Data Scientists and Statisticians

Overview


Data scientists are professionals who apply the tools of statistics and machine learning to data sets in order to extract knowledge from the dataset. Sometimes they are trying to uncover patterns or trends in the data, other times they are training a machine learning algorithm to forecast future data points.

The Modeling Corner contains resources for data scientists, including the following

  • Statistics
  • Machine Learning

User Tracks


The following examples demonstrate using the davinci platform to solve various data science problems.
Linear Regression
The linear regression is a standard workhorse of statisticians and data scientists.
start
Linear Discriminant
Demonstrates using a linear discriminant to recognize clusters in a dataset.
start
Clustering
Demonstrates using the k-means clustering algorithm to try to find clusters within a dataset.
start
Time Series
Time series analysis is a challenging area of probability and statistics. This example shows some basic techniques to analyze a simple time series.
start
Train a Machine Learning Algorithm for Digital Character Recognition
Uses tensor flow to create a neural network that is trained on a very simple representation of digit images.
start tensorflow start javascript
Neural Network from Scratch
This example shows a very basic method to construct a simple neural network using a linear algebra library and standard gradient descent.
start

Video Demo and Tutorial Playlist


  • Data
    • Introduction to Data
    • Exploring Data
  • Math
    • Math Notation (Latex) and Charting
  • Differential Equations
    • Crank Nicolson
  • Statistics
    • Statistics
    • Regression
      • Linear Regression
      • Logistic Regression
    • Time Series
        Stationarity, Unit Roots and Trends
  • Machine Learning
    • Linear Separable Patterns
    • Principal Components
    • Neural Networks and Deep Learning
      • Tensor Flow
      • Basic Neural Networks from Scratch - demonstrates creating a neural network using matrices and gradient descent.
      • Simple Character Recognition
      • Image Convolution
    • Clustering
    • Discriminant Analysis
  • Dynamic Visualizations - demonstrates ways to visualize data and relationships in the data