afs_doc_acc - AFS - Reference Guides

AFS Filters Description

Product
AFS
AFS_Version
7.12
Category
Reference Guides
language
English

The afs_doc_acc filter computes for each document of the corpus its proximity with every other documents. This computation generates data which are called ACC (Automatic Cross Content). The proximity is a number between 0 and 1. Two documents with a proximity close to 0 are totally different, and two documents with a proximity close to 1 are semantically very similar.

The filter is declared with the afs_doc_acc type. It is in the antidot-paf-misc package. It is a processor filter.

This filter will only work if instanced after the afs_doc_index filter.

This filter can be instantiated only once at any given moment. It will not read the "instances" parameter in the configuration.

The Document ACC filter specifications are described in the following table:

Parameter name

Mandatory

Type

Default

Description

nb

No

integer

100

Number of closest documents stored

input_layer

No

layer

MINING

The input layer containing the text to analyse. By default, will take all the text indexed by a preceding afs_doc_index filter.

Attention: This filter only works when instanced after an afs_doc_index filter.