Ta yaya analyzer yake shirya rubutu domin bincike?

Question

IT Interview Practice · Accepted Answer

Analyzer yana shirya rubutu domin bincike ta sauya shi zuwa tokens: sassa, sau da yawa kalmomi, da za a iya daidaitawa daga baya. ## Kalli kowane canji Ga pipeline da aka saita da gangan don cire HTML, raba kalmomi da mayar da su ƙananan haruffa: ```text Blue Shoes -> Blue Shoes -> [Blue, Shoes] -> [blue, shoes] ``` - Character filter yana cire HTML kafin rabawa. - Tokenizer yana raba rubutu zuwa kalmomi. - Token filter yana sauya kalmomin zuwa ƙananan haruffa. Tokens na ƙarshe suna nuna dalilin da ya sa `shoes` da ƙananan haruffa zai iya dacewa da input na asali. Wannan misali na musamman ne; ba kowane analyzer yake cire HTML ba. ## Bincika maimakon zato A Kibana Dev Tools, wannan buƙatar tana nuna tokens na pipeline: ```http POST /_analyze { "char_filter": ["html_strip"], "tokenizer": "standard", "filter": ["lowercase"], "text": "Blue Shoes" } ``` Nemi `blue` da `shoes` a cikin token entries da aka mayar. Sannan gwada haruffa masu alamu, harsuna da lambobin samfur na gaske. Index-time analysis yana shirya terms da aka ajiye; search-time analysis yana shirya tambayoyin mai amfani. Suna buƙatar terms masu jituwa ko da pipelines ɗin biyu sun bambanta. [Takardun hukuma](https://www.elastic.co/docs/manage-data/data-store/text-analysis).

Kalli kowane canji

Bincika maimakon zato

Tambayoyi masu alaƙa