Named Entity Recognition for Partially Annotated Datasets
This work addresses the challenge of noisy training data for Named Entity Recognition, which is incremental as it builds on existing methods for partially annotated datasets.
The paper tackled the problem of training Named Entity Recognizers on partially annotated datasets, which are too noisy for standard sequence taggers, by comparing three training strategies and proposing a method to derive new entity class datasets from Wikipedia without manual annotation, and manually annotated test datasets for food and drugs to verify the approaches.
The most common Named Entity Recognizers are usually sequence taggers trained on fully annotated corpora, i.e. the class of all words for all entities is known. Partially annotated corpora, i.e. some but not all entities of some types are annotated, are too noisy for training sequence taggers since the same entity may be annotated one time with its true type but not another time, misleading the tagger. Therefore, we are comparing three training strategies for partially annotated datasets and an approach to derive new datasets for new classes of entities from Wikipedia without time-consuming manual data annotation. In order to properly verify that our data acquisition and training approaches are plausible, we manually annotated test datasets for two new classes, namely food and drugs.