Solana Develops Automated Machine Learning Labeling Solution for Public Safety Canada
Solana Networks is pleased to announce that it has developed and delivered an automated Machine Learning labeling solution for Public Safety Canada.
Data scientists continuously face a key challenge when building machine learning models which require large volumes of high-quality training data to ensure accurate results. This effort is time consuming, complex and can be quite expensive, often requiring a high degree of human intervention.
Third party services are available to manually label datasets however these costly services are only an viable option for text, video and image annotation.
In the domain of network traffic, manual labeling of traffic flows while possible on a small scale is very time consuming, error prone, expensive and requires special expertise to do on a large scale. Such tools are not available in the market today, especially for encrypted network data which is now ubiquitous.
Solana’s automated solution, TrafficWiz Labeler, leverages multiple ‘seeding’ methodologies such as TLS fingerprinting, ML clustering, DPI plus others, coupled with human-in-the-loop techniques to label encrypted network traffic with very high confidence. The end result is the creation of highly accurate labeled datasets in a short amount of time.