Aditya Kurve

Title: Aditya Kurve: Inventor in Streaming-Data User Profiling.

Introduction

Aditya Kurve is an inventor based in Atlanta, Georgia, whose patent work addresses user profiling and authentication using streaming data. His invention applies feature embeddings and relative locality to classify user events, including phone calls.

Latest Patents

Kurve is associated with 1 patent: “Feature Embeddings with Relative Locality for Fast Profiling of Users on Streaming Data.” The patent describes systems and methods for classification using an explicit or approximate feature map. In one embodiment, the method selects an approximate feature map based on a subset of features and a cluster’s relative locality measure. It maps features from existing data points and a new phone-call data point into a feature space, classifies the new point according to its relative locality to a cluster, stores the classification, and uses it to authenticate the user during the call.

Career Highlights

Kurve works at Pindrop Security, Inc. His patent reflects work on applying data classification techniques to user profiling and authentication in a streaming-data context.

Collaborations

Kurve’s listed coworker is Hassan Kingravi. Their shared professional context is associated with Pindrop Security, Inc.

Conclusion

Aditya Kurve’s patent focuses on using feature embeddings and relative locality to support fast user profiling and phone-call authentication from streaming data.