4 ms·
Just a thought: Whether a prediction is objectively good or not is different to whether the prediction is seen as good or not. If you can think of a mechanism t
by derrida 16y ago
Just a thought: Whether a prediction is objectively good or not is different to whether the prediction is seen as good or not. If you can think of a mechanism to figure out whether a prediction was good or not after the event has happened, the system could then make a second order prediction about whether that user's a good predictor or not. Keep me up to date if you manage to solve any of these problems.
- Oyinko 16y agoWe worked on an algorithm that will check every prediction to know if they happened or not. If members record a prediction on Beansight and that it happens, they will be rewarded. If they record predictions that will not happen, they will lose credibility points.
- derrida 16y agoHow do you check if it has happened or not? Is this based on peoples votes after the predicted time of the event? What if someone wants to make a prediction that involves when something occurs?
- geeyom 16y agoHello Derrida, The algorithm to check whether the prediction occurred is based on each single vote. It weights more people with a great expertise score and weights more votes close to the deadline or votes that passed the deadline. If the prediction comes to an end before the deadline, then the algorithm will wait to reach the deadline to validate it. But we will soon let the community close a prediction if they decide it's over. Guillaume
- daniel1980fl 16y agoI think what he asking is how will your system know whether prediction happened or not. let say I predict June 1st they will be snow. 70% says nay, 30% says yes. June 1st comes, the snow fallen down -- how will your system know that and update this prediction? you not going to do it manually right? that is, if you plan to enlarge predictions database... you not going to have 10,000 employees checking each prediction. the only way to execute it is after the thing happen, give users second vote on whether prediction was right or not... but something this can be subjective... so users which "lost" can vote like they havent "lost", lol. as much as I like the idea, it tells me there is soo much daily system management (means spending $$$) versus I cannot see a stable, solid, long-term revenue stream.