TECHNOLOGY THAT WORKS.
Visr's versatile system utilizes cutting edge Artificial Intelligence and Machine Learning capabilities to identify distress indicators from language and image based content.
Natural Language Processing
Rather than keywords, Visr's system is trained based on millions of real world conversations across many different scenarios. By focusing on language, our system can identify unique and sometimes surprising situations that might otherwise have gone unnoticed.
While Natural Language is a strong start, understanding the nature of an issue is far more complex. Our system goes beyond single posts and learns from documented profile histories. With over one hundred classification types across thirteen categories, our system distinguishes between important differences like a violent threat versus insulting someone's character. The system also accounts for mood, and other indicators including anxiety, depression and more. Algorithms then help identify norms and changes over time, in order to best assess risk.
Deep Learning Technology
Perhaps the most fascinating piece is that today's artifical intelligence technologies are allowing even more insight than we could have imagined. In the case of Visr, we run neural networks that continuously learn from trends and information over time, providing new and previously unknowable findings and risk associations.