ML with Python Internship

1. DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
BABU BANARASI DAS UNIVERSITY, LUCKNOW
SESSION – 2022-23
INDUSTRIAL TRAINING PRESENTATION
[ MACHINE LEARNING WITH PYTHON ]
Submitted By -
Manish Kumar Shah
1190432099 CS42
Associated with:
Aptron,Noida
Mode : BBD Campus (OFFLINE

2. TABLE OF CONTENT
• Introduction
• Problem Statement
• Objective
• Dataset
• Methodology
• Dataflow
• Project Work
• Conclusion
• References
• Certificate

3. PROBLEM STATMENT
• The problem is to predict metal or rock objects
from sonar return data.
• Each pattern is a set of 60 numbers in the
range 0.0 to 1.0. Each number represents the
energy within a particular frequency band,
integrated over a certain period of time.
• The label associated with each record contains
the letter R if the object is a rock and M if it is a
mine (metal cylinder). The numbers in the
labels are in increasing order of aspect angle,
but they do not encode the angle directly.

4. • To design a system that can detect metal or
rock objects accurately.
• Use of Machine Learning Classification
Models –
K- Nearest Neighbor , Logistic Regression ,
SVM , Decision Tree to get the highest
accuracy .
• To build an interactive UI through which user
can provide input data & receive outcomes .
OBJECTIVE

5. DATASET
:
(Sonar, Mines vs. Rocks) Data Set from Kaggle
Number Of Attributes – 60
Number Of Instances – 208
DataSet Characteristics – Real & MultiVariate
Label associated -"R" for Rock and "M" if it is a Mine
Cylinder
It contains 111 patterns obtained by bouncing sonar
signals off a metal cylinder at various angles &
condition
It contains 97 patterns obtained from rocks under

6. METHODOLOGY

7. DATAFLOW

8. PROJECT
WORK - IMPORT LIBRARIES & READING THE DATASET

9. ROCK VS MINE DATA
DISTRIBUTION

10. Each Attribute
Distribution
Density
Distribution

11. SPLIT TRAIN-
TEST DATA

13. • For Exploratory Data Analysis & Data Cleaning used -
PANDAS , NUMPY
• To visualize the data and extract meaningful patterns , we used
following libraries –
MATPLOTLIB , SEABORN
• Used Machine Learning Model to make a model that can
classify a given sonar data with 60 attributes as either being a
ROCK or Mine Cylinder
• Used SCI-KIT LEARN Library – To import following models :
k-Nearest Neighbors , Logistic Regression , Support Vector
Machine , Decision Tree
PYTHON
LIBRABY/PACKAGES :

14. CONCLUSION
Used Classification Model to predict Rock vs Mine from
Sonar Data containing 60 attribute
Used algorithms like – SVM, kNN, Logistic Regression, Decision
Got Best Accuracy Of 82.8% using Logistic Regression Model
Developed an Input Output System for User friendly interface .

15. REFERENCES
•Content pdfs provided by APTRON, Noida .
•http://archive.ics.uci.edu/ml/datasets/connectionist+bench+(so
nar,+mines+vs.+rocks)
•Dataset from – Kaggle
•https://www.kaggle.com/code/sugamkhetrapal/project-3-sonar-
mines-vs-rocks/data
• https://scikit-learn.org/stable/auto_examples/index.html

16. CERTIFICAT
E:

17. THANK YOU 😁