Browse Prior Art Database

DPP Categorization engine

IP.com Disclosure Number: IPCOM000201615D
Publication Date: 2010-Nov-16
Document File: 2 page(s) / 31K

Publishing Venue

The IP.com Prior Art Database

Abstract

Rule-based classification/categorization based on the text messages' content

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 2

DPP Categorization engine

DPP Categorization engine is a rule-based categorization - algorithmical solution - based on the content of text messages. User defined rules are based on listing of words. Most beneficial difference in comparison with former existing solutions is possibility to use regular expressions and weights(importance) of the listed words. Above it from the user's perspective you can use logical junctions among these words.

USER INPUTS
users specify categories of tickets using words
between words logical junctions AND, OR and 'NOT'(Unwanted) are used word can be specified as regular expression using special characters

    * - asterisk (stands for 0-many characters)
? - question mark (stands for exactly 1 character)
every word has its 'weight' - this corresponds to the importance of the word from the user's perspective, e.g.

Important -> weight = 250

Might be -> weight = 25
Unwanted -> weight = -200

USER INPUTS - example
e.g. you want to define category related to Bakery there are these types of messages:

BAKE CRUNCHY ROLL

BAKE BREAD

PRODUCE BREAD

PRODUCTS MEATS

user's specifications:

Importance Words' description

Important BAKE,PRODUC* Important BREAD*,ROLL* Might be CR?NCHY Unwanted MEAT*

this can be rewritten like:
(Important(BAKE) OR Important(PRODUC*)) AND (Important(BREAD*) OR Important(ROLL*)) AND Mightbe(CR?NCHY) AND Unwanted(MEAT*)
the format how it will be stored in DB:

ID of Category Words' description Weight

935 BAKE,PRODUC* 250 935 BREAD*,ROLL* 250 935 CR?NCHY 25 935 MEAT* -200

CATEGORIZATION - schema
"Pricing" of categories -> Loading user inputs -> Categorization of messages

1


Page 02 of 2

Following connotations will use syntax forma...