Text Mining

Text Mining refers to the automatic extraction of data from different archived resources for the purpose of discovering new or previously unknown information in unstructured textual data. This extracted information is linked together to create new facts or to make logical associations that can be investigated in depth through the use of sophisticated technology, such as ZyLAB’s visualization, discovery and disclosure tools.

Innovative visualization tools help expedite the searching and finding of key information. More specifically, visualization enables relevant information and its interrelationships to be seen at the same time, in a given context. Advanced visualization tools enable users to analyze large data sets as a whole and can provide a threefold increase in the speed at which specific information in these data sets can be found.

Visualization does, however, require structure, but structure is not always available within complex data sets. Manual structuring is time consuming and expensive, so users need efficient structuring tools. For this reason, ZyLAB developed several manual structuring tools that have been integrated into standard ZyIMAGE products. The ZyIMAGE Text Mining tools provide (semi)-automatic structuring tools that further advance the process of giving structure to unstructured data.

Types of text mining

ZyLAB provides two levels of text mining functionality:

ZyLAB Standard Text Mining Suite (a standard component of the ZyIMAGE XML Wrapper module) provides users the following advantages:

• File property extraction
• Document property extraction
• Concept extraction
• Automatic language recognition
• Hashing capabilities for unique document identification (based upon SHA1)

ZyLAB Professional Text Mining Suite builds further upon the Standard Text Mining functionality to offer advanced features such as:

• Entity extraction
• Fact extraction
• Summarization
• Document categorization
• Automatic taxonomy generation (ATG)

The ZyIMAGE Professional Text Mining Suite provides users with an advanced set of text analysis tools that enable them to perform entity extraction, fact finding, summarization, automatic taxonomy generation and document categorization. This ZyIMAGE tool suite is unique because it also supports 31 languages, a very large set of linguistic capabilities for technology of this level. (See the product specification for the up-to-date language set.)

If you are interested in our solution or you have a question, please fill out the information request form, send us an email or call your local office.