人工智能服务的科技情报信息收集与处理优化研究  

Research on the Optimization of the Collection and Processing of Scientific and Technological Intelligence Information Based on Artificial Intelligence Services

作　　者：王伦 WANG Lun(Neijiang Institute of Scientific and Technological Information and New Technology Development,Neijiang,Sichuan Province,641000 China)

机构地区：[1]内江市科学技术情报和新技术开发研究所,四川内江641000

出　　处：《科技资讯》2024年第10期40-42,共3页Science & Technology Information

摘　　要：研究比较了循环神经网络系列算法、自然语言处理等多个算法在Penn Treebank数据集上的表现,采用ROUGE得分作为评价指标。结果显示,自然语言处理模型在自动摘要任务上表现最佳,能够生成更准确、更全面的摘要。此外,在不同类型的数据中,基于自然语言处理技术的人工智能抽取方法也表现出较高的抽取准确率,优于传统方法。这表明自然语言处理在科技情报信息收集与处理领域具有显著优势。This paper compares the performance of multiple algorithms such as the serial algorithm of recurrent neural networks and natural language processing(NLP)on the Penn Treebank data set,and uses the ROUGE score as the evaluation index.Results show that the NLP model performs best on the automatic summary task,which can generate more accurate and comprehensive summaries.In addition,in different types of data,the AI extraction method based on NLP technology also shows high extraction accuracy and is better than traditional methods,which shows that NLP has significant advantages in the field of the collection and processing of scientific and technological intelligence information.

关 键 词：人工智能 科技情报 收集与处理 自然语言处理 

分 类 号：G252[文化科学—图书馆学] TP18[自动化与计算机技术—控制理论与控制工程]