Articles
Uncover Document Processing Savings With Intelligent Classification
January 19, 2010
Click Here To Download:
•Uncover Document Processing Savings With Intelligent Document Classification
The document classification technology available varies and the applications can include anything from image processing techniques to latent semantic indexing (a technique in natural language processing that analyzes the relationships between a set of documents and the terms they contain). How your organization can use – and implement – this technology will be a key factor in your decision to buy certain programs. "Document classification software setup can vary from free-flow, drag-and-drop environments to more comprehensive ‘black box' solutions where you pre-instruct the software on how to classify documents," says Sam Schrage, president of AnyDoc Software. "There are also differences regarding the setup and ‘training' of the software. With some systems, this is done completely in advance, and adding new documents requires programming. However, there are a few systems where you can do both methods: the initial setup via online modules and training on the fly. This means, as new documents come into an enterprise, they can be identified in the daily workflow and the classification system is ‘trained' on them. With on-the-fly training, you don't have to move the documents out of the business process, have someone configure the classification system for that type of document, and then send it back through." Those systems requiring up-front setup might require specialized skills and extensive labor to create descriptions of specific document layouts – if there are hundreds or thousands of document layouts in circulation or numerous new layouts that appear on a recurrent basis, this method could be impractical. Click Here To Download:
•Uncover Document Processing Savings With Intelligent Document Classification
•Uncover Document Processing Savings With Intelligent Document Classification
The document classification technology available varies and the applications can include anything from image processing techniques to latent semantic indexing (a technique in natural language processing that analyzes the relationships between a set of documents and the terms they contain). How your organization can use – and implement – this technology will be a key factor in your decision to buy certain programs. "Document classification software setup can vary from free-flow, drag-and-drop environments to more comprehensive ‘black box' solutions where you pre-instruct the software on how to classify documents," says Sam Schrage, president of AnyDoc Software. "There are also differences regarding the setup and ‘training' of the software. With some systems, this is done completely in advance, and adding new documents requires programming. However, there are a few systems where you can do both methods: the initial setup via online modules and training on the fly. This means, as new documents come into an enterprise, they can be identified in the daily workflow and the classification system is ‘trained' on them. With on-the-fly training, you don't have to move the documents out of the business process, have someone configure the classification system for that type of document, and then send it back through." Those systems requiring up-front setup might require specialized skills and extensive labor to create descriptions of specific document layouts – if there are hundreds or thousands of document layouts in circulation or numerous new layouts that appear on a recurrent basis, this method could be impractical. Click Here To Download:
•Uncover Document Processing Savings With Intelligent Document Classification
