Reforming Traditional ML Algorithms with Spatio Temporal Analytics Capability for Big Data

Talk by: Lei Gao and Jing Xu (IBM)
Spatial and temporal information is commonly introduced in business analysis and provides valuable characteristics. To gain the insights from data analysis and optimize decision making, it is important to utilize this wealth of information, together with other external influential features. In this session a suite of spatio-temporal analytics engines based on Spark will be presented.

The suite includes several core evolutional algorithms, which extend the capability of classic machine learning models (such as regression, clustering and association rule) into spatio-temporal analysis area. They are based on Spark machine learning framework to produce the spatio-temporal analytics capability in the context of big data. They provide standard Spark ML APIs and can be smoothly used with other Spark modules. In addition, the suite also includes the manipulation and data preparation for widely used geographical data so as to make a complete analysis solution.

The session will also cover some business scenarios to demonstrate the functionality of introduced suite.

About: Databricks provides a unified data analytics platform, powered by Apache Spark™, that accelerates innovation by unifying data science, engineering and business.

Connect with us: