Data Science Consultant and Research Mentor, PhD.

Skanda V.

Senior Data Scientist and PhD. I am passionate about teaching cutting edge data science, machine learning, and NLP concepts. I love to constantly learn and have obtained multiple data science, ML, and deep learning certifications from Coursera, as well as undergone the highly selective Data Incubator (Top 1-3% selected) fellowship program.

My goal is to enable YOU to understand core concepts and help solve your problems. Please reach out if you have questions in other related subjects. My interests are inter-disciplinary.

I'm passionate to document what I learn to positively impact other learners. I frequently publish blogs in top Medium publications like "Towards Data Science," with 6000+ monthly views (https://skanda-vivek.medium.com/) using real-world data sets such as social media, road traffic, and airline delays. Here are some recent popular articles:

https://towardsdatascience.com/how-400k-tweets-show-that-simone-biles-wins-ee941bdb13e2
https://towardsdatascience.com/clustering-in-geospatial-applications-which-model-should-you-use-59a039332c45

Highlights from my TDI fellowship program include:
1) Developed random forest machine learning model to predict the impacts of collisions on Manhattan traffic using 1.4 million collisions and 23 million speeds from NYC Open Data API SQL database

2) Competitively selected to present a live demo of the project at Pitch Night: https://nbviewer.jupyter.org/github/skandavivek/Manhattan-predicting-fragility/blob/master/Visualization.ipynb

3) Web scraping of New York socialite blog to extract, parse, and filter over 120,000 photo captions and model social connections among 100,000 people using graph and page rank algorithms

4)Developed deep and convolutional neural networks in Tensorflow and Keras to classify images from the CIFAR-10 dataset.

5) Analyzed more than 100 GB of stack exchange posts data, and used NLP methods such as Word2Vec in Spark as well as Spark ML to predict if questions have the 10 most common tags