23-05-2023 дата публикации
Номер: CN116150704A
Принадлежит:
The invention discloses a multi-modal fusion representation method and system based on semantic similarity matching, and the method comprises the steps: obtaining a target text, carrying out the preprocessing, and extracting feature words in the target text; the feature words are expanded based on dictionaries, pictures and texts, a plurality of expanded dictionary vectors, expanded picture vectors and expanded text vectors are obtained, and corresponding feature vectors are generated; obtaining a reference word according to the current retrieval scene, performing traversal comparison on the reference word and the feature vector, obtaining a matching degree according to similarity calculation, and filtering to obtain the feature vector with the highest matching degree; and performing multi-modal weighted fusion on the dictionary feature vector, the picture feature vector and the text feature vector to form a feature word multi-modal feature vector in the current retrieval scene. According ...
Подробнее