INTERNSHIP REVIEW

MACHINE LEARNING

By:
Company Details:
M.Nandini
Upskillink , pvt lmt.
22102C01095
Department of Computer Application
 Contents
1. Introduction
2. Modules Description
3. System Analysis
4. Technology
5. Result
6. Conclusion
7. Bibliography

2
 Introduction
Upskillink is a cutting-edge education and skills development company dedicated to
providing hands-on and practical learning experiences in the field of artificial
intelligenc , with a specific focus on machine learning. As a leading provider of virtual
internships, upskillink offers a unique opportunity for aspiring machine learning to gain
realworld experience in artificial intelligence. Upskillink is a Non-profit organization
which enables Industry 4.0 ready digital workforce in India. Our vision is to fill the gap
between Academia and Industry by ensuring world class curriculum access to our
faculties and students. We want to completely disrupt the teaching methodologies and
ICT based education system in India. We work closely with all the important
stakeholders in the ecosystem Students, Faculties, Education Institutions and
Central/State Govern.

3
 Modules Description
The machine learning Virtual Internship likely covers a range of modules aimed at
providing participants with comprehensive skills and knowledge in machine learning
using python modules and libraries. Here’s the outline of what such a program might
include:
1.Introduction to Machine Learning
2.Supervised Learning
3.Unsupervised Learning
4.Neural Networks and Deep Learning
5.Libraries in python
6.Tensorflow and keras
7.Database management system

4
 System Analysis
Existing System:
The existing system uses a traditional machine learning approach to predict
household prices. The system collects historical data on household prices, features,
and location from various sources. The data is then preprocessed using techniques
such as handling missing values, encoding categorical variables, and
scaling/normalizing the data. Relevant features are engineered from the
preprocessed data, such as property type, number of bedrooms and bathrooms,
square footage, location, and proximity to schools, public transportation, and
amenities. A machine learning model, such as linear regression, decision trees, or
random forests, is trained on the engineered features to predict household prices.
The trained model is evaluated using metrics such as mean absolute error (MAE),
mean squared error (MSE), and coefficient of determination (R-squared).

5
Proposed System:
The proposed method uses a more advanced machine learning approach to predict
household prices. The system collects historical data on household prices, features,
and location from various sources, including online listings, public records, and
sensor data. The data is then preprocessed using techniques such as handling
missing values, encoding categorical variables, and scaling/normalizing the data.
Additionally, data augmentation techniques are used to increase the size and
diversity of the dataset. Relevant features are engineered from the preprocessed
data, including property type and characteristics, location and proximity to
amenities, economic and demographic indicators, and sensor data from IoT devices.
A deep learning model, such as a convolutional neural network (CNN) or recurrent
neural network (RNN), is trained on the engineered features to predict household
prices. The trained model is evaluated using metrics such as MAE, MSE, and R-
squared, as well as more advanced metrics such as mean absolute percentage error
(MAPE) and symmetric mean absolute percentage error (sMAPE).

6
Programming Languages: Technology
1. Python: Primary language for machine learning development . Machine Learning
2. Libraries: Scikit-learn: Provides algorithms for classification, regression,
clustering, and more. TensorFlow: Open-source deep learning library. High-level
neural networks API. PyTorchn : Dynamic computation graph and automatic
differentiation. LightGB : Fast and efficient gradient boosting framework . Data
Preprocessing ,Data manipulation and analysis. NumPy: Numerical computing.
3. Matplotlib and Seabron : Data visualization . Feature Engineering Libraries:
Category Encoders: Encode categorical variables . Feature tools : Automated
feature engineering . Model Selection and Hyperparameter Tuning Libraries:
GridSearch CV : Exhaustive search over specified hyperparameters . Randomized
Search CV : Randomized search over specified hyperparameters.
[These technologies can help you build a machine learning with Python]

7
Result

8
Result

9
 Conclusion
Python has solidified its position as the go-to language for machine learning due to its
simplicity, versatility, and a rich ecosystem of libraries. Its intuitive syntax, combined with
powerful libraries like NumPy, Pandas, Scikit-learn, TensorFlow, and PyTorch, empowers
data scientists and machine learning engineers to:
1.Efficiently Process and Analyse Data: Handle large datasets, clean, preprocess, and
extract valuable insights.
2.Implement Diverse Machine Learning Algorithms: Explore a wide range of algorithms,
from classical techniques like linear regression and decision trees to cutting-edge deep
learning models.
3.Rapidly Prototype and Experiment: Quickly iterate on ideas and build robust machine
learning models.
4.Visualize Data and Model Results: Gain deeper understanding through insightful
visualizations.
5.Deploy Models into Production: Integrate models into real-world applications, making
informed decisions.
The Future of Python in Machine Learning
As the field of machine learning continues to evolve, Python's role is poised to grow
further.
10
 Bibliography
The following books are referred during the analysis and execution phase of the
project.
Books:
1. Python Machine Learning by Sebastian Raschka (2015)
2. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow by
Aurélien Géron (2019)
3. Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville (2016)
4. Python Data Science Handbook by Jake VanderPlas (2016)6)"Big Data: Concepts,
Technologies, and Applications" by Gabor Fari and Orsolya Valló:An in-depth
analysis of big data technologies and their applications in modern data engineering.
5. Research Papers:
A Survey on Deep Learning for Predicting Household Energy Consumption by J. Li
et al. (2020)2. Household Energy Consumption Prediction using Machine
Learning and Deep Learning Techniques by S. S. Iyer et al. (2019)

11