Bridging the data gap with semantic search
Over the years, semantic search has shifted search engines away from displaying lists of document results based on keywords, to understanding the intention of those words and displaying targeted content that users really need.
Today, semantic search isn’t possible without AI technologies and content processing techniques that identify and extract entities, facts, attributes, concepts, and events to populate meta-data fields.
KMWorld recently held a webinar with Paul Nelson, innovation lead, Accenture, who discussed how semantic search bridges the gap between the user and the content.
Semantic search is a simple idea that means through better understanding we can get better search, Nelson explained. The search platform must bridge the gap between the user and the content they are looking for.
“What does it mean to mean, what is understanding?” Nelson said. “How does a computer understand you and what does it look like?”
The names of things and concepts can either be concrete or abstract. It’s abstract if the text is classified but not identified and it’s concrete if linked to a specific business ID.
For abstract understanding NLP will just identify entities:
- Question: Where is Lincoln? –this is probably a person
For concrete understanding NLU can identify when ambiguity exists and make intelligent guesses based on context:
- Question: Where is Lincoln? Answer: “We assume you are referring to Lincoln, Nebraska, which is here” or “did you want to know the final resting place of Abraham Lincoln, 16th President of the United States?”
Ambiguous classifications of search content can be more or less a fuzzy opinion about some content or fuzzy classification of content to abstract areas of meaning.
Accenture offers search and content analytics, said Paul Nelson. The platform offers search, content scanning and analysis, natural language, log analytics, and document understanding.
An archived on-demand replay of this webinar is available here.