Machine Learning, Data Science, Data Analysis
Resume:
Mir Muhammad Mujtaba ***.*.***********@*****.***
Bangaluru, India
github.com/mirmmujtaba
A conscientious and multi skilled student aspiring to enter the world of duty, responsibility and professionalism. Curious about the way Machine Learning is utilized to process the data and used for creating the solutions for real life problems as well as visualizing the data by the techniques of Exploratory Data Analysis.
EDUCATION
Advanced Certification course in Artificial
Intelligence and Machine Learning
IIIT - Hyderabad
B.Tech in Electronics and Communication
Islamic University of Science and Technology
Higher Secondary Part II (C.B.S.E)
Delhi Public School, Athwajan
Marticulation (J.K B.O.S.E)
Burn Hall School, Srinagar
M.O.O.C.S
Python for Data Science and Machine
Learning Bootcamp
- Udemy
Machine Learning A - Z
- Super Data Science
Machine Learning (Coursera)
Prof. Andrew N. G
SKILLS
Python Python for Data Science
Exploratory Data Analysis Machine Learning
DataScience
PERSONAL PROJECTS
Analysis of Indian GDP
Analyzing the GDP of different states and the areas of development on which the state-wise GDP depends.
Different states were divided into 4 groups C1, C2, C3 and C4 based on their G.D.P per capita. Observations were made regarding the biggest contributors to their GDP as well as recommendations based on the lowest contributors.
Github
Analysis of Indian Trade Data
Data regarding the exports and imports of India is given. Imports have shown a cyclic trend for the mentioned duration whereas Exports have shown an increasing trend for the duration 2010-2014 and 2015-1018 seperately.
Exports and Imports were positively correlated for the duration 2015-2018 which is a positive sign for an economy. USA is the major contributor to Indian exports.
China is the major contributor to Indian imports.
Github.
House Prices: Regression Techniques
I created a Linear Regression model for predicting house prices based on various features affecting the cost of the house. The original number of input features were 81. Some of the features had to be dropped because of the null values and to avoid the curse of dimensionality .
The features were selected based on the conditions that percentage of null values be less than 40% and the correlation of the features w.r.t the Sale Price of the house. Mean absolute error was calculated to be 1.53.
Github.
Digit Recognizer
Digit Recognition system using kNN algorithm based on the pixel values of the image.
The image size is 784 pixels and all the pixels were used as the input features for the algorithm
No features could've been dropped since that would be dropping information.
Accuracy was calculated to be 96.8%
Github.
Contact this candidate