Business Insight from Collection of Unstructured Formatted Documents with IBM Content Harvester

Today’s knowledge workers need to access, apply and reuse
content created by office productivity suite such as word
processor, spreadsheet and presentation. While the productivity
suite revolution in the 90’s freed individuals’ creativity
to generate content, it is increasingly difficult to effectively
manage and harvest individuals’ creation into knowledge
of the whole. One cannot glean into a project document
repository, for example, to get a summary of the status
of work items. Both keyword search and social tagging fall
short of functionality required to harvest and distill content
for reuse.

In this paper, we report the development and experiments
of IBM Content Harvester (CH), a tool to analyze and recover
templates and content from word processor created
text documents. CH is part of a bigger effort to collect
and reuse material generated in business service engagements.
Specifically, it works on unstructured formatted
documents and works by extracting content, cleansing off
sensitive information, tagging it based on user-defined or
domain-defined labels, and making it available for publishing
in any open format and flexible querying. As a result,
one can search for specific information based on tags, aggregate
information regardless of document source or formatting
peculiarities and publish the content in any format
or template. CH has been applied to a broad variety of
document collections containing hundreds of documents,
including live engagements, to promising effect.