Stephan Porada

Infos about me and stuff that I do

Works

This is an overview of the software projects I have worked on:

Bundesdata: N-Gram Viewer for the transcripts of the german Bundestag (and more)

I created this web application as part of my master thesis. The Bundesdata project aims to make the Bundestag plenary records accessible and analysable for all citizens in a structured and simple form and to provide interactive statistics on them.

For this undertaking, the XML versions of the Bundestag records provided by the Federal Government were automatically provided with additional information and structured. With the various tools on Bundesdata, the transcripts from 1949 to 2017 can be easily searched and their contents analysed.

The app can be visited on https://bundesdata.sporada.eu/

The app is in German because it covers German politics. The N-Gram Viewer is still fun to look at even if you do not understand German. It is inspired by the Google Books Ngram Viewer, so maybe this helps using and understanding the Bundesdata N-Gram Viewer.

nopaque: Web application for researchers bundling OCR and NLP tools

The web app nopaque was developed in a team effort by Patrick Jentsch and me during my time at Bielefeld University working at the INF | Digital Humanities project of the SFB 1288.

Our web application nopaque combines coordinated tools such as Optical Character Recognition (OCR), Natrual Language Processing (NLP) and a powerful Keyword In Context Search (KWIC) with the CQP query language.

nopaque offers the possibility to use all tools individually or as a workflow. All work steps are coordinated in such a way that individual services can be used on top of each other. The platform supports researchers in converting their files into formats that can be further processed, automatically enriching them with information and then analyzing them, so that nopaque maps a large part of the research processes in the humanities. With this toolbox we address researchers in the humanities from all disciplines and levels of knowledge. The data generated during the processes can be downloaded after each step in order to evaluate or further process them with other (external) tools.

nopaque is still being actively developed and the source code can be found here: https://gitlab.ub.uni-bielefeld.de/sfb1288inf/nopaque

nopaque can be visited on https://nopaque.sfb1288.uni-bielefeld.de/