|

InfoPro Home > SmartIndexing Technology™ > Extracted Proper Name Indexing
LexisNexis Extracted Proper Name Indexing
In addition to the tens of thousands of controlled terms for companies and people, LexisNexis offers access to millions of additional proper names discussed in the news.
Advanced linguistic and semantic processes recognize, extract and normalize proper names in news documents and add them to our indexing segments with appropriate relevance scores. Extracted proper names will be matched, whenever possible, to an appropriate controlled vocabulary term in the authority list.
Customers can search the data using the extracted terms just as they use our existing controlled vocabulary.
Frequently Asked Questions
How do I access the extracted indexing in my searches?
Use the same segments (COMPANY, PERSON or TERMS) and scoring you use for searching controlled Proper Name Indexing on LexisNexis.
Because the extracted names are not always standardized, start broadly, either with no score or with a proximity connector.
terms( name )
terms( breathasure )
company( name PRE/2 9*%)
company( breathasure PRE/2 9*%)
Why do I see variations on proper names?
The extracted indexing process does not require an authority list of names so any name is eligible for indexing.
Through normalization processes, the best form of a proper name (usually the longest form) as it appears in the document is added to the indexing segment. Extracted proper names will be matched, whenever possible, to an appropriate controlled vocabulary term in the authority list. If no controlled vocabulary term matches, the document will tag with the extracted term.
You may see variations in the extracted proper names because the extractor tool indexes the best form of the term found in an individual document. If two documents use two different versions of a company name (that is not on the authority list), extractor will index the best form found in each individual document.
When should I use a controlled term from the authority list?
For the companies and people we index through our authority list, the controlled term will always provide the most precise answer set. By definition, a controlled term normalizes variants into a standard applied across all indexed documents.
Extracted terms not mapped to terms in the authority list offer increased access to emerging companies and new players in an industry.
Some of the extracted indexing terms do not reflect the full name of the company. Why is that?
The extracted indexing technology does not require an authority list of established names. Rather, it pulls names from within the document itself. Authors will often refer to the same entity in a variety of ways, so extracted terms that cannot be accurately mapped to terms in the authority list may vary from document to document.
As always, controlled terms stay consistent. We will be enhancing and adapting the authority list, so a hot name extracted today may be a controlled term tomorrow.
Either way, the power of LexisNexis SmartIndexing Technology will help you locate the information you need on the companies, people, organizations, places and topics important to you today!
Send questions, feedback and suggestions for terms to the Indexing Team.
|