eClient 101 Customization and Integration

Chapter 1. Introducing Content Manager 9

Currently Information Integrator for Content provides two services:

򐂰 Information Mining Service

This service allows you to automatically analyze and organize documents on

content servers. Because nearly 80 percent of your business data is

unstructured, you cannot do this manually. Information Mining provides tools

such as automatic categorization, summarization and information extraction.

If the analysis results get stored together with the original document, you can,

for example, restrict searches to certain categories and display a summary for

each search result. You can find a description of the Information Mining

service features in 1.4, “Information Mining Service” on page 9.

򐂰 Workflow

You can use the workflow service to control the flow and performance of work

in your business. When users work with the results of federated searches,

they often must make decisions on what actions to perform. You can

determine in advance how you want users to perform the work. The actual

documents can reside on any of the supported content servers. You can

automate the workflow by setting up profiles and rules.

1.4 Information Mining Service

Industry studies show that employees, in general, spend 30 percent of their time

just looking for the information they need to do their jobs. What is more, most of

this information is unstructured and is buried within reports, e-mail, mail, or faxes.

One way to use these documents efficiently is to organize the information within

the documents and create metadata for documents. The metadata of a

document includes names, institutions, or places mentioned in a document or

category a document belongs to. Users can use the metadata to narrow their

searches of documents to certain topics or terms.

The Information Mining Service of Information Integrator for Content provides

text analysis components to extract metadata automatically, thus making mining

economically viable. The complete interpretation of only factual knowledge

stated in unrestricted natural language is still out of reach using current

technology. However, tools that apply pattern recognition techniques and

heuristics are capable of extracting valuable information from arbitrary free-text.

Extracting information ranges from identifying important terms, such as names,

institutions, or places mentioned in a document, to summarizing a document.

In the following section, we briefly describe the features, concepts and

architecture of the Information Mining Service. For a detailed description, go to

the Information Center and select Enterprise Information Portal ->

Administration -> Managing information mining.