Text feature extraction method, device, chat robot and storage medium based on fusion model

Abstract

The invention discloses a text feature extraction method based on a fusion model, a device, a chat robot and a storage medium. The extraction method comprises the following steps: S1, adopting Word2vec model to extract words, utilizing Skip-gram model to construct a word vector, and then maps the sentences composed of words into a sentence matrix; S2, using convolution neural network to extract the local features of the word vector; S3, extracting the global features related to the context of the word vector by using the BLSTM model; S4, extracting the deeper information features of the localfeatures and the global features through Attention mechanism, and fusing the extracted features; S5, extracting the text feature vector extracted from the network layer by layer by using soft-Max classifier for text classification. The invention not only solves the problem that the single convolution neural network ignores the semantic information of the words in the context, but also effectivelyavoids the problem that the gradient of the traditional circulating neural network disappears or the gradient is dispersed.

Description

technical field [0001] The invention belongs to the technical field of text processing, and in particular relates to a text feature extraction method, device, chat robot and storage medium based on a CNN-BLSTM-Attention fusion model. Background technique [0002] At present, artificial feature engineering and shallow classification models are still used for text classification in text feature extraction. The process of training a text classifier is as follows figure 1 shown. The process of converting data into information and then refining knowledge in machine learning problems determines the upper limit of the result, while models and algorithms are used to approach this upper limit. Feature engineering is different from classifier models, which is very time-consuming and does not have strong Versatility often requires a combination of understanding of feature tasks. [0003] Therefore, deep learning methods are generally used for text feature extraction, mainly includin...

AI Technical Summary

Problems solved by technology

[0008] In the identification of chat robot stock price query intentions, traditional SVM methods (such as image 3 Shown) uses pattern matching or basic features to train the model. The classification is simple but the accuracy is low. CNN makes better use of the chat context to extract features better than describe local features, but CNN has the ability to read text Further optimization of time series requirements

Images