Classifying articles in English and German Wikipedia

Classifying articles in English and German Wikipedia Nicky Ringland and Joel Nothman and Tara Murphy and James R. Curran School of Information Technologies University of Sydney NSW 2006, Australia {nicky,joel,tm,james}@it.usyd.edu.au Abstract Named Entity (NE) information is criti- cal for Information Extraction (IE) tasks. However, the cost of manually annotating sufficient data for training purposes, espe- cially for multiple languages, is prohibitive, meaning automated methods for develop- ing resources are crucial. We investigate the automatic generation of NE annotated data in German from Wikipedia. By incor- porating structural features of Wikipedia, we can develop a German corpus which accurately classifies Wikipedia articles into NE categories to within 1% F-score of the state-of-the-art process in English. 1 Introduction Machine Learning methods in Natural Language Processing (NLP) often require large annotated training corpora. Wikipedia can be used to au- tomatically generate robust annotated corpora for tasks like Named Entity Recognition (NER), com- petitive with manual annotation (Nothman et al., 2009). The CoNLL-2002 shared task defined NER as the task of identifying and classifying the names of people (PER), organisations (ORG), places (LOC) and other entities (MISC) within text (Tjong Kim Sang, 2002). There has been extensive research into recognising NEs in news- paper text and domain-specific corpora, however most of this has been in English. The cost of pro- ducing sufficient NER annotated data required for training makes manual annotation unfeasible, and the generation of this data is even more impor- tant for languages other than English, where gold- standard corpora are harder to obtain. German NER is especially challenging since various features used successfully in English NER, including proper noun capitalisation, do not apply to German language data, making NEs harder to detect and classify (Nothman et al., 2008). Furthermore, German has partially free word order which affects the reliability of con- textual evidence, such as previous and next word features, for NE detection. Nothman et al. (2008) devised a novel method of automatically generating English NE train- ing data by utilising Wikipedia’s internal struc- ture. The approach involves classifying all arti- cles in Wikipedia into classes using a features- based bootstrapping algorithm, and then creating a corpus of sentences containing links to articles identified and classified based on the link’s target. We extend the features used in Nothman et al. (2008) for use with German Wikipedia by cre- ating new heuristics for classification. We en- deavour to make these as language-independent as possible, and evaluate on English and German. Our experiments show that we can accurately classify German Wikipedia articles at an F- score of 88%, and 91% for entity classes only, achieving results very close to the state-of-the-art method for English data by Nothman et al. (2008) who reported 89% on all and 92% on entities only. Nothman et al.’s (2009) NER training corpus cre- ated from these entity classifications outperforms the best cross-corpus results with gold standard training data by up to 12% F-score using CoNLL- 2003-style evaluation. Thus, we show that it is possible to create free, high-coverage …

Content truncated to 500 words