_course,
resume_videos, interview_videos, cybersecurity_course, game_dev_course,
blockchain_course, cloud_computing_course, big_data_course,
networking_course,
robotics_course, quantum_computing_course, ai_course, vr_ar_course
)
import plotly.express as px
from yt_dlp import YoutubeDL
import json

22
 5.1.2 HTML Code (Snippet)
This HTML displays a Project Review Report with student and mentor details. It uses basic CSS for
clean formatting and structured layout. The content includes name fields, review date, and mentor
comments. Ideal for use in web apps or converting to PDF via Flask.

1. Home.html

{% extends "base.html" %} {% block content %}

<div class="container mt-5">
<div class="row justify-content-center">
<div class="col-md-8 text-center">
<h1 class="display-4 mb-4">Welcome to Resume Analyzer</h1>
<p class="lead mb-5">
</p>

<div class="row justify-content-center mt-4">

<div class="col-md-5 mb-4">
<div class="card h-100 shadow-sm">
<div class="card-body">
<h3 class="card-title">Student Login</h3>
<p class="card-text">
Access your dashboard, upload resume, and get recommendations
</p>
<a href="{{ url_for('login') }}" class="btn btn-primary"
>Student Login</a
>
<p class="mt-2">
<small
>New user?
<a href="{{ url_for('register') }}">Register here</a></small
>
</p>
</div>
</div>
</div>

<div class="col-md-5 mb-4">

<div class="card h-100 shadow-sm">
<div class="card-body">
<h3 class="card-title">TPO Login</h3>
<p class="card-text">
Access admin dashboard and manage student profiles
</p>
<a href="{{ url_for('admin_login') }}" class="btn btn-secondary"
>TPO Login</a
>
23
 </div>
</div>
</div>
</div>
</div>
</div>
</div>
{% endblock %}

24
 2. Login.html

{% extends "base.html" %} {% block content %}

<div class="row justify-content-center">
<div class="col-md-5">
<div class="card">
<div class="card-body">
<div class="text-center mb-4">
<img
src="{{ url_for('static', filename='img/logo.svg') }}"
alt="Logo"
style="
height: 60px;
filter: invert(26%) sepia(92%) saturate(1276%) hue-rotate(198deg)
brightness(91%) contrast(96%);
"
/>
<h2 class="mt-3">Login</h2>
</div>

{% if message %}
<div class="alert alert-info">{{ message }}</div>
{% endif %} {% if error %}
<div class="alert alert-danger">{{ error }}</div>
{% endif %}

<form method="POST" action="{{ url_for('login') }}">

<div class="mb-3">
<label for="email" class="form-label">Email</label>
<input
type="email"
class="form-control"
id="email"
name="email"
required
/>
</div>
<div class="mb-3">
<label for="password" class="form-label">Password</label>
<input
type="password"
class="form-control"
id="password"
name="password"

25
 required
/>
</div>
<div class="d-grid">
<button type="submit" class="btn btn-primary">Login</button>
</div>
<div class="mt-4 text-center">
<p>
Don't have an account?
<a href="{{ url_for('register') }}">Register here</a>
</p>
<hr />
<p class="small text-muted">
For TPO Login, use the credentials provided by your institution
</p>
</div>
</form>
</div>
</div>
</div>
</div>
{% endblock %}

26
3.Admin_Lgin.html

{% extends "base.html" %} {% block content %}

<div class="login-container">
<div class="login-form">
<div class="card">
<div class="card-body">
<div class="text-center mb-4">
<img
src="{{ url_for('static', filename='img/logo.svg') }}"
alt="Logo"
style="
height: 60px;
filter: invert(26%) sepia(92%) saturate(1276%) hue-rotate(198deg)
brightness(91%) contrast(96%);
"
/>
<h2 class="mt-3">Admin Login</h2>
</div>

{% if error %}
<div class="alert alert-danger">{{ error }}</div>
{% endif %}

<form method="POST">
<div class="mb-3">
<label for="username" class="form-label">Username</label>
<input
type="text"
class="form-control"
id="username"
name="username"
required
/>
</div>
<div class="mb-3">
<label for="password" class="form-label">Password</label>
<input
type="password"
class="form-control"
id="password"
name="password"
required
/>

27
 </div>
<div class="d-grid">
<button type="submit" class="btn btn-primary">Sign In</button>
</div>
</form>
</div>
</div>
</div>
</div>
{% endblock %}

28
Chapter 6
RESULT DISCUSSION:
The Project Review System processes review sheets, evaluating student projects at different stages.
The system efficiently captures mentor feedback, stores it in a structured format, and automatically
generates PDFs for record-keeping and analysis.

