AI.Hack - a series of hackathons in Moscow and St. Petersburg

Hi, Habr! Very soon AI and hackathons will be held in Moscow and St. Petersburg.

It is not the first time when artificial intelligence becomes the subject of a hackathon, because solutions that can be built with its help are essentially limited only by the imagination and skill of teams.

This time the tasks are set:
')

• Gazprom Neft
• Speech Technology Center
• Leaked
• BestFit.ai
• Robomed

But first things first.

Gazprom Neft

From 2016, the company has been conducting detailed logging of customers' transactions - payment for gasoline cards and goods sold in small stores at the gas station territory.

Participants for work will be available about 1.5 GB of data.

But what you need to do:

Predicting customer churn

St. Petersburg

Using the data provided, create an algorithm that will be able to identify among customers those who are planning to stop refueling at the Gazprom Neft filling stations. That is, it is necessary to create a predictive user model from the horizons of 1 month. From the options presented by the teams, the best one will be selected.

Clustering customers based on the purchase of petroleum products and the search for patterns of consumption of related goods and services

St. Petersburg

It is necessary to analyze and create the maximum possible number of customer segments based on the purchase of petroleum products.

For each of the selected segments to find the specific patterns of consumption of related goods and services.

Dataset is the transaction flow of the Gazprom Neft filling station network sales that you use to solve a case study.

• NP sales
• Sales STIU combined in nomenclature groups

The task consists of two parts:

1. Find the maximum number of patterns in the purchase of petroleum products and divide customers into segments based on each pattern
2. Within the customer segments of paragraph 1, find the maximum number of regularities in the size and fact of purchase / non-purchase of related products from details to “EKTU nomenclature Subgroup 4”

Task for Moscow

Moscow

Details will be known after March 12th.

Speech Technology Center

The work will be with the data obtained from the surveillance cameras for pedestrian crossings

Intellectual system of visual control

St. Petersburg

We need to create a "smart surveillance camera." It should monitor the movement of pedestrians and vehicles, record violations - speeding, red light, abnormal trajectories.

Leaked

Building a rating system

Moscow

The system will be used for rating drug manufacturers. The basis will be letters about the seized medicines, taking into account the reasons for withdrawal, withdrawal and termination of circulation.

Information is not provided in finished form - instead, a list of open sources is given.

BestFit.ai

Car image classification

Moscow

Development of a model for the classification of car images according to various characteristics related to the peculiarities of the perception of car design by a person. The task is at the intersection of machine learning and a new scientific direction - digital psychology. The solution will be developed based on real data from our experimental database BestFit.ai

Robomed

The company will present 2 tasks, one for Moscow and St. Petersburg. Direction - analysis of dataset impersonal selection of medical subjects.

Where and when

St. Petersburg

March 9-11 (acceptance of applications is over)
Gazprom Neft
Vilensky Lane, 14

Moscow

March 16-18 (collection of applications for participation - until March 14)
VEB commune
2nd Donskoy passage, 9

Register and learn more here .