Variable Length-Based Genetic Representation to Automatically Evolve Wrappers

The Web has been the star service on the Internet, however the outsized information available and its decentralized nature has originated an intrinsic difficulty to locate, extract and compose information. An automatic approach is required to handle with this huge amount of data. In this paper we present a machine learning algorithm based on Genetic Algorithms which generates a set of complex wrappers, able to extract information from the Web. The paper presents the experimental evaluation of these wrappers over a set of basic data sets.

• Type: Journal
• Year: 2010
• Where: PAAMS
• Authors: David F. Barrero, Antonio González-Pardo, María D. R-Moreno, David Camacho
• Keywords: Basic Data Sets, Decentralized Nature, Intelligent Agents, Intrinsic Difficulty, PAAMS 2010