Coveo Platform 7.0 Language Support and Stemming Improvements

The Coveo Platform 7 language and stemming improvements provide more precise results as the index now knows exactly where any variant of any term occurs within indexed documents.

With the Coveo Platform 7:

• All terms are indexed in their original form including all common small words as well as terms containing accented characters and casing variants.

• Stemming is rather performed at query time by expanding search terms (see About Stemming).

• You can configure what specific or automatically detected languages are used to expand query terms by stemming (see Configuring the Culture of a Search Hub With the .NET Interface Editor).

• Stemming algorithms are language-specific, now supporting 17 languages (see Supported Languages).

You can now:

• Perform an exact match search for a phrase that contains only short common words.

  Example: Search for to be or not to be and find documents containing this famous phrase. This was previously impossible as the searched phrase contains essentially common words that were not indexed.

• Perform a true exact match search for a phrase.

• Include wildcard characters in phrase searches.

• Use the NEAR operator combining many terms.