Intro

🚀 Passionate Data Professional & Problem Solver

Welcome to my professional journey! I'm Rahul Nair, a dedicated data professional with a passion for unraveling complex challenges and turning them into actionable insights. 📊

Skills Snapshot:
📈 Analytical Mindset: Proficient in transforming raw data into meaningful stories, I thrive on extracting trends and patterns to drive strategic decision-making.
🤖 Machine Learning Maven: Experienced in developing and implementing machine learning models for marketing, risk assessment, and voter outreach, adding a touch of innovation to every project.
💻 Tech Savvy: Comfortable with a diverse tech stack, from SQL, Python, and Tableau to cloud platforms like AWS. I believe in leveraging technology to unlock new possibilities.
🌐 Cross-Functional Collaboration: A team player at heart, I've successfully collaborated with diverse teams, including data scientists, product managers, and IT professionals, fostering a collaborative environment.

About Me:
I am someone who thrives in dynamic environments, embracing challenges as opportunities to learn and grow. My journey is marked by a relentless pursuit of excellence, reflected in my commitment to delivering impactful solutions. I take pride in my ability to communicate complex findings in a clear and compelling manner, ensuring that insights drive action.

Beyond the Data:
Outside the realm of data analytics, I am an avid learner, always seeking to broaden my horizons. Whether it's volunteering for impactful projects or spearheading initiatives, I find fulfillment in contributing to positive change. Please do check out my Projects and blogs.

Work

Restaurant Recommendation using Yelp Dataset

This was my academic project for CSP-571 "Data Preparation And Analysis". In this project I built a personalized recommender web app using Yelp dataset of restaurants. Various models were tested like Pure Collaborative, Approximate Nearest Neighbour, K-NN, Naive Bayes and Hybrid Maxtrix Factorization on various hyperparameters which were tuned using the library "scikit optimizer". AUC was chosen as an evaluation metric for the models which is a decision-support metric that checks whether customers like the item or not. In this case, figuring out customer preference in general is more important and practical. And for deployment, I used Angular8 and Flask frameworks.

Click the link to view the project.

StackOverflow Data Analysis

This was my academic project for CSP-554 "Big Data Technologies". In this project I developed a tag predictor using pyspark which predicts tag based on the post. NLP techniques were implemented to achieve that. Apart from that, certain analysis were also performed on the StackOverflow Data using HIVE and PIG.

Click the link to view the project.

Facial Expression Identifier

I did this project during my UG. In this, it detects facial expression by capturing your image by webcam and then predicting the emotion, i.e, happy, angry, surprised or sad. It uses inception v3 model for image classification and haar-cascade for face detection.

Click the link to view the project.

Plant Disease Detector

I did this project during my Internship back in my UG. It's a web-based API which detects the disease the plant has whose image is being put as input. It is implemented in Python, using inception v3 model for training the dataset. The web interface was devleoped using Flask and Django

Click the link to view the project.

US accidents data analysis

This is my pet project. In this, I built a web-app with a basic layout to predict the severity of an accident. Various machine learning models such as Logistic Regression, Decision Tree Classifier, AdaBoost Classifier, etc. were tested and out of that XGBoost Classifier came out to be the best one. Hyperparameter tuning of these models were done using Bayesian Optimization technique. I tested the models using F1_Beta score which is a weighted harmonic mean of precision and recall, reaching its optimal value at 1 and its worst value at 0. In this, the beta parameter determines the weight of recall in the combined score. beta < 1 lends more weight to precision, while beta > 1 favors recall (beta -> 0 considers only precision, beta -> +inf only recall). In this case, I was more interested in knowing "how many relavant items are selected" rather than "how many selected items are relevant". Since this is predicting severity if an accident, it paramount that all the relevant cases are covered even with a few false positives. As for deployment, we used HTML and Flask frameworks.

Click the link to view the project.

Statistical Analysis of Life Expectancy

There have been studies on factors that affect life expectancy like gender, ethnicity, etc., but there hasn’t been much study done on factors like immunization, which can matter a lot as there are countries where the awareness for immunization is less which may lead to lower life expectancy. Other factors can also play a role like the GDP of that country, etc. In this project, I did a study that puts focus on factors like immunization, social, economic and other health related factors and see if there is any correlation among these factors and predict what can be the life expectancy of an individual who is from a particular country with all the above mentioned factors provided.

Click the link to view the project.

Clustering and Regression Analysis of Gerrymandering

In this project, I applied clustering technique (weighted k-means) to see if I can come up with a better districting plan, and regression analysis to see which factors play an important role in determining voters favoring a particular party.

Click the link to view the project.




Dashboards

UFC Analysis

This dashboard is currently in progress. This basically shows few distribution of certain attributes which if closely analyzed might be factor for increasing the chances of winning. There are are more analysis to be added.

Click the link to view the story.

Explaining Data Science to an 11-year old kid

Recently, I had an assignment of explaining my major to an 11-year old kid (based on FLAME Challenge). This is my take on explaining Data Science.

Impact of Gerrymandering on US elections

This is part 1 of Gerrymandering series in which we will discuss what gerrymandering is, how it impacts elections, an exploratory data analysis on PA and the way computation can be used to tackle this issue.

Extensive Analysis of the Pennsylvania districts

This is part 2 of Gerrymandering series in which , we will do more analysis and focus on PA state.

Clustering technique for Redistricting

This is the final part of the Gerrymandering series in which I will discuss a clustering technique to generate a district plan.

Elements

Text

This is bold and this is strong. This is italic and this is emphasized. This is superscript text and this is subscript text. This is underlined and this is code: for (;;) { ... }. Finally, this is a link.


Heading Level 2

Heading Level 3

Heading Level 4

Heading Level 5
Heading Level 6

Blockquote

Fringilla nisl. Donec accumsan interdum nisi, quis tincidunt felis sagittis eget tempus euismod. Vestibulum ante ipsum primis in faucibus vestibulum. Blandit adipiscing eu felis iaculis volutpat ac adipiscing accumsan faucibus. Vestibulum ante ipsum primis in faucibus lorem ipsum dolor sit amet nullam adipiscing eu felis.

Preformatted

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';

Lists

Unordered

  • Dolor pulvinar etiam.
  • Sagittis adipiscing.
  • Felis enim feugiat.

Alternate

  • Dolor pulvinar etiam.
  • Sagittis adipiscing.
  • Felis enim feugiat.

Ordered

  1. Dolor pulvinar etiam.
  2. Etiam vel felis viverra.
  3. Felis enim feugiat.
  4. Dolor pulvinar etiam.
  5. Etiam vel felis lorem.
  6. Felis enim et feugiat.

Icons

Actions

Table

Default

Name Description Price
Item One Ante turpis integer aliquet porttitor. 29.99
Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
Item Three Morbi faucibus arcu accumsan lorem. 29.99
Item Four Vitae integer tempus condimentum. 19.99
Item Five Ante turpis integer aliquet porttitor. 29.99
100.00

Alternate

Name Description Price
Item One Ante turpis integer aliquet porttitor. 29.99
Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
Item Three Morbi faucibus arcu accumsan lorem. 29.99
Item Four Vitae integer tempus condimentum. 19.99
Item Five Ante turpis integer aliquet porttitor. 29.99
100.00

Buttons

  • Disabled
  • Disabled