Kunal Tamang

Email: adtwnn@r.postjobfree.com

LinkedIn: https://www.linkedin.com/in/kunal-tamang-57b684120/ Contact: 786-***-****

I am a PhD candidate at University of Miami with 5+ years of experience working on social network analysis. My research work is specially focused on modeling citation network and identifying potential breakthrough publications at their early stage. I am a self-driven, self- motivated personality who enjoys both teamwork and self-initiative, as well as being flexible to new and challenging situations and have a strong desire for personal growth. As a proactive and fast learning data lover, I am looking for an opportunity to work in a dynamic environment, utilizing logical & analytical skills in a proficient way to help the corporate achieve business goals.

Modeling Network Evolution for Scientific Innovations

• Research on dynamics of common social networks like citation network.

• Defining the factors responsible for the papers towards large impact.

• Evaluating different citation-based measures to show the dependency of impactful papers on its reference papers and the timing of its citation.

• We define new metric potential showing that papers impact solely depends on potential rather than other factors.

• Based on our metric potentiality, we propose a model giving analytical solution whose theoretical results can capture network quantities in real world. Early-Stage Identification of potential breakthrough papers

• Identifying potential Nobel prize papers at an early stage.

• Describing newly defined metric as a parameter to define the phenomenon of Nobel papers using American Physical Society and Web of Science Database.

• Using ML Algorithms like XGBoost, Support Vector Machine, Random Forest and Neural networks based on features defined in our research to classify Nobel papers from regular papers.

• Using Metric like ROC curve, precison and recall measuring the performance of model. ECG beats Classification using Neural Network and other ML Algorithms

(Work submitted as Master thesis project for LMJU online program)

• Using ML algorithms like KNN, Random Forest, and XGBoost as well as neural networks like CNN and MLPNN in Arrhythmia classification.

• Uses open-source MIT-BIH Arrhythmia dataset containing 48 records from 47 patients hosted by PhysioNet.

• Comparing all the classification algorithms using performance metrics like Accuracy, precision, F1- score, ROC curve and AUC.

• To describe the best performing models and summarize the future scope and limitations of the work.

EDUCATION:

• Ph.D. in Physics, University of Miami, FL, USA (Expected in 05/2023)

• M.S. in Physics, University of Miami, FL, USA (05/2019)

• M.S. in Data Science, Liverpool John Moore’s University, England (07/2022) CERTIFICATIONS:

• Introduction to Data Science in Python (Coursera)

• Applied Plotting, Charting and Data Representation in Python (Coursera)

• Neural Networks and Deep Learning (Coursera)

• Post Graduate Diploma in Data Science (IITB, Bangalore, India) EXPERIENCE:

v Graduate Research Assistant at Department of Physics, University of Miami (Aug. 2017-present).

§ Simulation in Python to understand the behavior of Citation Network.

§ Modeling the citation network to understand the potential of scientific papers to get future citations as well as to be considered as breakthrough research.

§ Using different ML Algorithms to predict the breakthrough papers based on citation network of APS and Web of Science Database.

§ Using different performance metrics to evaluate the performance of ML models. v Graduate Teaching Assistant at Department of Physics, University of Miami (Aug. 2017-present)

v Lecturer of Physics and Mathematics at Secondary and Higher Secondary Level School (Chitwan, Nepal)

TECHNICAL SKILLS:

SKILL SETS DESCRIPTION

Programming language Python, R

Database SQL, MySQL, MongoDB

Programming language Python, R

Database SQL, MySQL, MongoDB

Operating System Windows, MacOS, Linux

Visualization Tools/package Tableau, Matplotlib, plotly, Seaborn, powerBI Machine Learning Scikit-learn, TensorFlow, Keras

Other Software/Tools Git, GitHub, Machine learning, Deep learning, Network Science

Contact this candidate