Closed

Simple Java or C++ Classifier (Naive Bayes) Program - repost

This project was awarded to snubmooseman for $55 USD.

Get free quotes for a project like this
Employer working
Awarded to:
Project Budget
$10 - $30 USD
Total Bids
2
Project Description

Hi, I need some programming done in either Java or C++. The program is a simple classifier, with real-world application.

2 versions of Naïve bayes classifier
- discretize continuous attributes by creating a number of equally sized bins and counting the number of samples (on a per-class basis) that fall into each bin. With these values, you can estimate conditional probabilities. the number of bins used a run-time parameter.

- assume a Gaussian distribution represents the class-conditional probability for continuous attributes. You can then calculate sample means and variances (again on a per-class basis) using the class-conditional probability equation

Bagging / ensemble classifier
bagging algorithm provided to construct 60 datasets from training sets (provided). Data points selected using a random uniform probability with replacement. Each constructed dataset will be used to train on classifier for the final ensemble of 60 classifiers.

Also create decision trees to classify data.

Note, code should not use any pre-built libraries.

Looking to make some money?

  • Set your budget and the timeframe
  • Outline your proposal
  • Get paid for your work

Hire Freelancers who also bid on this project

    • Forbes
    • The New York Times
    • Time
    • Wall Street Journal
    • Times Online