Paypal, Inc.

Jiaxin Fang

Yu Zhang

Pengshan Zhang

Xia Zhang

Haoran Zhang

Junshi Guo

Xiaojun Luan

Jun Li

Xin Wang

Ying Yue

Kun Wang

Changle Lian

Yangxing Liu

Lance Lian

Techniques are disclosed for detecting risk via a server system that receives, based on user activity at a device, an indication of a trigger event corresponding to a potential electronic communication. After receiving the trigger event indication and prior to receiving an indication of initiation of the electronic communication, the system executes a set of computations for the communication, including performing a multi-hop query to a graph database storing a graphical representation of a plurality of communications and storing results of the execution. In response to initiation of the electronic communication, the system retrieves, using information corresponding to entities involved in the electronic communication, one or more portions of the results of executing the set of computations the database. The system determines, based on the retrieved results, whether to approve the electronic communication. The disclosed techniques may advantageously decrease unsecure electronic communications relative to real-time risk detection techniques.

Graph computing for electronic communication risk detection

A method for operating a graph database, including receiving, by a computer system, a query to a particular graph database, the query identifying a plurality of vertices of the particular graph database. The method further includes performing, by the computer system, hash operations on two or more of the plurality of vertices to generate respective hash values and dividing, using the respective hash values, the query into a plurality of sub-queries, each corresponding to a subset of the plurality of vertices. The method also includes sending, by the computer system, ones of the plurality of sub-queries to a plurality of database repositories for the particular graph database.

Dual write and dual read access to graph databases

Methods and systems are presented for assisting a user to identify and evaluate features for use in a machine learning model configured to perform a task. Based on graph data associated with a graph data structure, a user interface is provided on a device. Based on user inputs received via the user interface, a feature candidate for the machine learning model is determined. The feature candidate is associated with a particular way of traversing the graph data structure to obtain attribute values associated with one or more vertices and/or one or more edges in the graph data structure. Based on the attribute values, a value corresponding to the feature candidate can be calculated. The value can be used to evaluate the effectiveness of the feature candidate in performing the task. The feature candidate can then be incorporated into the machine learning model as one of the input features.

Graph-based feature engineering for machine learning models

Growing community of inventors

Shanghai, China

Jiaxin Fang