AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Using Predictive Datasets to Predict the Market
In the end, the model can be packaged as a dataset itself when it's ranking. So we have an algorithm, a property algorithm, so we're calculating the sentiment on stocks based on public news. We cover 10,000 global stocks approximately and so we end up with a list of sentiment which is a daily file basically. Then you might want to say, to decide what kind of strategy you want to use that for. You might think if it's a positive sentiment, then I want to invest or if it’s a positive sentiment with an incremental number of news, then it's interesting - even this depends also on your time horizon.