Scientific research and academic news keyword matching recommendation method based on deep learning
The invention discloses a scientific research news and academic paper keyword matching recommendation method based on deep learning. The adopted recommendation method is based on a news library and apaper library which are constantly rich. A keyword library is formed by combining picture recognition...
Gespeichert in:
Hauptverfasser: | , , , , , , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention discloses a scientific research news and academic paper keyword matching recommendation method based on deep learning. The adopted recommendation method is based on a news library and apaper library which are constantly rich. A keyword library is formed by combining picture recognition, a word segmentation and probability statistics method and a centence2vec method, so that a word2vec model is constructed, the similarity of the content of the word2vec model is judged by language processing, and recommendation is performed according to the similarity. When the user browses the scientific research news, some academic papers with the relevancy from high to low can be recommended to the user, and the expectation of supporting news content viewpoint question co-papers is achieved; otherwise, when the user browses the academic paper, the scientific research news similar to the academic paper can be recommended to the academic paper, so that the reader can know the latest development or information abo |
---|