Predicative Analytics MeetUp: Apache Spark

Discussion & Demo: How Apache Spark is Being Used for Machine Learning & AI

Apache Spark is a powerful open source processing engine built around speed, ease of use, and sophisticated analytics. It was originally developed at UC Berkeley in 2009. Spark comes packaged with higher-level libraries, including support for SQL queries, streaming data, machine learning, deep learning and graph processing. These standard libraries increase developer productivity and can be seamlessly combined to create complex workflows.

Databricks, founded by the team that created Apache Spark, will be providing an overview of Spark, the current state of the Project and how Spark is being applied to various Predictive Analytics’ use cases.

Featured Presenter: Steven Yu, Principle Engineer from Databricks, will then perform a live Deep Learning demo leveraging the Databricks platform & Apache Spark

Demo Abstract: We install surveillance cameras as a security measure to improve safety and provide real-time monitoring. However, it’s difficult to watch these cameras 24/7 and be immediately reactive. But what if we could?

In this session we will explore a deep learning pipeline with Databricks & Apache Spark to analyze security camera footage and identify suspicious behavior. We will ingest, preprocess and featurize a data set of videos to train a machine learning model.

Agenda:

6:00pm -- Networking

6:30pm to 8:00pm -- Interactive Discussion and Demo

8:00pm -- Closing Comments and Questions