Case recognition method based on question and answer template

The invention discloses a case recognition method based on a question and answer template, which is in butt joint with a quality inspection platform of a customer service platform and processes the data offline at regular time. A dialogue text enters the system in a streaming manner, and is temporarily stored in two Solr or two Solr Core after being preprocessed. The fact that the to-be-processeddata size is large, and a Solr full-text retrieval function is used subsequently, so that the approximate sentences are quickly filtered out, the pressure is reduced for a subsequent semantic similarity calculation module, and mainly, the semantic similarity calculation module is relatively time-consuming. The technology in the invention can be applied to the links of customer service personnel training and customer service quality detection, makes up for the lack of real materials in the existing training link, and automatically outputs the cases conforming to the specified question and answer template. And the metho