The world and data are not static. But most machine learning models are. Once they are in production, they become less relevant with time. The data distributions evolve, the behavioral patterns change, and models need updates to keep up with new reality. The usual process is to retrain the models at defined intervals.

It is not easy to build a machine learning model. It is even harder to deploy a service in production. But even if you managed to stick all the pipelines together, things do not stop here. Once the model is in use, we immediately have to think about operating it smoothly. It is now delivering the business value, after all!

Steel production is an area that has been studied for decades, and as such the industry has remained very conservative. Despite the big data revolution beginning in the early 2000s, "old-school" industries like steel-making have largely shunned any form of data-driven applications. Fortunately, things change, and here's an example of how data analytics technologies, born within the internet industry, can be applied to an offline practice like turning pig iron into steel. When we began work with Magnitogorsk Iron and Steel Works (MMK), one of the world's largest steel producers and a leading steel company in Russia, a lot of time was spent looking for a challenge that if solved, could (a) positively impact business revenues, and (b) be completed in reasonable time.The challenge that was eventually uncovered and able to meet these criteria, is one well-known to all metallurgists: how much of each ferroalloy to add during steel-making process in order to ensure the required chemistry of the steel at the lowest possible cost. This chemistry is dictated by the international standards for steel – a list of required ranges for the amounts of each element in the final mix.

The use of data within an organisation to improve elements of the business such as the supply chain, improve decision making, and to make cost savings, is becoming more widely accepted as being vital. It is vital in respect to the business remaining competitive, vital to remaining relevant, and vital to the future of the business. One of the industries that has been looking significantly at the use of its data is the manufacturing industry, and stepping back one level to the steel industry. Magnitogorsk Iron and Steel Works (MMK), is the third largest steel company in Russia with a revenue of 9.3bn. Established in 1870, the company has taken to using machine learning technology from Yandex Data Factory to creative a competitive advantage that will see it being competitive for years to come.

One of the world's biggest banks has revealed it is leading the fight against criminals using state of the art AI technology. The Bank of Russia has teamed up with Yandex Data Factory to use the latter's machine learning services to identify unlicensed money lenders, and the websites hosting them. So far, Yandex Data Factory's custom search algorithm has helped to reveal 2,500 suspicious organisations, and played a significant role in spotting any inaccuracies, meaning that potentially fraudulent organisations will not be missed in the future. The system works by analysing key words across an existing Yandex database, which identified seven million web pages related to finance topics. The algorithm is then able to assign a web page to its correct category, before identifying if an organisation is licenced, which Yandex says occurred correctly in 98 percent of cases.

The Bank of Russia is using machine learning technology to identify unlicensed money lenders, and the websites hosting them. The technology, developed by Yandex Data Factory, has helped to reveal 2,500 suspicious organisations. The system uses algorithms to search out websites hosting illegal cash loan providers and unregulated financial activity by indexing web pages related to microfinance and consumer loans. Yandex uses keyword analysis across a search index of some seven million web pages related to finance topics. In order to help build the specialised search model, Bank of Russia experts sorted through and categorised 8,000 web pages.