TIE Processing Goals

NameClassOutput ExtensionFurther ArgumentsDescription
adjustde.fu_berlin.ties.xml.XMLAdjusterxmlTries to fix corrupt XML documents, especially documents containing nesting errors
answersde.fu_berlin.ties.extract.AnswerBuilderansBuilds answer keys from from an annotated text (in XML format)
class-trainde.fu_berlin.ties.classify.ClassTrainclsClassifies a list of files, training the text classifier on each error
extractde.fu_berlin.ties.extract.ExtractorextExtracts relevant information from texts
preprocessde.fu_berlin.ties.preprocess.PreProcessoraugPreprocesses documents by converting them to a suitable XML format and adding lingustic information
re-evalde.fu_berlin.ties.eval.ReEvaluatorextRe-evaluates evaluated extractions (useful for switching the match mode -- eval.match.all)
shufflede.fu_berlin.ties.eval.ShuffleGeneratorCreates random "shuffles" of input arguments (e.g. files or URLs)
stripde.fu_berlin.ties.xml.dom.XMLStrippertxtStrips all markup from an XML document and stores the resulting plain text
trainde.fu_berlin.ties.extract.TrainerTrains the classifier used to extract information
train-evalde.fu_berlin.ties.extract.TrainEvalmetricsTrains an extractor and evaluates extraction quality