A neural network for semantic labelling of structured information
作者：
Highlights：
• A novel approach to perform semantic labelling with neural networks is presented.
• The existing proposals focus on features engineering instead of classification techniques.
• Neural networks handle well the large number of features and allow nonlinearity.
• Experimental results show consistent improvement in every tested scenario.
• Tests with different subsets of features compare their usefulness and impact.
摘要
•A novel approach to perform semantic labelling with neural networks is presented.•The existing proposals focus on features engineering instead of classification techniques.•Neural networks handle well the large number of features and allow nonlinearity.•Experimental results show consistent improvement in every tested scenario.•Tests with different subsets of features compare their usefulness and impact.
论文关键词：Semantic labelling,Information integration,Neural networks
论文评审过程：Received 11 February 2019, Revised 10 October 2019, Accepted 24 October 2019, Available online 2 November 2019, Version of Record 8 November 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.113053
