Anjuum Khanna – Top 3 Machine Learning Projects for Beginners
                     Anjuum Khanna – Top 3 Machine Learning 
Projects for Beginners
Anjuum Khanna – In the IT sector, it is more helpful to work on 
practical projects than theoretical knowledge. It is important to 
get theoretical knowledge, but in the end, this knowledge we will 
apply in our projects. Working on real world projects helps us with 
how the algorithm works, if we made a slight change to this code 
how it would affect the projects.
In this blog post, you will discover how beginners like you can gain 
incredible progress in applying Machine learning to real-world 
problems with these awesome machine learning projects for 
beginners recommended by Anjuum Khanna. 
Top 3 machine learning projects for beginners that cover the core 
aspects of machine learning such as regression, unsupervised 
learning. In all these machine learning projects you will start with 
real world datasets that are freely accessible.
Top 3 Machine Learning Projects for Beginners
1)   Sales Forecasting using Walmart Dataset
This project is available on github, created by Gagandeep Singh 
Khanju, This is a Regression based modelling project to forecast 
the sales of Walmart. This project was created on Jupyter 
notebook, and for this project he used the “
Walmart Store Sales Forecasting” dataset, which was available on 
kaggle.
Walmart is probably the biggest retailer worldwide and it is 
significant for them to have precise conjectures for their 
deals in different departments. Since there can be 
numerous components that can influence the deals for each 
division, it becomes basic that he distinguish the key factors 
that have an impact in driving the deals and use them to 
build up a model that can help in estimating the deals with 
some exactness. According to him “ In this project, he 
conducted multiple linear regression to predict the future 
sales. There were several different factors that he analyzed 
in his regression model starting with a full model with all 
the variables and then moving towards a reduced model by 
eliminating insignificant variables. He used several different 
exploratory analyses to identify the key variables for his 
regression equation such as correlation plots, heatmaps, 
histograms etc.”
2)   BigMart Sales Predictions
This project is available on github, created by Gurudev 
Aradhye. This is a Regression based modelling project which 
can be tried to solve using two approaches XGBoost with 
hypertunning  and Random forest with hypertunning.
This project was created on Jupyter notebook, packages 
which he used in the project are pandas, numpy, sklearn, 
matplotlib and for this project you can use the “
BigMart Sales predictions” dataset, which was available on 
kaggle. According to him, “These two algorithms had their 
own importance and uses. The xgboost is used in many 
competitions. Here hyper tuning is performed with Greedy 
Search which initially takes some initial parameter values then 
it will search for parameter values which increases the 
accuracy of the model. Some details about problems are, The 
goal is to find item sales at Outlet of different types & located 
at different locations, It includes tasks such as data 
visualization, cleaning and transformation, feature 
engineering.
3) Music Recommendation system
This project is available on Github, created by Sarath Sattiraju.
This is a simple Music Recommendation System based 
on an unsupervised learning system which analyses 
multiple users playlists and gives recommendations for a 
particular playlist of a user. This model is a user-to-user 
based recommendation system. The dataset considered 
for this project is the music analysis dataset FMA. This 
project was created on Jupyter notebook, Clustering 
algorithms were used to provide predictions for the 
data. Recommendations were given based on the 
frequent genre, frequent artist, top 10 songs.
About Anjuum Khanna, Tech Blogger
Anjuum Khanna a strategic leader with a proven track 
record of over 19 years in spread heading profitable 
ventures within Fintech, eCom Start-ups, BPOs, 
Telecom & D2H, spearheaded domestic & Global 
Business Operations with large team sizes. 
Championed change management & enterprise wise 
automation initiatives within organizations in India & 
Middle East. Presently working as Vice President at 
Mswipe Technologies.