Hybrid forecasting of geopolitical events<sup>†</sup>
-
- Daniel M. Benjamin
- USC Information Sciences Institute Marina del Rey California USA
-
- Fred Morstatter
- USC Information Sciences Institute Marina del Rey California USA
-
- Ali E. Abbas
- University of Southern California Los Angeles California USA
-
- Andres Abeliuk
- USC Information Sciences Institute Marina del Rey California USA
-
- Pavel Atanasov
- Pytho, LLC New York New York USA
-
- Stephen Bennett
- University of California Irvine Irvine California USA
-
- Andreas Beger
- Predictive Heuristics Seattle Washington USA
-
- Saurabh Birari
- USC Information Sciences Institute Marina del Rey California USA
-
- David V. Budescu
- Fordham University Bronx New York USA
-
- Michele Catasta
- Stanford University Stanford California USA
-
- Emilio Ferrara
- USC Information Sciences Institute Marina del Rey California USA
-
- Lucas Haravitch
- University of Southern California Los Angeles California USA
-
- Mark Himmelstein
- Fordham University Bronx New York USA
-
- KSM Tozammel Hossain
- USC Information Sciences Institute Marina del Rey California USA
-
- Yuzhong Huang
- USC Information Sciences Institute Marina del Rey California USA
-
- Woojeong Jin
- USC Information Sciences Institute Marina del Rey California USA
-
- Regina Joseph
- Sibylink New York New York USA
-
- Jure Leskovec
- Stanford University Stanford California USA
-
- Akira Matsui
- USC Information Sciences Institute Marina del Rey California USA
-
- Mehrnoosh Mirtaheri
- USC Information Sciences Institute Marina del Rey California USA
-
- Xiang Ren
- USC Information Sciences Institute Marina del Rey California USA
-
- Gleb Satyukov
- USC Information Sciences Institute Marina del Rey California USA
-
- Rajiv Sethi
- Barnard College Columbia University New York New York USA
-
- Amandeep Singh
- USC Information Sciences Institute Marina del Rey California USA
-
- Rok Sosic
- Stanford University Stanford California USA
-
- Mark Steyvers
- University of California Irvine Irvine California USA
-
- Pedro A Szekely
- USC Information Sciences Institute Marina del Rey California USA
-
- Michael D. Ward
- Predictive Heuristics Seattle Washington USA
-
- Aram Galstyan
- USC Information Sciences Institute Marina del Rey California USA
Abstract
<jats:title>Abstract</jats:title><jats:p>Sound decision‐making relies on accurate prediction for tangible outcomes ranging from military conflict to disease outbreaks. To improve crowdsourced forecasting accuracy, we developed SAGE, a <jats:italic>hybrid forecasting system</jats:italic> that combines human and machine generated forecasts. The system provides a platform where users can interact with machine models and thus anchor their judgments on an objective benchmark. The system also aggregates human and machine forecasts weighting both for propinquity and based on assessed skill while adjusting for overconfidence. We present results from the Hybrid Forecasting Competition (HFC)—larger than comparable forecasting tournaments—including 1085 users forecasting 398 real‐world forecasting problems over 8 months. Our main result is that the hybrid system generated more accurate forecasts compared to a human‐only baseline, which had no machine generated predictions. We found that skilled forecasters who had access to machine‐generated forecasts outperformed those who only viewed historical data. We also demonstrated the inclusion of machine‐generated forecasts in our aggregation algorithms improved performance, both in terms of accuracy and scalability. This suggests that hybrid forecasting systems, which potentially require fewer human resources, can be a viable approach for maintaining a competitive level of accuracy over a larger number of forecasting questions.</jats:p>
Journal
-
- AI Magazine
-
AI Magazine 44 (1), 112-128, 2023-03
Wiley
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1360017282231194752
-
- ISSN
- 23719621
- 07384602
-
- Data Source
-
- Crossref
- KAKEN