3 years, 5 months ago

Semester Project for Decision Systems 2020/2021 Course

This is a semester project for students enrolled in the Decision System 2020/2021 course at the Faculty of Mathematics, Informatics and Mechanics at the University of Warsaw.


The goal of the competition is to create an efficient model for predicting whether the total 14-days sales of a particular product, offered by the Fitfood company at one of their FitBoxy locations in Poland, will exceed four pieces.

More detailed competition rules are given in Terms and Conditions.

The description of the data and evaluation metric is in the Task description section. 

The submission system will open on Monday, December 28.

Terms & Conditions
Rank Team Name Is Report Preliminary Score Final Score Submissions
True 0.8953 0.894000 15
True 0.8951 0.893000 10
True 0.8919 0.890000 5
Krzysztof Grabiec
True 0.8894 0.886600 7
Golden Bacon MAIers
True 0.8867 0.885600 14
Patryk Skwiercz
True 0.8795 0.877800 24
True 0.8783 0.876500 16
True 0.8785 0.876500 5
Krzysztof Ostrowski
True 0.8731 0.871500 7
True 0.8738 0.870800 12
Kajetan Fornalik
True 0.8671 0.864800 3
True 0.8501 0.846600 10
True 0.8197 0.815700 6
Please log in to the system!

Provided data describe a short-term sales history of products at various point of sales (PoS). The target attribute will_it_sell tells if in the following 14 day period the total sales of a given product at a particular location will be at least 4 pcs. 

The data is very similar to the one from the second graded task, with the difference that the sets in this challenge do not contain any random probes (which were deliberately added to the data from the second graded task for evaluation purposes). 

The data tables are provided as two CSV files with the ';' separator sign. They can be downloaded after the registration for the challenge. Both files (training and test sets) have exactly the same format but all the values from the will_it_sell column in the test set are missing.

The evaluation metric will be AUC. During the challenge, your solutions will be evaluated on a small fraction of the test set, and your best preliminary AUC score will be displayed on the public Leaderboard. 

The submission format: the solutions need to be submitted as text files with predictions. The file should have exactly the same number of rows as the test data table. In each row, it should contain exactly one real number expressing the likeliness that the correct target value for the corresponding test set instance is 1.

In order to download competition files you need to be enrolled.

This forum is for all users to discuss matters related to the competition. Good manners apply!

There is no topics in this competition.