Google using machine learning to filter spam on Gmail

Donna Miller
February 8, 2019

Neil Kumaran, product manager of counter-abuse technology at Google, said: "By complementing our existing ML [machine learning] models with TensorFlow, we're able to refine these models even further, while allowing the team to focus less on the underlying ML framework and more on solving the problem: ridding your inbox of spam".

Google claims that its proprietary machine-learning framework, TensorFlow, is blocking an additional 100m Gmail spam messages on a daily basis.

Since Google has already been using AI for a long time to block spam messages, this is not huge news.

Whitaker won't testify to House Intel panel unless he's promised no subpoenas
In a separate letter sent to Nadler, Assistant Attorney General Stephen Boyd demanded a response on the subpoena question. Jerrold Nadler of NY , approved a tentative subpoena to ensure that Whitaker would appear and answer questions.

There is no one definition for 'spam,' and that what looks like a spammy message to one user might be a much-awaited mail for another one, and Google is fully aware of the fact. Moreover, Google, Intel, SAP, Airbnb, and Qualcomm are the top users of this software. The company reports it's blocking over 99.9 percent of spam, phishing, and malware from reaching its 1.5 billion Gmail users. According to the company, it is successful at identifying image-based messages, emails with hidden embedded content, and messages from newly created domains that try to hide a low volume of spammy messages within legitimate traffic.

Google has been using AI along with rule-based filters to detect spam for years. "What one person considers spam another person might consider an important message (think newsletter subscriptions or regular email notifications from an application)", it says. Google is aiming to simplify the process with TensorFlow. Kumaran also says that TensorFlow will help Gmail personalise spam filters, learn from user patterns on what they judge as spam, and provide better customised results. TensorFlow Extended (TFX) is one of these components that allows Google to deploy ML pipelines in a quick and standardized fashion while TensorBoard allows it to monitor model training pipelines and quickly evaluate new models to determine their usefulness. "But CxOs want to see proof points before adopting technologies, so that's why Google is showcasing its internal uptake of TensorFlow for fighting spam".

Other reports by

Discuss This Article