Btm-Layout *St stage, Madiwala,Bangalore
Gmail: adartn@r.postjobfree.com
GitHub: https://github.com/Harisuryatm
LinkedIn: https://www.linkedin.com/in/hari-s-129465a0/ HARISURYA TM
OBJECTIVE
Seeking for a career to utilize my knowledge, personal skills to gain comprehensive understanding of reputed organization so as to take responsibility and contribute significantly.
I educate, refine and drive myself to be a better person
Self-taught myself in Machine Learning after my graduation.
I stay calm when faced with adversity.
I am constantly learning because I never settle. EDUCATION B-TECH, ELECTRONICS AND COMMUNICATION, 2014 - 2018 Amrita school of Engineering, Coimbatore (DEEMED) GPA : 6.50
HIGHER SECONDARY XII, COMPUTER SCIENCE, 2013 - 2014 Saratha higher secondary school, Erode district
PERCENTAGE : 86 %
TRAININGS
AND
PRACTICAL
KNOWLEDGE
ALGORITHMS USING PYTHON, INTRODUCTION TO MACHINE LEARNING, ARTIFICIAL INTELLIGENCE
NPTEL (online) – Python 3
July – 2018
Description : Based on implementing from scratch to better understand the core mathematics of Supervised Algorithms . Hands-on Machine Learning with Python Book
Description : This Book is the best resource and let me to dive into projects and more insights and strategies to achieve (Data pre-prococessing, feature Engineering and feature selection) the model with good accuracy.
Page 2 HariSurya TM
PROJECTS Customer Churn Prediction Model
This case requires trainees to develop a model for predicting customer churn at a fictitious wireless telecom company and use insights from the model to develop an incentive plan for enticing would-be churners to remain with company. This dataset helps to analyse and to build a perfect model with better accuracy. Link : https://github.com/Harisuryatm/Customer-Churn-Prediction Employee Attrition Prediction Model
aim to predict whether an employee of a company will leave or not, using Logistic Regression algorithm. The Dataset used IBM HR Analytics Employee Attrition, which was downloaded from Kaggle. The dataset contains 1470 rows (records) and 35 columns
(features) was split, using 70% for training the algorithm and 30% for testing it, achieving an accuracy of 88%.
Link : https://github.com/Harisuryatm/Employee-Attrition-Prediction-Model- Loan Dataset Analyses.
The Real time Dataset consists of some customer information and their associated loan information from company. Task here is to clean the data, exploratory data analysis on ANDREW NG ‘S MACHINE LEARNING COURSE
Coursera (online)
July – 2019
Description : Gained knowledge about workings of Neural Networks and various optimizers to tune the model. And mostly, how to handle the data so that model’s performance becomes high. This course is the gateway to do my projects basically. SKILLS
LANGUAGE : Python 3, OpenCV
BACKEND : MySQL (RDMS)
IDE : Spyder, PyCharm, Jupyter
DATA VISUALIZATION : Seaborn, Matplotlib, ggplot MACHINE LEARNING : scikit-Learn, TensorFlow, Keras ( Neural Networks, Deep Learning (CNN))
DATA PROCESSING : Numpy, Pandas
data mining, manipulating data sets and building statistical models, proficiency with large datasets.
I hereby confirm that the skills given above is true to the best of my knowledge. Page 3 HariSurya TM
it and infer from the data about the trends in loans taken and products bought. Link : https://github.com/Harisuryatm/Loan_Dataset_Cleansing Classifying whether the person is at risk of Diabetes or Not using Neural Network. (Dataset taken from Kaggle).
Modeled Neural Network with two Hidden Layers to predict whether women is a risk of Diabetes or Not with 80 percent (without feature Engineering) accuracy using keras. Link : https://github.com/Harisuryatm/Project-2_using_DeepNeuralNetwork Hand-written Digits classification (MNIST dataset) Trained a Stochastic Gradient Descent Classifier and compared it with Random Forest using ROC- AUC to predict Handwritten Digits with good accuracy. (Binary as well as Multiclass classification). Link : https://github.com/Harisuryatm/Support-Vector-Machine/blob/master/Voice%20classification%20(SVM).ipynb Housing price prediction. (Dataset taken from Kaggle). Modeled Housing price using Random Forest Regression that shows insights about how to handle missing values,categorical values in the features. Housing Price data set contains 1,460 homes, each with a few dozen features of types: float, integer, and categorical . Task is to predict Sales Price of the particular house for our taste. Link : https://github.com/Harisuryatm/Random-Forest-Algorithm/blob/master/Housing%20dataset%20(RandomForestRegressor).ipynb