DKPro TC: A Java-based Framework for Supervised Learning Experiments on Textual Data

Abstract

We present DKPro TC, a framework for supervised learning experiments on textual data. The main goal of DKPro TC is to enable researchers to focus on the actual research task behind the learning problem and let the framework handle the rest. It enables rapid prototyping of experiments by relying on an easy-to-use workflow engine and standardized document preprocessing based on the Apache Unstructured Information Management Architecture (Ferrucci and Lally, 2004). It ships with standard feature extraction modules, while at the same time allowing the user to add customized extractors. The extensive reporting and logging facilities make DKPro TC experiments fully replicable.

Controversy awareness on evidence-led discussions as guidance for students in wiki-based learning

“…Furthermore, such analyses should be conducted in conjunction with the individuals' personal need for cognitive closure, which was identified as a meaningful determinant to guide one's information-seeking behaviour when dealing with ambiguous contents. The presented approach might be enriched by developing and implementing methods of automating the generation of visualisations representing controversy status information on wiki talk page discussions with the help of natural language processing (Bär et al, 2011;Daxenberger et al, 2014). Collaborations in this area between computational and psychological researchers could be fruitful to draft and test cognitive group awareness tools focused on evidence-led controversies for real-world deployment opportunities, such as on Wikipedia or Wikiversity talk pages.…”

On the Role of Discourse Markers for Discriminating Claims and Premises in Argumentative Discourse

“…We trained three Machine Learning algorithms (Naive Bayes (NB), Random Forests (RF) and Support Vector Machine (SVM)) on the three datasets annotated by the annotators A1, A2 and A3, using 10-fold cross-validation and the DKPro TC framework (Daxenberger et al, 2014). For the classification experiments, we used all 88 documents in the annotated corpus (including the documents from the pre-study).…”

Recognizing Insufficiently Supported Arguments in Argumentative Essays

“…As a learner, we use the same support vector machine as for SVM-bow. For feature extraction and experimentation, we use the DKPro TC text classification framework (Daxenberger et al, 2014). We tried various features which have been used previously for assessing the quality or the persuasiveness of arguments (cf.…”