Named Entity Recognition for Russian Historical Texts

With the raise of big data, machine learning and crowdsourcing, the volume of existing datasets for different machine learning problems have greatly increased. The natural language processing field is not an exception; so, as a result, most of the researches have transitioned into investigating and applying different deep architectures for it. One of the main issues of this trend is as follows: it is hard to adopt such approaches for somewhat poorly studied languages, which do not have training data enough as for natural language processing perspective. In this paper, we investigate some modern approaches to named entity recognition as for Russian language and show that for historical texts their results are much lower than for general ones. In addition, we propose our own algorithm that improves the results of for these historical texts.