Andrés Fernando SANOJA VARGAS

@proceedings{Sanoja2008,

title = {Overview of extratos: yet another service oriented information extraction system for the web },

author = {Andrés Sanoja and Claudia León},

editor = {Jornadas Chilenas de Computación},

url = {https://web.archive.org/web/20090219124306/http://lahuen.dcc.uchile.cl/~jcc2008/libro.pdf},

isbn = {978-956-319-507-1},

year  = {2008},

date = {2008-11-15},

publisher = {Universidad de Magallanes},

abstract = {This article describes the design and implementation of Extratos, a Service Oriented Information Extraction System for web content sharing, based on web services as extractors and BPEL business process generation. Some insights from archaeological sciences are applied to the design of the system. It is organized in five subsystems: Xpathula, Lab, Node, Web Portal and Executor and the external entities web browser, web page and orchestration engine. Our solution present a web extraction process, from the perspective of users and software systems, with three phases: design, generation and execution. The goal of the design phase is to help the user "discover" text from web pages, transform them in

Text References, conform Pages and compose a Mashup, and the corresponding extraction procedure. The goal of the generation phase is, based on design, is convert the mashups as the result of a service oriented process. For the execution of the process is given a service oriented infrastructure to allow access to software clients through to mashups, using web services standard protocols },

keywords = {bpel, distributed systems, web content extraction},

pubstate = {published},

tppubtype = {proceedings}

}