1. Home Page

29
30
2.Student Login:

31
Output:

32
3.Admin Login:

31
Output:

34
33
Chapter 7:
Glossary:
 Resume Analyzer – A system that reads and interprets resumes using Natural Language Processing
to extract structured data.
 Natural Language Processing (NLP) – A branch of AI used to analyze and understand human
language, enabling the system to identify useful information from resumes.
 spaCy – An open-source Python library used in the project for processing and extracting named
entities such as names, degrees, and skills from text.
 PyPDF – A Python library used to extract raw text from PDF resume files.
 Named Entity Recognition (NER) – An NLP technique used to locate and classify entities such as
names, organizations, locations, and dates within text.
 Candidate – The person whose resume is uploaded and analyzed by the system.
 Recruiter / TPO – The user who accesses analyzed resume data to evaluate candidate
qualifications and suitability.
 Dashboard – A user interface for the Training and Placement Officer (TPO) to view structured
resume data and scores.
 PDF Resume – The input document format accepted by the system for text extraction and analysis.
 Regex (Regular Expressions) – A pattern-matching tool used to extract specific data like emails
and phone numbers from raw text.
 Text Extraction – The process of converting PDF content into plain text using libraries like
PyPDF.
 Structured Output – The formatted and organized result of resume analysis, which includes fields
such as name, skills, and education.
 Resume Score – A numerical evaluation of a resume’s quality or relevance based on extracted
content and predefined metrics.
 Authentication – The process that allows only authorized users (students and TPOs) to log in and
access the system functionalities.
 Course Recommendation – A system feature that suggests additional courses based on the skills
detected in a student's resume.
Chapter 8
Conclusion
The Resume Analyzer using Natural Language Processing provides an efficient, automated solution
for Training and Placement Officers (TPOs) in colleges to screen and analyze student resumes. By
using Python libraries such as spaCy and PyPDF2, the system extracts essential details like name,
contact info, education, skills, and experience from PDF files and presents them in a structured
format. The inclusion of a resume scoring feature makes it even more helpful in evaluating
candidates based on predefined parameters.

This tool reduces manual workload, minimizes human errors, and speeds up the shortlisting process.
It demonstrates how NLP can be practically applied in academic and recruitment environments. The
modular design ensures maintainability and offers a base for future enhancements tailored to campus
placement needs.

By using this system, colleges can move towards a more digitized and data-driven placement process.
It brings transparency, speed, and consistency, making it easier for recruiters to find the right
candidates and for students to better understand their resume strengths.

Future Scope:
1. Multiformat and Multilingual Support: The system can be extended to accept .docx, .txt, and
image-based resumes through OCR, as well as support regional languages to cater to a wider range of
students.
2. Web-Based Portal for TPOs: A user-friendly web dashboard can be developed for TPOs to
upload resumes, track resume scores, and filter candidates based on different attributes like skills,
branch, or academic year.
3. Integration with Company Portals: The system can connect with external job portals or
employer APIs to automatically match student profiles with available job opportunities.
4. Cloud-Based Deployment: Hosting the system on cloud platforms like AWS or Google Cloud
will enable access across departments or institutions, allowing centralized and scalable placement
management.
5. Student Feedback and Analytics Module: A feedback system from recruiters and tracking
analytics for placed students can be added to improve future recruitment strategies and understand
hiring trends.

35
Chapter 9
References

 Rajashree Shettar, Sachin JH (2024).

Resume Parsing System using NLP and Semi-Structured Supervised Networks.
Journal of Intelligent Data Analysis and Computational Statistics, 1(3).
🔗 https://matjournals.net/engineering/index.php/JoIDACS/article/view/902

 Prof. C.S. Wagh et al. (2023).

Resume Ranking System Using Machine Learning & NLP.
International Journal of Scientific Research in Engineering and Management (IJSREM).
🔗 https://ijsrem.com/download/resume-ranking-system-using-machine-learning-nlp/

 Kamal Chandwani et al. (2024).

Smart Resume Analyzer using NLP.
International Journal of Advanced Research in Engineering, Science and Management (IJARESM),
12(5).
🔗 https://www.ijaresm.com/smart-resume-analyzer-using-nlp

 Md Tanzim Reza, Md. Sakib Zaman (2022).

Analyzing CV/resume using Natural Language Processing and Machine Learning.
Bachelor Thesis, BRAC University.
🔗
https://www.researchgate.net/publication/365299910_Analyzing_CVresume_using_natural_language
_processing_and_machine_learning

 Pratham Rasal et al. (2023).

Resume Parser Analysis Using Machine Learning and Natural Language Processing.
International Journal for Research in Applied Science and Engineering Technology (IJRASET).
🔗 https://www.ijraset.com/research-paper/resume-parser-analysis-using-machine-learning

38