BuiltinTEXT is a technology created and distributed by Management-Ware Solutions, developed to locate words and word phrases found within texts, and to link them to relevant commercial information.
Extraction: A typical analysis process begins by extracting all the relevant page content and attributes, including: text, HTML properties, location on page, URL, Title, Meta tags, custom meta tags, etc. Every such attribute conveys a specific weight to the algorithms that analyze the data.
Discovery: The extracted data is then scanned through the discovery process in relationship with the proprietary taxonomy and via a dynamic part-of-speech analysis to identify keyword relationships and significance.
Classification: The page and its best matching keywords are classified with their best matching topics. Both keywords and topics are then scored for relevancy.
Ranking: The resulting output of keywords and topics are then ranked, based not only on keyword and topic relevancy but on other parameters as well, such as topic and keyword conversion rate and click through rate, advertisement CTR and conversion, and potential revenue.
|