Method for video searching

A method for video searching, the method being implemented using a server storing a plurality of video files, a plurality of tag datasets that are associated with the plurality of video files, respectively, a plurality of tag words that are in a natural language, a plurality of word embeddings that correspond with the plurality of tag words, respectively, and, with respect to each of the plurality of tag words, a plurality of co-occurrences that are associated with the tag word, a frequency of one of the plurality of co-occurrences that appears in a text file being positively related to a frequency of an associated one of the plurality of tag words that appears in the same text file, each of the plurality of co-occurrences serving as a tag word, each of the plurality of tag datasets including a plurality of tags, each of which corresponds with a video segment of an associated video file, each of the plurality of tags serving as a tag word, each of the plurality of tag words being in a first language family, and having a number of first language family co-occurrences that are in the first language family and a number of non-first language family co-occurrences that are in languages not in the first language family, the method comprising:

• a) in response to receipt of a search query that is in the natural language from a user device operated by user action, obtaining a searching word embedding that represents the search query;
• b) obtaining a reference word embedding based on the searching word embedding, and obtaining a tag word associated with the reference word embedding as a reference tag word, wherein the reference word embedding is one of the plurality of word embeddings stored in the server and having a highest similarity with the searching word embedding;
• c) obtaining, based on the reference tag word and the plurality of co-occurrences stored in the server, a number of associated co-occurrences that are associated with the reference tag word, and making the reference tag word and the number of associated co-occurrences serve as candidate tag words;
• d) obtaining a target word embedding that corresponds with one of the candidate tag words, and selecting, based on the target word embedding, at least one target tag word embedding that has a similarity with the target word embedding higher than a predetermined threshold;
• and
• e) selecting, based on the at least one target tag word embedding, at least one target video segment from one of the plurality of video files that is represented by a tag associated with the target tag word embedding, and presenting the at least one target video segment to the user device;
• wherein the method further comprises, after step a), determining whether the search query is in the first language family;
• and
• step c) includes:
• in a case where the search query is in the first language family, obtaining, based on the reference tag word and the number of first language family co-occurrences, a number of associated first language family co-occurrences that are associated with the reference tag word, and making the reference tag word and the number of associated first language family co-occurrences serve as the candidate tag words, and
• in a case where the search query is not in the first language family, obtaining, based on the reference tag word and the number of non-first language family co-occurrences, a number of associated non-first language family co-occurrences that are associated with the reference tag word, making the reference tag word and the number of associated non-first language family co-occurrences serve as the candidate tag words, and translating the candidate tag words into the first language family.