Text search method, device and system based on fragment embedding and storage medium

The invention discloses a text search method, device and system based on fragment embedding and a storage medium, and relates to the technical field of text search. The method comprises the following steps: acquiring original text data and preprocessing the original text data to obtain preprocessed...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: TANG ZHONGZHU, LIN ZHENGYU
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a text search method, device and system based on fragment embedding and a storage medium, and relates to the technical field of text search. The method comprises the following steps: acquiring original text data and preprocessing the original text data to obtain preprocessed text data; the pre-processed text data is subjected to fragment embedding processing and stored in a vector database, the vector database is searched in the retrieval process to obtain a corresponding search result, and a fragment embedding processing method comprises the steps that part-of-speech tag processing and blocking processing are conducted on the pre-processed text data, and discrete text data are obtained; performing vector processing on the discrete text data, and mapping the discrete text data into a real number vector; and performing distributed representation vector search on the real number vector, storing the real number vector into a vector database, and searching the vector database in a retrieva