Wisdom of the Crowds for Forecasting: Future Event Pre- diction Based on Summarizing Text-based Forecasts
Abstract: Future Event Prediction (FEP) is an essential activity whose demand and application ranges across domains. While methods like simulations, predictive and time-series forecasting have demonstrated promising outcomes, their application in forecasting complex global events is not entirely reliable owing to the inability of numerical data to capture the semantic information of events. One effective way is to gather collective opinions on the future to make effective forecasts as cumulative perspectives carry the potential to evaluate probability of upcoming events, or at least they can serve as a complement to other approaches. We organize the existing research and frameworks that aim to support Future Event Prediction based on Crowd Wisdom (FEP-CW) through aggregating individual forecasts, the challenges involved, available datasets, as well as the scope of improvement and future research directions in this domain.
Submission Length: Regular submission (no more than 12 pages of main content)
Previous TMLR Submission Url: Link: https://openreview.net/forum?id=uNkU3vtnDT¬eId=uNkU3vtnDT Changes: Corrected the format to the TMLR format
Changes Since Last Submission: The last submission did not follow the TMLR format. The present work has been corrected of that.
Assigned Action Editor: ~Erin_J_Talvitie1
Submission Number: 3101
Loading