The document summarizes an internship project on mushroom classification using machine learning algorithms. The intern was assigned to build a model to classify mushrooms as edible or poisonous based on features in a dataset. Decision tree and random forest classifiers were used and evaluated based on graphs of results. The random forest and decision tree algorithms provided better classification accuracy compared to other algorithms for this mushroom dataset. In conclusion, the project aimed to develop a machine learning approach to identify edible versus poisonous mushrooms based on their characteristics.
1. DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
INTERNSHIP REVIEW
ON
“MUSHROOM CLASSIFICATION USING MACHINE LEARNING”
Under the Guidance of: Seminar By:
Dr. M. K Shanker Ganesh Nayana R [1EP17CS050]
Associate professor,
Department of CSE,
EPCET
2. CONTENTS
About the Company
Introduction
Task Assigned
Aim of the Project
System Design
Functional Requirements
Implementation
Results
Conclusion
References
3. ABOUT THE COMPANY
Name: PraLoTech Solutions LLP
Domain : Machine Learning with Python
Topic : Mushroom Classification
Duration : 1 month
Mission of the Organization : “To enable its customers to achieve
total e-commerce through innovative solutions using the cutting
edge technologies and to provide world class IT and ITES
services at affordable costs to the customers with fast turnaround
time and to continually improve the service delivery at the client
service centre’s managed by us.”
4. ABOUT THE COMPANY
Pralotech Solutions is completely dedicated to the success of
our customers and does not permit external forces to diminish
our focus and commitment.
To achieve the highest level of customer satisfaction, we follow
basic principles to deliver solutions with impact.
Motive
The extensive software development in system integration,
web page design, web page development, e-commerce, iphone
development, android development, blackberry development,
content management system, open source technologies like
joomla customization, wordpress customization, zencart
customization, angular front end customization etc.
5. INTRODUCTION
Mushroom is one of the fungi types’ food that has the most
potent nutrients on the plant.
Mushrooms have major medical advantages such as killing
cancer cells.
• This study aims to find the
most appropriate technique
for mushroom
classification, and
mushroom will be classified
into two categories,
poisonous and edible.
7. Nowadays, there are different challenges to develop systems
that analyze a huge and complex data to make better decisions.
This study aims to find new approach working to classify the
mushrooms based on different features using the different
techniques of Machine Learning (ML).
In the proposed approach, we used the training dataset that
contain the mushroom data to classify it into poisonous and
nonpoisonous(edible).
The data set contains
mentioned features of the
mushroom which can be
seen in the image.
8. TASK ASSIGNED
In the first week of internship, the company assigned to learn
the basics of Machine Learning like Categories of ML:
Supervised, Unsupervised and Semi supervised
And in the later days we learnt different types of algorithms in
machine learning.
They assigned to build a model for Mushroom Classification.
The proposed technique for mushroom classification is decision
tree and random forest. classifier assumes the presence of
particular feature of a class which is not related to the presence
of any feature. This is the supervised classification technique.
9. Aim of the Project:
This project aims at developing a machine-learning algorithm
that will determine if a certain mushroom is edible or poisonous
by its specifications like cap shape, cap color, gill color, etc.
using different classifiers.
To do so, I have used the following classification methods:
Decision Tree Classifier
Random Forest Classifier
10. SYSTEM DESIGN
Fig 1: Phases
The figure shows the research phases for the proposed approach.
12. IMPLEMENTATION
Implementation Using Machine Learning:
In Machine learning has two phases, training and testing.
And build a machine learning model and apply different
algorithms like decision tree and random forest.
After building
the trained
model we
evaluate the
results in term
of graphs.
Fig 2: Methodology
13. RESULTS
Fig. 3 : Distribution of class for
white Mushroom
Fig. 4 : Odor of Mushroom
15. CONCLUSION
Even if a mushroom is considered edible, it is wise to only
consume a small amount if it is a species the person has not eaten
before, in case the person is sensitive and has an adverse reaction.
It is deduced that Random forest and Decision tree Classification
algorithm provides better results for Mushroom Classification
dataset when compared with other classification algorithms.
16. REFERENCES
M. Alameady, “Classifying Poisonous and Edible Mushrooms in the Agaricus,”
International Journal of Engineering Sciences & Research Technology, vol. 6, no. 1,
pp. 154–164, 2017.
M. Tawarish and K. Satyanarayana, “A Review on Pricing Prediction on Stock Market
by Different Techniques in the Field of Data Mining and Genetic Algorithm,”
International Journal of Advanced Trends in Computer Science and Engineering, vol.
3, no. 23– 26, 2019.
A. Deshpande and R. Sharma, “Multilevel Ensemble Classifier using Normalized
Feature based Intrusion Detection System,” International Journal of Advanced Trends
in Computer Science and Engineering, vol. 8, no. 3, pp. 874–878, 2019.
D. Chowdhury and S. Ojha, “An Empirical Study on Mushroom Disease Diagnosis : A
Data Mining Approach,” International Research Journal of Engineering and
Technology(IRJET), vol. 4, no. 1, pp. 529–534, 2017.
S. Beniwal and B. Das, “Mushroom Classification Using Data Mining Techniques,”
International Journal of Pharma and Bio Sciences, vol. 6, no. 1, pp. 1170– 1176, 2015.
“Mushroom Dataset.”, Retrevided from http://www.mushroom.world/.