Question sentence recommendation method and system

Abstract

The invention provides a question sentence recommendation method. The method includes the following steps: S1, receiving corpus data, wherein the corpus data are multi-round question and answer data;S2, transforming the corpus data to generate positive example pairs, and generating counter example pairs through random sampling and combination with the corpus data; S3, carrying out word vectorization on the positive example pairs and the counter example pairs through a word2vec model to respectively acquire sentence vector matrices; S4, inputting the sentence vector matrices into a hidden layer, and carrying out dot product operating on the sentence vector matrices and a weight matrix to obtain new sentence vector matrices; S5, inputting the sentence vector matrices into a convolutional neural network, and carrying out convolution and pooling sampling operation to obtain semantic vectors of the sentences; and S6, carrying out non-linear transformation on the semantic vectors of the sentences, calculating cosine similarity of the semantic vectors of the positive example sentence pairs and cosine similarity of the counter example sentence pairs, and finally, acquiring a prediction model. The invention also provides a question sentence recommendation system used for realizing the above-mentioned method.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a question recommendation method and system. Background technique [0002] In the existing question answering system, the user usually asks a question first, and then the system will give an answer. In this way, the system is often passive, looking for answers based on the user's questions. see figure 1 , which is a flowchart of an existing question matching method. Currently, a keyword-based question matching method is usually used for question matching, and the specific steps are as follows: [0003] 1) Perform word segmentation and part-of-speech tagging on the questions submitted by players [0004] 2) Use the stop word list to filter out the stop words in the question [0005] 3) Classify the questions submitted by the players to determine the type of questions [0006] 4) According to the question type, the keywords are appropriately expanded [0007] 5) Divide...

AI Technical Summary

Problems solved by technology

[0009] 1. Difficult to choose good keywords correctly

[0010] 2. It is difficult to determine the weight of keywords

[0011] 3. Although the expansion of keywords improves the recall rate of the system, if the expansion is not appropriate, the correct rate of retrieval will be greatly reduced

[0012] 4. When keywords have polysemy, the accuracy of question matching is not high

[0013] 5. For more complex questions, it is difficult for the system to judge the real intention of the user by relying on keywords

Images