/MSDS7333

Quantifying The World

Primary LanguageJupyter Notebook

MSDS7333

Quantifying The World

This is our last MSDS course at SMU. Team consist of four members to work together on case studies throughout the semester.

Team Members:

  1. Sachin Chavan
  2. Tazeb Abera
  3. Gautam Kapila
  4. Sandesh Ojha

1) Case Study 01

This case study is based on Real time location system. The idea is to learn to build Indoor positioning system using dataset provided which contains measurements of signal strength at 166 different position from 6 access points and the main idea is to build IPS to determine location of the object using signal strength.

Techology : R, Latex

Reference : Deborah Nolan: Duncan Temple Lang. Data Science in R.Chapman and Hall/CRC, 2015

2) Case Study 02

This case study is about analyzing relationship between age and physcial performance of road racers.

Techology : R, Latex

Reference : Deborah Nolan: Duncan Temple Lang. Data Science in R.Chapman and Hall/CRC, 2015

3) Case Study 03

This case study is about evaluation of tree based model's ability to classify spams or no-spams emails.

Techology : R, Latex

Reference : Deborah Nolan: Duncan Temple Lang. Data Science in R.Chapman and Hall/CRC, 2015

Python begins

4) Case Study 04

This case study is building ARIMA model using python. Use historical flu data and provide four week farcast.

Techology : Python, Latex

Reference : Online

5) Case Study 05

This case study is all about studying impact of missing values. Building and comparing models with and without impuatiations for three common missing types.

MCAR - Missing Completely at Random

MAR - Missing at Random

MNAR - Missing Not at Random

Techology : Python, Latex

Reference : Online

6) Case Study 06

This case study is about evaluating deep neural network for classification problem and higgs boson dataset shall be used for signal vs background classification problem using TensorFlow.

Techology : Python, Latex

Reference : Online

7) Case Study 07

Perform analysis and build a model on the unknown dataset. This is binary classification models with 50 unknown features. Analyze and find important features and build model and calculate and compare cost of each model.

Techology : Python, Latex

Reference : Online