Racheli Lazar

Title: Racheli Lazar and Unsupervised Anomaly Detection

Introduction

Racheli Lazar is an inventor based in Tel Aviv, Israel, whose patent work includes a method for identifying unusual patterns in sequences of actions. She is associated with Intuit, Inc. The available information lists 1 patent.

Latest Patents

Lazar’s patent, “Unsupervised Anomaly Detection Using Lookahead Pairs,” describes systems and methods for detecting anomalies in action sequences. The method receives a sequence of actions and generates lookahead pairs (LAPs) of interest. Each pair identifies a target action, an origin action, and the number of actions between them.

The system compares these pairs with a LAP database to determine whether to flag them as anomalies. It refrains from flagging a pair when its number of recorded observations exceeds a threshold, and flags it when observations fall below the threshold or when the pair is absent from the database.

Career Highlights

The patent reflects Lazar’s contribution to unsupervised anomaly detection, an approach that can identify uncommon patterns without relying on manually labeled examples. Her listed affiliation is Intuit, Inc.

Collaborations

The available information connects Lazar’s patent work with Intuit, Inc. It does not specify additional collaborators or co-inventors.

Conclusion

Racheli Lazar’s listed patent presents a database-based approach to detecting anomalies in sequences of actions through lookahead pairs and observation thresholds.