基于融合词向量模型的特色文献分类

  • 打印
  • 收藏
收藏成功


打开文本图片集

中图分类号:TP391.1 文献标识码:A 文章编号:2096-4706(2025)08-0157-05

Abstract: In library service work, when facing local characteristic literature with a smalldata volume,library managers need to spend a great deal of time and efort manually organizing such local characteristic literature.In order to achieve automatedpre-clasificatinofcharacteristicliterature,thispaperproposestheCGBmodel,whichisanutomatedclasiiation modelforliteraturewithasmalldatavolume.TakingthecharacteristicliteraturedatasetofGuizhouProvinceas theexperimental object,the model conducts pre-training through GloVeand BERT,fuses the generated vectors,extracts andrepresents features throughTextC,andlasifsharactersticitratureofferentdatasales.Experimentalsultsidicatethatteaacy of the model with fused word vectors isat least 4 % higherthanthatof thebenchmark model.

Keywords: local characteristic literature; text classification; text vectorization

0 引言

在图书馆服务工作中,为展现地方特色建立地方文献库,图书馆管理人员需要将具有地方特色的文献从海量文献中挑选出来,与中图分类法不同,地方特色文献融合了多种类型文献,如:政治、科技、历史、小说等,却又与地方特色密切相关,将此类文献进行归纳整理需要耗费大量的时间与精力。(剩余6798字)

monitor