Data Science Consultant at Neos with focus on data analysis, visualization and predictive modelling. Passionate about data-driven decision making, deep learning and computer vision, but also about developing applications powered by machine learning. During the last year, contributed to multiple projects at Neos. Among those still in progress, the most significant one is most certainly CloudVane – a solution for tracking and predicting cloud usage costs. Thrilled to be working in one of the most promising fields of the 21st century, with many of its advancements yet to come.
How to use ML for cloud services costs prediction?
Since the early stages, Neos has been leveraging advanced analytics, machine learning and artificial intelligence concepts and tools turning terabytes of data into knowledge and predictions, helping business users make better decisions. Besides working on custom client projects related to ML implementation, at Neos, we are using ML in our own products. The latest one is CloudVane, whose primary goal is to make all cloud costs visible and controllable, enabling users to govern, analyse and manage spend, usage, and security across the entire organization.
In order to implement all this, we need not only to gather large amounts of data in Near-real time, but also to analyse and predict spending patterns and identify anomalies.
As part of the workshop, we will demonstrate how to utilize Oracle Machine Learning technologies upgraded with a custom NeosML library, for time series analysis and forecasting in a cloud spend control domain.
The workshop’s agenda can be summarized as follows:
- Cloud cost management challenges + CloudVane
- Available datasets and usage of ML
- Oracle Machine Learning
- NeosML library
- A practical example of using OML and NeosML to forecast time series
If you were searching for a workshop that covers machine learning, time series, and cloud – search no more! In 90 minutes, you’ll see how time series analysis is used to analyse the present and predict the future, in a fully automated fashion.