Dialogue interaction method and device and electronic device

Abstract

The invention provides a dialogue interaction method and device and an electronic device, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining acandidate answer set corresponding to a dialogue operation; wherein the candidate answers in the candidate answer set comprise at least one retrieval answer and / or at least one generated answer; inputting the dialogue questions corresponding to the dialogue operation and the candidate answer set into a pre-trained sorting model; wherein the sorting model is obtained based on dialogue corpus sample training; calculating the matching degree of the dialogue question and each candidate answer in the candidate answer set through a sorting model; generating a sorting result of the candidate answersbased on the matching degree of each candidate answer; and replying the dialogue operation according to the sorting result. According to the invention, the man-machine conversation effect can be effectively improved, and the user satisfaction is improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a dialog interaction method, device and electronic equipment. Background technique [0002] Human-machine automatic dialogue is an important application in the field of artificial intelligence. At present, natural language processing technology is mostly used to realize automatic dialogue. In human-computer dialogue technology, it is usually first to set up multiple matching sub-models, and different sub-models are used to calculate the matching degree between candidate answers generated in different ways and user questions; then the matching degree of multiple sub-models and the preset The artificial features related to the answer are input to the tree model based on GBDT (Gradient Boosting Decision Tree) or Xgboost, and the scores of each candidate answer are output through the tree model; finally, the candidate answers are sorted according to the scores...

AI Technical Summary

Problems solved by technology

[0003] However, the inventors have found through research that, in the above method, since the process of calculating the matching degree of different sub-models is not uniform, the differences between the sub-models lead to inaccurate ranking results of candidate answers, which makes it difficult for users to answer the questions. Satisfied; at the same time, the resource consumption of multiple sub-models is too large and the deployment is cumbersome, resulting in low efficiency in the generation of answer sorting results, unable to respond to users in a timely manner, resulting in poor man-machine dialogue and poor user experience

Images