Paper Number

2338

Paper Type

SP

Abstract

Recent legislation has been proposed both in the EU and in the US with the aim of regulating AI. The content of the legislation makes clear the usefulness of explainable AI (XAI). We use the context of bitcoin value prediction to demonstrate the feasibility and value of an explainable system, with the goal of advancement toward an XAI trading algorithm. Using a broad set of 55 signals culled from over 4000 social media and news sources, as well as 29 technical financial indicators, we create a system that allows the human-in-the-loop to identify the best predictors of bitcoin value. We show that it is possible to outperform the market using a simple buy-or-sell trading strategy. This work addresses a gap in sentiment analysis by taking a perspective that draws upon signaling theory and sensemaking theory.

Share

COinS
 
Jun 14th, 12:00 AM

An Explainable Algorithm to Compare the Predictive Power of Social Media and News Signals with Traditional Technical Indicators

Recent legislation has been proposed both in the EU and in the US with the aim of regulating AI. The content of the legislation makes clear the usefulness of explainable AI (XAI). We use the context of bitcoin value prediction to demonstrate the feasibility and value of an explainable system, with the goal of advancement toward an XAI trading algorithm. Using a broad set of 55 signals culled from over 4000 social media and news sources, as well as 29 technical financial indicators, we create a system that allows the human-in-the-loop to identify the best predictors of bitcoin value. We show that it is possible to outperform the market using a simple buy-or-sell trading strategy. This work addresses a gap in sentiment analysis by taking a perspective that draws upon signaling theory and sensemaking theory.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.