[Pre-proposal screening]

The pre-proposal screening is a review of the proposal materials sent in by the applicants for contest entry.

The professional fields of the applicants, their experiences in prediction and analysis, and strengths and viewpoints of their methods are considered in the pre-proposal screening.

If the Contest Hosts determine that the contents of the proposal materials do not fit the purpose of this Contest, the applicant will not be selected as a Participant.

Note, however, that if an applicant’s proposal is not accepted, the applicant may submit a revised proposal for reconsideration (re-application of entry) within the entry application period.

 

[First Selection / Final Selection]

In the Contest review processes, the judging committee will evaluate a Participant’s prediction results, explanatory report on the method, and presentation video. The selections of successful Participants will be based on the following selection method and evaluation points by item.

 

◆ Selection method

First Selection

 

Final Selection (2019.02.13 Some misprints were corrected.)

 

◆  Evaluation points by item

Prediction accuracy

(Quantitative evaluation based on the prediction data)

An ideal method has fewer and smaller gaps, on average, between the predicted values and actual values. The "Root Mean Square Error (RMSE)" in the entire period is used as the index.

Evaluation point: The RMSE calculated based on the errors between the prediction results (fi) and actual values for the target prediction date and time (ai) should be small.

◯ Method (practicality)

(qualitative evaluation based on the Participant’s explanatory report describing the method, along with the presentation video)

Evaluation point: A practical method is proposed while certain levels of prediction accuracy are secured. An ideal method is simple, requires comparatively less time and effort for calculation, uses fewer data types, and directly predicts a wide area without relying on data accumulated at separate area points, etc.

 

◯ Method (expandability)

(qualitative evaluation based on the Participant’s explanatory report describing the method, along with the presentation video)

Evaluation point: Certain levels of prediction accuracy are realized thanks to a creative idea, such as a method based on a unique viewpoint or the unique ideas of the Participant, a method that uses unique data, etc.

◯ Other

(qualitative evaluation mainly based on the Participant’s explanatory report describing the method, along with the presentation video)

Evaluation point: The Participant’s explanations and discussions on the proposed method, the reasons for using it, or the like.

 

This page is not for entry. Users who want to enter the contest, please find the submit button on Application Guidelines.

→ Back to Application Guidelines