Gazprom Neft invites: GPN Data Science Meetup
At first glance, it seems that only effective managers work in Gazprom Neft, and in IT there are maximum sysadmins and seconded implementers of everyones Oraklov and SAPs. From the second - you can find your own development (web fronts, mobile applications, software for managing gas stations), and, for example, the date of the Scientists. Under the cut - about the internal DS mitap, which we decided to make external.
First, we recall about the business. Gazprom Neft is a big company, it has exploration and production, there are oil refineries, there is logistics and sales - we deal with them. In this part of the company are responsible for the delivery and sale of motor fuels - from factories to end customers. Our most visible part for motorists is the gas station network. This is a large retail (more than 1,800 stations in Russia, the CIS and Europe) and we sell not only gasoline (for example, we sell coffee in large quantities).
')
Tasks in DS like any retail: sales forecasting, pricing, assortment optimization, logistics, marketing and geo-analytics, client profiles, outflow management, video and audio analytics, some credit scoring. The work is carried out in different divisions in small groups, while the tasks, methods and tools often overlap. In order to share experiences, level out competencies and discuss common problems and interests, we organized the Scientists Date Guild.
Every 3-4 weeks, guild meetings are held by free and not-so-Scientists of the company, listening to and discussing the reports of colleagues on their projects, topics, tools, and best practices. Recently, we began to invite external speakers. And now we are ready to invite the willing listeners.
On Friday, June 7 , we invite practicing Scientists and all who are interested in the topics below, to the mitap in our lecture hall (St. Petersburg, Vilensky Lane, 14). You will hear three reports, and you will also be able to present your topic on the next mitap.
Register here
Reports this time:
“Interpreted ML: the task of estimating the function of price elasticity of demand”
Ruslan Yermagambetov, Data Scientist, Gazpromneft
Ruslan will talk about what tools exist today for interpreting models - “black boxes”, that is, extracting the nature of the relationship between target and features; consider the most popular ones (Feature Importance in “wooden” models, LIME and SHAP). Then it will tell how these methods are used in the pricing project to evaluate the function of the elasticity of the client’s demand for price.
“Generative-concurrent neural network (GAN) architecture, examples of their application”
Maxim Balashov, ITMO University student, specialist of the Center for Advanced Analytics, Gazpromneft
A report on the popular architecture of neural networks: the main details of the architecture, there will be examples of applications of architecture modifications.
"Recognition of license plates at Gazpromneft gas stations
Alexander Popkov, Specialist of the Center for Competences in Advanced Analytics (Gazpromneft), Master in St. Petersburg State University in the direction of "Information Business Analytics"
Alexander will tell about the task as a whole, will analyze in detail the application of the classical Viola-Jones method for detecting an object in an image (car number).
Link to broadcast here .
First, we recall about the business. Gazprom Neft is a big company, it has exploration and production, there are oil refineries, there is logistics and sales - we deal with them. In this part of the company are responsible for the delivery and sale of motor fuels - from factories to end customers. Our most visible part for motorists is the gas station network. This is a large retail (more than 1,800 stations in Russia, the CIS and Europe) and we sell not only gasoline (for example, we sell coffee in large quantities).
')
Tasks in DS like any retail: sales forecasting, pricing, assortment optimization, logistics, marketing and geo-analytics, client profiles, outflow management, video and audio analytics, some credit scoring. The work is carried out in different divisions in small groups, while the tasks, methods and tools often overlap. In order to share experiences, level out competencies and discuss common problems and interests, we organized the Scientists Date Guild.
Every 3-4 weeks, guild meetings are held by free and not-so-Scientists of the company, listening to and discussing the reports of colleagues on their projects, topics, tools, and best practices. Recently, we began to invite external speakers. And now we are ready to invite the willing listeners.
On Friday, June 7 , we invite practicing Scientists and all who are interested in the topics below, to the mitap in our lecture hall (St. Petersburg, Vilensky Lane, 14). You will hear three reports, and you will also be able to present your topic on the next mitap.
Register here
Reports this time:
“Interpreted ML: the task of estimating the function of price elasticity of demand”
Ruslan Yermagambetov, Data Scientist, Gazpromneft
Ruslan will talk about what tools exist today for interpreting models - “black boxes”, that is, extracting the nature of the relationship between target and features; consider the most popular ones (Feature Importance in “wooden” models, LIME and SHAP). Then it will tell how these methods are used in the pricing project to evaluate the function of the elasticity of the client’s demand for price.
“Generative-concurrent neural network (GAN) architecture, examples of their application”
Maxim Balashov, ITMO University student, specialist of the Center for Advanced Analytics, Gazpromneft
A report on the popular architecture of neural networks: the main details of the architecture, there will be examples of applications of architecture modifications.
"Recognition of license plates at Gazpromneft gas stations
Alexander Popkov, Specialist of the Center for Competences in Advanced Analytics (Gazpromneft), Master in St. Petersburg State University in the direction of "Information Business Analytics"
Alexander will tell about the task as a whole, will analyze in detail the application of the classical Viola-Jones method for detecting an object in an image (car number).
Link to broadcast here .
Source: https://habr.com/ru/post/454588/
All Articles