Data science mini projects

This repo contains a list of data science related projects that I have completed, the topic covers the most common used techniques and concepts, the objective is aimed to help:

• explore/express precisely the insight from a real world dataset
• understand/choose the "right" model with the "right" metric in the "right" way
• practice scalable data with parallel processing engine such as Pyspark or Dask

Categories

• Data Visualization

  • AirBnb Hawaii EDA
  • AirBnb Reviews Analysis(Timeseries, Flask, CodePipeline, Beanstalk)

• Hypothesis Testing

  • AirBnb Superhost Earning Analysis

• Naive Bayes

  • 20newsgroups - Multinomial Naive Bayes classification from scratch
  • MNIST - Multivariate Gaussian classification from scratch

• Linear Regression

  • BostonHousing - Linear Regression(Pyspark)

• Logistic Regression

  • Click-thru-Rate-Prediction
  • Calibrated Estimation

• Gradient Descent

  • Coordinate Descent from scratch

• NLP

  • Word Embedding from scratch
  • Book Reviews Wordcount(Pyspark)
  • Trip Advisor Review - Text Classification(Pyspark)