Financial news is a key resource for investors to base decisions on. While quarterly results are the measure of the health of a company, the experts’ interpretations of the reported numbers, as written down in financial analyses, create a lot of additional nuance.
Humans are very good at collecting information from written statements, however, to do this consistently and at scale is a challenge. In recent years, Natural Language Processing (NLP) has become a hot topic and various commercial offerings are now available that interpret financial news in a more automated and structured fashion. The output from these approaches is mostly centered on keyword counts and sentiment scores.
At Causality Link we believe there’s far more information in text than is currently extracted by the current class of NLP solutions. This article discusses which elements we extract from text, how we convert those into actionable statistics and how this enables financial institutions to get more value out of financial news.