Using cognitive search
Puzzel's Knowledgebase uses cognitive search which is a semantic search, unlike keyword search. It works really well on set of words explaining what you are looking for instead of a single word. Here is an explanation of the difference:
Typical Keyword Search
"Exact words matching"
Looks for article that exactly match the keywords you type without any meaning or understanding of the context.
If you search for:
"Best places to eat pizza New York"
A keyword search engine looks for pages with exact matches like:
- "best"
- "places"
- "eat"
- "pizza"
- "New York"
Even if the article says "Top restaurants serving pizza in NYC" — if it doesn't have your exact words, it might not show up.
Semantic Search
"Understands the meaning behind your query."
- Uses Natural Language Processing (NLP) & AI.
- Understands synonyms, context, intent.
- Looks for conceptually similar content — not just exact matches
If you search for the same exact text as above:
"Best places to eat pizza New York"
Semantic search might also show pages that say:
- "Top-rated pizzerias in NYC"
- "Where to find delicious pizza in Manhattan"
- "Popular New York pizza joints locals love"
Even though the words aren’t exact, it understands you're looking for pizza recommendations in NYC.