NLP Challenges for SSIX

NLP in the context of the SSIX project faces all the challenges associated with social media (cp. “NLP & Social Media” section). Furthermore, the financial domain requires a specific set of linguistic tools to properly analyse the domain-specific language. This includes recognising multi-word expressions (e.g. “Return on Investment”) and identifying Named Entities such as companies or stock tickers (“Apple Inc.”, “AAPL”). Entity linking is also important In the context of sentiment analysis, for example to resolve that the names “Google” and  “Alphabet” as well as the stock tickers “GOOGL”, “GOOG”, “GOAF” and “GOCF” all refer to the same underlying entity.