4th Brazilian Competition on Knowledge Discovery in Databases (KDD-BR 2020)

The Brazilian competition on Knowledge Discovery in Databases (KDD-BR) was firstly launched in the 2017 joint editions of the Brazilian Conference on Intelligent Systems (BRACIS), the Brazilian Symposium on Databases (SBBD) and the Symposium on Knowledge Discovery, Mining and Learning (KDMiLe), at Uberlândia-MG, Brazil. This first edition involved classifying images captured by one of the monitoring stations of the EXOSS Citizen Science organization, which monitors meteors crossing the southern skies. The second edition was held in partnership with the IBM research center in São Paulo, as part of the BRACIS and KDMile events. The objective was to predict the production of palm oil harvests from the company AGROPALMA . Last year , KDD-BR was a joint activity of BRACIS and ENIAC. The competition involved predicting the similarity between the partitions produced by the manual clustering of a set of molecular markers and those obtained by an auto-clustering tool, using data provided by the Corteva Agriscience company.

This year the fourth edition of the KDD-BR competition will be one of the activities of the BRACIS conference, which will take place in Rio Grande, RS, from October 20 to 23, 2020. Due to the COVID pandemic and the sanitary and economical related issues, the competition award session will be a virtual (online) conference. The challenge will be revealed soon.

4th KDD-BR (Brazilian Knowledge Discovery in Databases) competition:

Predicting unavailability of cars in a car rental agency

The 4th KDD-BR (Brazilian Knowledge Discovery in Databases) competition is one of the joint activities of the 2020 editions of BRACIS (Brazilian Conference on Intelligent Systems), KDMiLe (Symposium on Knowledge Discovery, Mining and Learning) and ENIAC (Encontro Nacional de Inteligência Artificial e Computacional), which will organized by the Universidade Federal do Rio Grande, RS, Brazil, from October 20th to October 23th, 2020.

The competition involves predicting the unavailability of cars in a car rental agency. When a car is unavailable, its status can be either "in maintenance" or "being washed". The goal is to predict the number of cars entering and leaving each of the two status, for each of the four shifts in a day, for one week ahead. The dataset was provided by the Localiza Hertz company.

The top three teams will be invited to present their solutions at a competition award session at the BRACIS 2020 conference. Due to the COVID-19 pandemic and the sanitary and economical related issues, BRACIS 2020 and all collocated events will be virtual (online).

Important Information:

• The dataset is available now at https://www.kaggle.com/c/kddbr-2020
• The solution submission deadline is: October 1st, 2020 (anywhere on Earth)
• The final results will be based on the solutions posted until October 1st, 2020. The actual final positions of the ranking will be disclosed during the event.