Automatic Annotation of Parameters from Nanodevice Development Research Papers
Sae Dieb, Masaharu Yoshioka, Shinjiro Hara, Marcus C. Newton · 2014
In utilizing nanodevice development research papers to assist in experimental planning and design, it is useful to identify and annotate characteristic categories of information contained in those papers such as source material, evaluation parameter, etc.In order to support this annotation process, we have been working to construct a nanodevice development corpus and a complementary automatic annotation scheme.Due to the variations of terms, however, recall of the automatic annotation in some information categories was not adequate.In this paper, we propose to use a basic physical quantities list to extract parameter information.We confirmed the efficiency of this method to improve the annotation of parameters.Recall for parameters increases between 4% and 7% depending on the type of parameter and analysis metric.