Data Management Experts Share Best Practices for Machine Learning

Machine learning is on the rise at businesses hungry for greater automation and intelligence with use cases spreading across industries.

At the same time, most projects are still in their early phases as companies learn how to deal with selecting data sets and data platforms to architecting and optimizing data pipelines.

DBTA recently held a webinar with Gaurav Deshpande, VP of marketing, TigerGraph, and Prakash Chokalingam, product manager, Databricks, who discussed key technologies and strategies for dealing with machine learning.

There are several trends affecting machine learning, according to Chokalingam. These include:

  • Cloud
  • Ubiquity
  • Democratization

Companies deal with data challenges such as data corruption, read scan inefficiency, slow ingestion, schema management, data versioning and rollbacks.

The way to combat these issues is to integrate a system that contains:

  • Data stores
  • Sand box and collaboration
  • ETL tool
  • Processing engine
  • Version management
  • Schema enforcement

At TigerGraph, graph technology is what’s used to handle machine learning and its data, Deshpande said.

Graph is a natural model for interconnected data he explained. It is an organic way of modeling data for a variety of relationships and transactions. It can identify key data and process massive amounts of data along with using the power of relationships and deep analytics to provide insights.

TigerGraph offers a platform that can offer unsupervised machine learning. With TigerGraph users can gain the following benefits:

  • Real-time performance
  • Transactional (mutable) graph
  • Scalability for massive datasets
  • Deep link multi-hop analytics
  • Ease of development and deployment
  • Enterprise grade security

An archived on-demand replay of this webinar is available here.