Shed: a Framework for Reproducible Text Mining Research

TitleShed: a Framework for Reproducible Text Mining Research
Publication TypeTalks
AuthorsEmmery, C., & Daelemans W.
Place PresentedDemonstrated at ATILA 2015, Antwerp, Belgium
Year of Publication2015
Date Presented15-10-2015

Shed is a collection of Python code that is intended to make different steps of Text Mining research less script-heavy. Its core purpose is to offer a uniform interface for handling a broad range of methods commonly used in pipelines that deal with textual data. Therefore, it will ship with homebrewn methods, wrappers for popular Natural Language Processing and Machine Learning tools, and the ability to be easily extendable with new functionality. All of this incorporated into a single environment that stores every configuration detail so that results are reproducible, and the models quickly deployable to be applied to new data.

Slides219.34 KB