1. Sohom Ghosh
in.linkedin.com/in/sohomghosh
+91 8001734384
sohom1ghosh@gmail.com
github.com/sohomghosh
SUMMARY
– Data Science Enthusiast with relevant research,
industrial experience & Publications in International
Journals [ISSE (Springer, NASA Journal), IJARCST]
& Conferences [ICACNI (Springer), ICACCE (IEEE)]
– Completed courses on Big Data Analytics, Data
Mining [IIT-KGP]; Social and Economic Networks
[Stanford University (Coursera)]; Machine Learning
for Data Science and Analytics [Columbia Univer-
sity (edX)]; Artificial Intelligence
– Expertise in Text Analytics, Recommendation
System & Statistical Modeling
TECHNICAL SKILLS
TECHNOLOGIES R, Python, SQL, Hadoop,
Spark, Tableau, MS-Office
TECHNIQUES Regression, Random Forest,
SVM, GBM, Neural Net, Deep
Learning, Clustering etc.
WORK EXPERIENCE
APRIL 2016 – PRESENT
FN MathLogic Consulting Services Pvt. Ltd, Gurgoan
Analyst
• Project: Prediction of whether a customer will
re-buy an asset product; Classified the data using
Random Forest, Deep Neural Net, GBM in R (h2o)
• Data Visualization: Prepared dashboards & re-
ports using MS-Excel & Shiny R
• Text Analytics: Topic Modeling, Text Classifica-
tion [Done as part-time intern Apr - Jun’16]
• Capability Development: Machine Learning, Model
Assessment, Ensemble Learning, Deep Learning, Au-
tomation, Time Series, Optimization, Cloud (AWS)
SEPTEMBER 2015 – JANUARY 2016
Novel R & D India (P) Ltd., Kolkata
Big Data Faculty (Part-Time)
• Courses Taught: Big Data Analytics - Hadoop, R
DECEMBER 2013 – May 2016
Heritage Institute of Technology, Kolkata
Undergraduate Student Researcher
• Sentiment Analysis on Movie Reviews [IJARCST,
Vol 3, Issue 1, pp 41-46] (journal)
– Classified reviews by Lexicon, Machine Learn-
ing (SVM, Neural Net, Random Forest), Deep
Learning (word2vec) based approaches, En-
sembled them using Deep Neural Network
– Devised an algorithm to suggest words to re-
viewers by analyzing the title of reviews
• Recommendation System based on Product Pur-
chase Analysis
[ISSE, Springer London, ISSN:1614-5054, Vol 12, Is-
sue 3, pp 177-192] (NASA journal)
[ICACNI, SIST Springer, ISBN: 978-81-322-2538-6, Vol
43, pp 581-591] (conference)
– Analyzed various properties of Amazon Co-purchase
Network (Clustering Co-efficient, Degree Dis-
tributions, Popularity Trend etc.)
– Analyzed dynamic buying patterns & developed
algorithms to recommend products
• Solving Real Life Problems using Machine Learn-
ing Techniques
• Predicted Sale of Products in stores across
different cities (Used XGBoost, Deep Net)
• Predicted Customer Churn in a Telecom Net-
work (Used Random Forest, SVM, Neural Net)
• Extraction & Analysis of Publication Data of
Conferences [IEEE ICACCE-2015, pp 588-593]
• Analysis of Computer Science publications
[WIS & COLLNET 2015] (poster)
– Analyzed content of research papers to develop
a Recommendation System
– Examined the collaboration characteristics &
trends of research for 60 years
JUNE 2015 – JULY 2015
Indian Statistical Institute, Kolkata
Summer Research Intern
• Prediction of Cancellations of Taxi Reservations
Developed a predictive model for classifying new book-
ings as to whether they will eventually get canceled
due to unavailability of cabs; Used Naive Bayes’, SVM,
Neural Net, Random Forest in R, Weka
EDUCATION
2016 B. Tech (Computer Science & Engg.)
Heritage Institute of Technology, 8.22/10
2012 Senior Secondary Education (CBSE)
Sarvodaya Sr Secondary School, 80.80 %
2010 Secondary Education (ICSE)
St. Xavier’s School, 91.57 %
PERSONAL DETAILS
ADDRESS: K-36, opp. Presidium School, Sec-51,
Gurgoan - 122018, Haryana, India
HOBBIES: Learning from MOOCs, Solving Data
Science Challenges, Playing
Mouthorgan & Tabla