Rules-Based Approach: Difference between revisions
Dgreenwood (talk | contribs) No edit summary |
Dgreenwood (talk | contribs) No edit summary |
||
| Line 1: | Line 1: | ||
This approach uses [[Data Extractor]]s to find key words, phrases, or other text-based information in order to identify and classify a document (assigning a '''[[Document Type]]''' to a document). For example, a document with a centered header of "Purchase Report" might be classified as a "Purchase Report" '''Document Type''' with this approach. One could build a [[Data Type]] extractor using regular expression to match the phrase "Purchase Report" centered at the top of a document to identify it. | This approach uses [[Data Extractor]]s to find key words, phrases, or other text-based information in order to identify and classify a document (assigning a '''[[Document Type]]''' to a document). For example, a document with a centered header of "Purchase Report" might be classified as a "Purchase Report" '''Document Type''' with this approach. One could build a [[Data Type]] extractor using regular expression to match the phrase "Purchase Report" centered at the top of a document to identify it. | ||
The "rules" are set using the '''''Positive Extractor''''' and '''''Negative Extractor''''' properties of a '''Document Type''' object in a '''[[Content Model]]'''. If set as the '''''Positive Extractor''''' and the extractor returned a result on a document, it would be classified as a "Purchase Report" '''Document Type'''. The '''''Negative Extractor'''''' works the opposite way. If the extractor finds a result on a document, it would be ''prevented'' from being classified as that '''Document Type'''. | |||
Revision as of 15:14, 6 October 2020
This approach uses Data Extractors to find key words, phrases, or other text-based information in order to identify and classify a document (assigning a Document Type to a document). For example, a document with a centered header of "Purchase Report" might be classified as a "Purchase Report" Document Type with this approach. One could build a Data Type extractor using regular expression to match the phrase "Purchase Report" centered at the top of a document to identify it.
The "rules" are set using the Positive Extractor and Negative Extractor properties of a Document Type object in a Content Model. If set as the Positive Extractor and the extractor returned a result on a document, it would be classified as a "Purchase Report" Document Type. The Negative Extractor' works the opposite way. If the extractor finds a result on a document, it would be prevented from being classified as that Document Type.