Comment text-based deep learning recommendation method

Abstract

The invention discloses a comment text-based deep learning recommendation method. The method comprises the steps of obtaining semantic features and feature vector matrixes of users and articles by applying a BERT model; performing convolution, maximum pooling and full connection operations on the vector matrixes by using a BLSTM model and combining with a CNN to obtain final representations of user features and article features respectively; splicing the obtained user feature representations and the obtained article feature representations as input through an MLP full-connection network, and generating a recommendation list by applying Top-N sorting. According to the invention, word embedding feature extraction is carried out by using the BERT model, a problem of mismatch of one word withmultiple meanings is avoided, the BLSTM model avoids the problems that one-way LSTM cannot acquire semantic information from back to front and expression of sentences lacks integrity, semantic encoding of the sentences is carried out in the forward direction and the reverse direction respectively, and more accurate sentence vector representation is obtained; and more accurate implicit representation is obtained, local semantic features are extracted through CNN, and effective recommendation is carried out.

Description

Technical field[0001]The invention relates to the technical field of text classification recommendation systems, in particular to a personalized recommendation method based on deep learning of text classification.Background technique[0002]With the continuous development of Internet technology, users are increasingly demanding personalized recommendation services. How to mine user preferences based on the user's existing comments and behavior data has become a hot field in the development of the Internet.[0003]Traditional recommendation algorithms based on ratings can no longer meet the shopping needs of users. When purchasing items on e-commerce platforms, user reviews of items can better reflect the user's shopping experience. The review text contains a lot of valuable information, which largely describes the characteristics of the item and the sense of user experience. E-commerce platforms can use the recommendation system to match the characteristics of the item with the sense of...

AI Technical Summary

Problems solved by technology

[0004] In the feature extraction of existing recommendation systems, the polysemy of a word is generally not considered, and the order of words in the sentence is not considered. Although LSTM can encode long texts, it cannot be encoded from the end to the end. According to the previous information, the feature extraction in the current recommendation system mainly faces the following problems:

[0006] (2) One-way LSTM cannot obtain semantic information from back to front, and the expression of sentences lacks integrity.

Images