Rice University researchers have invented a more efficient method for preventing the online spread of misinformation on social media, using Bloom filters trained with artificial intelligence.
Rice's Anshumali Shrivastava and Zhenwei Dai utilized machine learning to show their Adaptive Learned Bloom Filter (Ada-BF) needs 50% less memory to equal the performance of learned Bloom filters.
According to Shrivastava, Ada-BF’s reduced need for memory means additional capacity for real-time filtering systems.
Shrivastava said, “We need half of the space, so essentially, we can handle twice as much information with the same resource.”
From Rice University
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA