A clustering approach to extract data from HTML tables

作者：

Highlights：

• User-friendly HTML tables are a popular means to publish data.

• It is difficult for software agents to leverage them automatically.

• We present a new method to extract their data automatically.

• Its approach is totally unsupervised and builds on genetic clustering.

• It is as effective as the best supervised proposal, but far more efficient.

摘要

•User-friendly HTML tables are a popular means to publish data.•It is difficult for software agents to leverage them automatically.•We present a new method to extract their data automatically.•Its approach is totally unsupervised and builds on genetic clustering.•It is as effective as the best supervised proposal, but far more efficient.

论文关键词：HTML tables,Data extraction,Clustering,Genetic algorithms

论文评审过程：Received 30 October 2020, Revised 10 June 2021, Accepted 27 June 2021, Available online 13 August 2021, Version of Record 13 August 2021.

论文官网地址：https://doi.org/10.1016/j.ipm.2021.102683