Hello ..!! My name is Manas Thakre. Thanks for visiting my website. I am a graduate student at University of Washington and enrolled into MS in Information Management program. Graduating in June-2018, I specialize in Data science, Business Intelligence and User Experience Design.
I have a professional work experience of ~ 9 years doing business data analytics, customer analytics/ insights, market research, strategic consulting and project management . I have worked for different companies like GE Healthcare, Novartis, Ericsson, Wockhardt and Frost and Sullivan. In 2016, I decided to join University of Washington to pursue MS in Information Management.
I am passionate about generating useful information and insights from the data, and enjoy leading the project teams at work.
I have experience of working with technologies and concepts like R, Python , Machine Learning, Classification algorithms, Big Datasets, SQL , IBM Watson, d3.js , Healthcare analytic tools (IMS Dataview, IMS Padds), IBM Cognos, Tableau, PowerBI and RapidMiner.
I have proficiency in using popular data analytics libraries from R and Python, like ggplot, dplyr, scikit learn, numpy, pandas, matplotlib, maxLik, scipy, seaborn and other modules.
The data science courses that I completed at UW include INFX 573: Data Science and Statistics Using R taught by Prof. Ott Toomet,
INFX 574: Machine Learning Algorithms and Econometrics Using Python taught by Prof. Jevin West
and INFX 575: Data Science: Scaling, Applications, Responsibility. taught by Prof. Bill Howe.
The creation of this website is, in fact, a part of my assignment for INFX 575 course :)
Please feel free to visit my LinkedIn and GitHub page, or reach me at manast@uw.edu
This is a tweet sentiment analysis engine based on Python, and written in Javascript.
Enter the twitter username of your interest to see what kind of sentiments their tweet displays.
Disclaimer: This engine is not built using any ntlk libraries and hence the code is entirely made up of custom written functions and built from scratch. I am using a json dictionary file which is static, to map the tweets to the words in this dictionary.
Click here to visit the GitHub repository for this project
I along with two fellow researchers, Dania Tanzil and Timoth Pace, worked on this 6 month long capstone project with T-Mobile.
Our capstone aimed at applying advanced machine learning techniques to identify advertisement traffic from the cellular network traffic.
The idea behind classifying advertising traffic is relevant in today’s business era where digital marketing has become the predominant medium to increase
outreach and connect directly with customers. Our team developed a novel approach to label & predict this advertisement traffic (accuracy ~97%),
thereby paving the way for further research into identifying and blocking malicious/spam advertisements.
The relevant activities we performed included getting data from network captures, data pre-processing, feature extraction, ad-labeling,
applying ML algorithms and evaluating model performance. To understand the top-level overview of project and our research experience please
Click here to read the news briefing of the Project.
This news briefing is published by University of Washington, Information School on their website (Dated: 22-May-2018).
This project includes making a local clean copy of database by doing ETL processes on remote database, and then using the data to answer business questions and understanding business of retailers. The final visualizations are made in Tableau.
To view the presentation .
Click here to visit the GitHub repository for this project
Written entirely in d3.js, this visualization attempts to see how the world performed in terms of 4 key health indicators. Nations have struggled through many decades, to bring down their Infant Mortality Rates (IMR) and to increase the chances of live births. With the help of Visualization, I am trying to explore if there are any associations/ patterns that emerge when we try to compare two different parameters. By including Healthcare GDP, I am also trying to see if the economic status of the country has any role on IMR.
Click here to visit the GitHub repository for this project
In this Python Script, I attempt to use a graphing "turtle" to plot some basic bar charts using the sample data from the files stored in the data folder of this repository. You can use other data of your interest to do the same by tweaking my code wherever needed. The functions written in the script guide the turtle movements to draw all the key elements of this chart. Enjoy Graphing :)
Click here to visit the GitHub repository for this project
This is a mini project aimed at just showcasing on how we can use SQL to find answers to some interesting questions from a dataset. I have implemented this project previously as well using R's dplyr, but here I use SQL to pull and sructure the information I need. Links to the tables used for this task are placed on the github repository for this project.
Click here to visit the GitHub repository for this project
This script is inspired by a real-world situation and is deliberately somewhat vague. The task is to understand the data (in NOAA's .nc format), convert it into a suitable dataframe, and find the tools that produce the desired output. In this script, I perform data wrangling to prepare precipitation and temperature maps for month of January and July (Year 2017).
The files can be found on the repository. Click here to visit the GitHub repository for this project
This project aims at designing and evaluating the concept of RELAY as a set of wrist bands, that offers minimal communication capabilities stressing more emphasis on user safety and feel good. This is not a sophisticated Apple Watch or Android wear.
It is a simple, affordable, useful and sympathetic way of silent communication between you and your loved ones who care for you and your safety.
It empowers you by emotionally connecting you to the people who care about at times when you are alone.
Please watch the animated video below that explains the concept behind Relay.
Click here to read my Medium Blogpost to get the complete details of this project.
