Matching degree determination method, device, equipment and medium
A technology for determining the method and matching degree, which is applied in the fields of instrumentation, computing, electrical and digital data processing, etc. It can solve the problems of information annihilation, unfavorable historical dialogue and estimation of the matching degree of candidate responses, and achieve the effect of improving the accuracy.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0030] figure 1 It is a flow chart of a matching degree determination method provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of determining the matching degree between the target candidate replies and the historical dialogue, so as to determine the final reply from the candidate replies based on the matching degree. The method can be executed by a device for determining matching degree, and the device can be implemented by software and / or hardware. see figure 1 , the matching degree determination method provided in this embodiment includes:
[0031] S110. Based on the vector representation of the target candidate reply and the vector representation of at least one historical conversation, determine an association vector describing the association information between each historical conversation and the target candidate reply.
[0032] Wherein, the target candidate reply is the candidate reply to be scored.
[0033] Specific...
Embodiment 2
[0059] figure 2 It is a flow chart of a matching degree determination method provided in Embodiment 2 of the present invention. This embodiment is an optional solution proposed on the basis of the foregoing embodiments. see figure 2 , the matching degree determination method provided in this embodiment includes:
[0060] S210. Based on the dependency relationship among the components in the sentence, respectively update the vector representation of the target candidate reply and the vector representation of at least one historical dialogue, and generate a reply relationship vector and at least one dialogue relationship vector.
[0061] Specifically, the extraction of dependencies between components in a sentence can be realized based on a gated recurrent unit.
[0062] The vector representation of the target candidate reply is input to the gated recurrent unit, and the reply relationship vector is output; the vector representation of the historical dialogue is input to th...
Embodiment 3
[0067] image 3 It is a flowchart of a matching degree determination method provided in Embodiment 3 of the present invention. This embodiment is an optional solution proposed on the basis of the foregoing embodiments. see image 3 , the matching degree determination method provided in this embodiment includes:
[0068] S310. Based on the vector representation of the target candidate reply and the vector representation of at least one historical conversation, determine an association vector describing the association information between each historical conversation and the target candidate reply.
[0069] S320. Integrate the depth correlation matrix along the direction of the historical dialogue, and integrate the integration results along the direction of the target candidate reply to generate a reply integration vector.
[0070] Wherein, the depth correlation matrix is composed of at least one correlation vector.
[0071] Specifically, multiple correlation vectors desc...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


