Alternate Source Fusion ASF

The amount of financial data and news feeds is so big that it is an imperative to provide an intelligent system that leverages knowledge and exploits it efficiently for people.

Vermont Harding's Alternative Source Fusion System ASF gathers and mines streaming real-time news. Unstructured textual information from news providers is being processed in order to extract domain-specific qualitative information at the right abstraction level – for example, as sentiment indicator for asset positive or asset negative news. The result of such an information extraction is being evaluated by a machine-learning component which overlooks econometric interdependencies and allows to track changing market dynamics over time.

Natural Language Processing – An intelligent information agent extracts information from streaming real-time news for relevancy and qualitative direction. For Forex, our information agents maintain a domain-specific text corpus, including abbreviations, allowing to consider news in other languages than English as well.

Sentiment Tracking – Extracted information is being evaluated by a higher-level market model that considers information for tracking sentiment changes over time. Relevant information contributes to autonomous trading decisions, irrelevant information remains unconsidered. The model is also able to learn the reliability and correctness of extracted information in relation to a selected asset.

The Benefits of ASF

The benefits of ASF include:

• Information becomes available more quickly.
• Information is not being overlooked or lost.
• Information from other languages is also being considered.

Integration with proprietary platforms

ASF usually would integrate with VERMONT HARDING’s autonomous trading strategies. For those enterprises that are concerned about integrate it with their proprietary trading platforms, we deliver information agents as stand-alone components.