SUNEETH KUNCHE

732-***-****

** ********* **, ****** ********, NJ 08831 ad1s2s@r.postjobfree.com

SUMMARY

Data Science graduate student. Possess advanced coding & programming skills in Machine learning and python. Internship at Decovita ceramics as data analyst. Internship is based on Predicting market value of product and demand of product in Vietnam using Machine models like k-means, logistic regression. EDUCATION

Rutgers University, New Brunswick, NJ

Master of Science in Statistics-Data Science Sep. 2023-May .2025 National Institute of Technology Agartala

8.29/10 Aug. 2019 – May. 2023

RESEARCH PROJECTS

Machine Learning: Credit Card Fraud Detection 2022 Supervised machine learning use logistic regression model to predict credit card is fraud or legit. I used Logistic regression algorithm from sklearn library. I understand that the data is over fitted. I reduce the data to minimize overfitting because to train the model the data should have equal number of fraud credit cards when compared to legit credit cards. By using accuracy score to predict the cost function. To conclude, accuracy rate of prediction is 90%. Data analytics: comparison between single and double pilot injection of diesel engine in python 2022 The project use pandas, matplotlib, numpy libraries for comparative analysis. I collected data of single and double pilot injection. I import the data by using pandas and data manipulating. By using matplotlib implementing line graph and bar graph to represent engine pressure, crank angle and heat release rate. To achieve LTC (low temperature combustion rate) and suggest which pilot injector can attain LTC (low temperature combustion rate). The project includes Data manipulation and numpy for scaling the data into certain range. To conclude, Double pilot injector can attain LTC (low temperature combustion rate) when compared matplotlib.pyplot graphs with Single pilot injector. K-Means: Customer Segmentation based on annual salary and spending in mall 2023 The project use pandas, matplotlib, numpy, seaborn, sklearn.cluster to import kmeans. Taking a group of customers data, predicting number of k (centroid) need for group data by using Elbow method. The k value is 5 through Elbow method. Finally, clustering data: spending score vs annual income. The scatter plot represents 5 groups: 1 group represent of people who has low annual income and spend less, 2 group represent of people who has low annual income and spend more in mall, 3 group represent of people who has average annual income and average amount of spending in mall, 4 group represent of people who has high annual income and high spending in mall, 5 group represent of people who has high annual income and low spending in mall. Conclusion from project is that we can offer discount to 5 group to range there spending in mall compared with their Annual income TECHNICAL AND OTHER SKILLS

Core Domain Expertise: Data science and statistics Computing and Programming: Python, Pandas, Numpy, Matplotlib, sklearn, Tensorflow, Machine learning, Deep learning, Data analytics: Data cleaning and manipulating, Data visualization, R. Communication: English, Hindi and Telugu

CERTIFICATES

Machine Learning ((Organized by Stanford University) 2022

Supervised Machine Learning (Regression and Classification), Advanced Learning Algorithms, Unsupervised Learning, Recommenders, and Reinforcement Learning. Certified by Coursera Neural Networks and Deep Learning (Organized by Stanford University) 2023

Understanding neural network with tensor flow

Improving Deep Neural Networks ((Organized by Stanford University) 2023

Understanding hyper parameters to tuning models