METHOD FOR PROCESSING DATA, AN ELECTRONIC DEVICE, AND A COMPUTER PROGRAM PRODUCT
Embodiments of the present disclosure relate to a method for processing data, an electronic device, and a computer program product. The method includes determining a first feature representation of data to be processed; determining a second feature representation that matches the first feature representation from a plurality of candidate feature representations based on the first feature representation, wherein the plurality of candidate feature representations and corresponding candidate data are stored in a storage system in association; and determining target data that matches the data to be processed from the candidate data based on the second feature representation. Through the method, matching data can be accurately determined, and the accuracy and efficiency of data retrieval can be improved.
1 . A method, comprising:
determining, by a system comprising a processor, a first feature representation of data to be processed;
based on the first feature representation, determining a second feature representation that matches the first feature representation from a plurality of candidate feature representations, wherein the plurality of candidate feature representations and corresponding candidate data are stored in a storage system in association; and
based on the second feature representation, determining, from the corresponding candidate data, target data that matches the data to be processed.
2 . The method according to claim 1 , wherein the data to be processed is first data, wherein the plurality of candidate feature representations is stored as metadata of the corresponding candidate data, and wherein the determining the target data comprises:
determining target metadata corresponding to the second feature representation from the metadata of the corresponding candidate data; and
determining, from the corresponding candidate data, that second data corresponding to the target metadata is the target data.
3 . The method according to claim 1 , wherein the determining the first feature representation comprises:
determining preprocessed data corresponding to a second type, based on a first type of data to be processed, wherein the preprocessed data is used to determine the data of the first type as a feature representation of the data; and
determining the first feature representation of the data to be processed according to the preprocessed data.
4 . The method according to claim 3 , wherein the preprocessed data that determines the data to be processed as the first feature representation comprises a same model as a model for determining the target data as the second feature representation.
5 . The method according to claim 1 , wherein the determining the second feature representation comprises:
determining that a feature representation of a matching degree with the first feature representation greater than a threshold, of the plurality of candidate feature representations, is the second feature representation.
6 . The method according to claim 5 , wherein the matching degree is determined based on one of a Euclidean distance between the feature representation and first feature representation and a cosine distance between the feature representation and first feature representation.
7 . The method according to claim 1 , wherein the data to be processed comprises at least one of:
image data representative of an image,
video data representative of a video,
audio data representative of audio, and
textual data representative of text.
8 . A device, comprising:
at least one processor; and
a memory coupled to the at least one processor and having instructions stored thereon, wherein the instructions, when executed by the at least one processor, cause the device to perform operations, comprising:
determining a first feature representation of data to be processed;
determining a second feature representation that matches the first feature representation from a group of candidate feature representations based on the first feature representation, wherein the group of candidate feature representations and candidate data corresponding to the group of candidate feature representations are stored and associated in a storage system; and
determining target data that matches the data to be processed from the candidate data based on the second feature representation.
9 . The device according to claim 8 , wherein the data to be processed is first data, wherein the group of candidate feature representations is stored as metadata of the candidate data, wherein the determining the target data comprises:
determining target metadata corresponding to the second feature representation from the metadata of the candidate data; and
determining, from the corresponding candidate data, that second data corresponding to the target metadata is the target data.
10 . The device according to claim 8 , wherein the determining the first feature representation comprises:
determining preprocessed data corresponding to a first type, based on a second type of data to be processed, wherein the preprocessed data is used to determine the data of the second type as a feature representation of the data; and
determining the first feature representation of the data to be processed according to the preprocessed data.
11 . The device according to claim 10 , wherein the preprocessed data that determines the data to be processed as the first feature representation comprises a same model as a model for determining the target data as the second feature representation.
12 . The device according to claim 8 , wherein the determining the second feature representation comprises:
determining that a feature representation of a matching degree with the first feature representation greater than a threshold, of the group of candidate feature representations, is the second feature representation.
13 . The device according to claim 12 , wherein the matching degree is determined based on at least one of a Euclidean distance or a cosine distance.
14 . The device according to claim 8 , wherein the data to be processed comprises at least one of:
an image,
a video,
audio, and
text.
15 . A computer program product, wherein the computer program product is tangibly stored on a non-transitory computer-readable medium and comprises machine-executable instructions, and the machine-executable instructions, when executed, cause a machine to perform operations, comprising:
determining a first feature representation of data to be processed;
based on the first feature representation, determining a second feature representation that matches the first feature representation from candidate feature representations, wherein the candidate feature representations and corresponding candidate data are stored in a storage system in association; and
based on the second feature representation, determining, from the corresponding candidate data, target data that matches the data to be processed.
16 . The computer program product according to claim 15 , wherein the data to be processed is first data, wherein the candidate feature representations is stored as metadata of the corresponding candidate data, and wherein the determining the target data comprises:
determining target metadata corresponding to the second feature representation from the metadata of the corresponding candidate data; and
determining, from the corresponding candidate data, that second data corresponding to the target metadata is the target data.
17 . The computer program product according to claim 15 , wherein the determining the first feature representation comprises:
determining preprocessed data corresponding to a second type, based on a first type of data to be processed, wherein the preprocessed data is used to determine the data of the first type as a feature representation of the data; and
determining the first feature representation of the data to be processed according to the preprocessed data.
18 . The computer program product according to claim 17 , wherein the preprocessed data that determines the data to be processed as the first feature representation comprises a same model as a model for determining the target data as the second feature representation.
19 . The computer program product according to claim 15 , wherein the determining the second feature representation comprises:
determining that a feature representation of a matching degree with the first feature representation greater than a threshold, of the candidate feature representations, is the second feature representation, wherein the matching degree is determined based on one of a Euclidean distance between the feature representation and first feature representation and a cosine distance between the feature representation and first feature representation.
20 . The computer program product according to claim 19 , wherein the data to be processed comprises at least one of:
image data representative of an image,
video data representative of a video,
audio data representative of audio, and
textual data representative of text.