Content Disclaimer
Copyright @2014.
All Rights Reserved.
StatsToDo : Classification by Naive Bayes Probability Program

Links : Home Index (Subjects) Contact StatsToDo

Related link :
Classification by Bayes Probability Explained Page
Classification by Basic Bayes Probability Program Page

Program Help & Hints
Data: Attributes ± Outcome Designation
Data Input for Analysis of reference Data
    The data is for a table with 2 columns
    Each row contains data from a case from the reference data
    Col 1 = Series of + (Positive) and - (Negative) for attributes
    Col 2 = Single character or word for Outcome Outcome Designation

Data Input for Interpretation
    The data is for a table with single columns
    Each row contains data from a case from the input data for interpretation
    Column with series of + (Positive) and - (Negative) for attributes

Program 1.

Reference Table of Counts

Array of of Sample Size for Each Outcome

Input Table of Counts for Analysis of reference Table
    This is an alternative input for analysis
    The table is a count of the reference data
    The number of columns is the number of outcomes
    The first row contains the outcome names
    Each following row represent an attribute
    Each cell is the count of the attribute (row) for that outcome (Col)
Program 2.

Probabilities P(+|o)
Table of Probabilities

Array of of A priori Probabilities

Array of Costs Coefficients

Input Table of Probability of attributes for each outcome P(+|o)
    This is a table of probabilities P(+|o)
    The number of columns is the number of outcomes
    The first row contains the outcome names
    Each following row represent an attribute
    Each cell is the Probability of Positives in the outcome P(+|o)
    To Calculate:
    - the attributes must be in the Data input textarea
    - The number of attributes in the data and in this table must be compatible

Array of Apriori Probabilities
    Probability for each outcome before attributes are known
    Single row, number of columns = number of outcomes
    Columns separated by spaces of tabs
    Values representing relative probabilities

Array of Costs
    Cost of wrongly missing a outcome
    Single row, number of columns = number of outcomes
    Columns separated by spaces of tabs
    Values representing relative costs

Program 3.

Program 4.