codesimian
Class Email

java.lang.Object
  codesimian.Email

public class Email
extends java.lang.Object

part of a college class project. mostly useless

Field Summary

java.lang.String	body Exactly as given to the constructor.
int	bodyStarts index in textIndex[] the email body starts.
protected NaturalLanguage	nl
double	spamFraction Range 0.0 (not at all like spam) to 1.0 (worst spam ever).
java.lang.String	subject Exactly as given to the constructor.
int[]	textIndex Each unique word has a number
int	uniqueWords how many unique words could be in the values of textIndex[], even if some of those values are not in textIndex[].
double	weight Relatively how important this email is to figuring out what divides spam and nonspam.

Constructor Summary

Email(java.lang.String subject, java.lang.String body, double spamFraction, NaturalLanguage nl)

Method Summary

static Email[]	getEmailsFromFile(java.lang.String fileName, NaturalLanguage useThisInEmailConstructors) The file must contain my own email input format: $$$startNewEmail $$$spam .87 $$$subject email subject here $$$text non-subject text here Put newlines wherever you want.
void	readAgain(NaturalLanguage nl)
java.lang.String	textIndexsToString()
java.lang.String	toString()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

spamFraction

public double spamFraction

Range 0.0 (not at all like spam) to 1.0 (worst spam ever).

subject

public java.lang.String subject

Exactly as given to the constructor.

body

public java.lang.String body

Exactly as given to the constructor.

textIndex

public int[] textIndex

Each unique word has a number

uniqueWords

public int uniqueWords

how many unique words could be in the values of textIndex[], even if some of those values are not in textIndex[]. Useful to scale a floating point number to the range of possible words.

nl

protected NaturalLanguage nl

bodyStarts

public int bodyStarts

index in textIndex[] the email body starts. Before that is the subject.

weight

public double weight

Relatively how important this email is to figuring out what divides spam and nonspam. Any positive value.

Constructor Detail

Email

public Email(java.lang.String subject,
             java.lang.String body,
             double spamFraction,
             NaturalLanguage nl)

Method Detail

readAgain

public void readAgain(NaturalLanguage nl)

textIndexsToString

public java.lang.String textIndexsToString()

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

getEmailsFromFile

public static Email[] getEmailsFromFile(java.lang.String fileName,
                                        NaturalLanguage useThisInEmailConstructors)

The file must contain my own email input format:
$$$startNewEmail
$$$spam .87
$$$subject email subject here
$$$text non-subject text here
Put newlines wherever you want. Spam ranges 0.0 to 1.0.