Content mining

In the face of so much data, how do you find and extract the pertinent data that you require? Today, search engines and portals cannot completely meet this demand.

To improve distribution channel visibility, all editors, on-line shops and enterprises must improve their content indexation. In addition they must also develop new search methods which can be used on all types of multimedia content. One of these applications concerns custom publishing (mashup).

Techniques to enhance your data

Thanks to our expertise in using, structuring and enhancing content (content mining), Jouve offers a fully comprehensive service.

Based on syntactic, semantic and linguistic analysis technologies, these solutions allow extraction and indexing within the text of:

• Entity names
• Descriptive key words from content
• Specific semantic relations

They also allow:

• The categorisation or filing of articles or documents according to a filing plan
• The extraction of terms to enhance your terminologies

Jouve’s industrial approach can guarantee the quality of the extracted information. By combining several text mining engines and countless terminological resources, the Jouve system only makes an automatic choice when there is no risk of a mistake. In case of doubt, an alert is generated and a manual check is performed.

Jouve’s system has validation interfaces that can be used by Jouve personnel (in Europe or offshore), or can be used by our customers. Some decisions may require expert knowledge.

Jouve can create specialist teams (medical doctors, for example) or integrate our customer’s experts in the process.

High added-value services to win new customers

The enhancement of your metadata optimises the visibility for search engines and allows you to propose innovative services, such as:

• “Pull” services:
  • Searches by theme or concept
  • Keyword searches that do not necessarily exist in the text
  • Complex searches by combining generic criteria (geographical locations, names, dates or periods …) or technical (illness and/or drug names, chemical formulae …)
  • Hyperlink navigation in documents, collections or different databases
  • Searching by category
• “Push” services:
  • Subscription to themes with alert mechanisms
  • Proposition of similar contents (articles, documents …)

By capitalising on these linguistic and semantic technologies, and by integrating this with your traditional creation processes and content enhancing (conversion, pre-press), Jouve opens new business opportunities by using your content differently.

Known in all the domains, this unique expertise has also been used in Quaero. The goal of this European project is to develop a new industrial digitisation platform for the conversion and annotation of documents.

Jouve offers libraries, editors, Internet users, patent offices and industrials to enhance their content by giving their data a new dimension.