Extracting firm alliance networks via text mining
-
SeriesResearch Master Defense
-
SpeakerJacob Rauch
-
LocationOnline
-
Date and time
September 28, 2020
16:00 - 17:00
Abstract: We
propose a text mining model for automatically extracting alliances between
firms from news articles. We leverage the pre-trained language model RoBERTa
(Liu et al., 2019) and a large amount of labeled examples from the SDC alliance
data to fine-tune the model. The resulting system is able to detect alliance
announcements in documents, extract the participating firms, and flag alliances
according to their purpose. We show that our model is highly accurate in the
firm name recognition and relation classification tasks. We run inference on a
large corpus of news articles and show that the model can add valuable data to
existing sources.