NaiveBayes (sigma-component 2.10 API)

java.lang.Object
- com.articulate.sigma.nlp.NaiveBayes

```
public class NaiveBayes
extends Object
```

Field Summary

Fields
Modifier and Type	Field and Description
`HashMap<String,HashMap<String,HashMap<String,Integer>>>`	`conditionalCounts`
`HashMap<String,HashMap<String,HashMap<String,Float>>>`	`conds`
`ArrayList<ArrayList<String>>`	`input`
`ArrayList<String>`	`labels`
`HashMap<String,HashMap<String,Float>>`	`means`
`HashMap<String,HashMap<String,ArrayList<Float>>>`	`numericValues`
`HashMap<String,Integer>`	`priorCounts`
`HashMap<String,Float>`	`priors`
`HashMap<String,HashMap<String,Float>>`	`ssd`
`HashMap<String,HashMap<String,Float>>`	`totals`
`ArrayList<String>`	`types`

Constructor Summary

Constructors
Constructor and Description

NaiveBayes(String filename)

Constructors
Constructor and Description
`NaiveBayes(String filename)`

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`void`	`calcConditionals(int sum)` Compute conditionals in the format of a class name key then the probalbities of the values for each "column" which is a numerical key.
`void`	`calcPriors(int sum)` Calculate the prior probabilities of each class given the numbers of instances of each class.
`String`	`classify(List<String> values)` Given the conditional and prior probabilities, and a particular instance set of attributes, compute which class that instance is mostly likely to fall into.
`void`	`createConditionalCounts()` Count the number of occurrences of each class.
`void`	`createMeans()`
`void`	`createPriorCounts()` Count the number of occurrences of each class.
`void`	`createStandardDeviation()` Note that this computes the sample standard deviation sigma = sqrt( (1/(N-1)) sum(1,n,(xi-meanx)*(xi-meanx)))
`void`	`createTotals()` Create totals per class of each variable that is continuous.
`int`	`findTrainingSetSize()`
`void`	`initialize()`
`static void`	`main(String[] args)`
`static float`	`probDensFunc(float mean, float ssd, float x)` Compute P(x\|y) given the mean, sample standard deviation and x
`float`	`probDensFunc(String clss, String label, float x)` Compute P(x\|y) given the mean, sample standard deviation and x

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - input
```
public ArrayList<ArrayList<String>> input
```
  - labels
```
public ArrayList<String> labels
```
  - types
```
public ArrayList<String> types
```
  - means
```
public HashMap<String,HashMap<String,Float>> means
```
  - ssd
```
public HashMap<String,HashMap<String,Float>> ssd
```
  - totals
```
public HashMap<String,HashMap<String,Float>> totals
```
  - numericValues
```
public HashMap<String,HashMap<String,ArrayList<Float>>> numericValues
```
  - priorCounts
```
public HashMap<String,Integer> priorCounts
```
  - conditionalCounts
```
public HashMap<String,HashMap<String,HashMap<String,Integer>>> conditionalCounts
```
  - priors
```
public HashMap<String,Float> priors
```
  - conds
```
public HashMap<String,HashMap<String,HashMap<String,Float>>> conds
```
- Constructor Detail
  - NaiveBayes
```
public NaiveBayes(String filename)
```
- Method Detail
  - probDensFunc
```
public static float probDensFunc(float mean,
                                 float ssd,
                                 float x)
```
    Compute P(x|y) given the mean, sample standard deviation and x
  - probDensFunc
```
public float probDensFunc(String clss,
                          String label,
                          float x)
```
    Compute P(x|y) given the mean, sample standard deviation and x
  - createPriorCounts
```
public void createPriorCounts()
```
    Count the number of occurrences of each class.
  - createConditionalCounts
```
public void createConditionalCounts()
```
    Count the number of occurrences of each class. The class name must be the last element of each row of the input data. {i100={interest={appearance=2, health=1, ...}, ...}, i500={interest={appearance=3, health=4,...}, ...} }
  - createTotals
```
public void createTotals()
```
    Create totals per class of each variable that is continuous.
  - createMeans
```
public void createMeans()
```
  - createStandardDeviation
```
public void createStandardDeviation()
```
    Note that this computes the sample standard deviation sigma = sqrt( (1/(N-1)) sum(1,n,(xi-meanx)*(xi-meanx)))
  - findTrainingSetSize
```
public int findTrainingSetSize()
```
  - calcPriors
```
public void calcPriors(int sum)
```
    Calculate the prior probabilities of each class given the numbers of instances of each class.
  - calcConditionals
```
public void calcConditionals(int sum)
```
    Compute conditionals in the format of a class name key then the probalbities of the values for each "column" which is a numerical key. So below, the probability of getting the value "appearance" as the value of variable 1 for class {'i500': {1: {'appearance':0.33,'health':0.44}, 2: {...
  - classify
```
public String classify(List<String> values)
```
    Given the conditional and prior probabilities, and a particular instance set of attributes, compute which class that instance is mostly likely to fall into.
  - initialize
```
public void initialize()
```
  - main
```
public static void main(String[] args)
```

Class NaiveBayes

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

input

labels

types

means

ssd

totals

numericValues

priorCounts

conditionalCounts

priors

conds

Constructor Detail

NaiveBayes

Method Detail

probDensFunc

probDensFunc

createPriorCounts

createConditionalCounts

createTotals

createMeans

createStandardDeviation

findTrainingSetSize

calcPriors

calcConditionals

classify

initialize

main