Scientific and technical literature quotation recommendation method based on deep learning

Abstract

The invention discloses a scientific and technical literature quotation recommendation method based on deep learning. The method comprises the following steps: sequentially performing information extraction, noise removal, word list index construction and vectorization processing on text data; firstly, performing semantic vectorization representation on a to-be-matched text through Bi-LSTM, then performing interactive coding on the text by adopting an Attention mechanism, and performing feature extraction on interactive information of the text through a multilayer CNN network so as to obtain final matching degree information of the text; in the first stage, forming a related quotation recommendation set through vector space similarity of texts, and in the second stage, using a text reasoning matching method for carrying out language understanding on candidate sets so as to obtain an accurate relevancy ranking list. According to the method, the problems of semantic comprehension, content integrity and accuracy in existing text reasoning matching can be solved, high-quality semantic features and a digital text input form are provided, and feature extraction is performed on interactive information of the text, so that final matching degree information of the text is obtained.

Description

technical field [0001] The invention belongs to the technical field of information retrieval and analysis, and in particular relates to a method for recommending citations of scientific and technological documents based on deep learning. Background technique [0002] Scientific and technological innovation capability is a decisive factor in the development of a country's scientific and technological undertakings, the core of national competitiveness and an important foundation for strengthening the country and enriching the people. As the scientific and technological literature resource that condenses the achievements of scientific and technological innovation activities, it is an important carrier for the dissemination of scientific and technological knowledge, and it is also the basic source and important support for further improving the ability of scientific and technological innovation. It has become one of the country's valuable strategic resources. [0003] In the pro...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to provide a method for recommending scientific literature citations based on deep learning, which uses a text-based vector space model to quickly screen a collection of documents that match the target document. In order to narrow the recommendation range and reduce the calculation load of the model, the algorithm based on text reasoning matching is used to sort the candidate document collection more accurately to achieve a more personalized recommendation effect, which solves the semantic incomprehension and problems existing in the existing text reasoning matching. Incomplete and inaccurate content

Images