Paypal, Inc.

Yunjing Dai

Jie Huang

Jun Gu

Zhenyu Qiao

Si Feng

Methods and systems are presented for detecting groups within a graph using computer-based multi-threading techniques. These techniques provide technical improvements in computing power and efficiency for analysis of large graphs. A group detection system accesses a graph. Threads are instantiated to perform task related to a group detection process based on the nodes in the graph, where a thread is instantiated for a corresponding node. Each thread determines a neighbor count representing a number of neighbor nodes having one degree of separation from the corresponding node. Each thread also generates a list comprising an identity of the corresponding node and identities of the neighbor nodes. The thread transmits the list only to threads corresponding to a first subset of the neighbor nodes having more neighbors than the corresponding node, but not to threads corresponding to a second subset of the neighbor nodes having less neighbors than the corresponding nodes.

Graph-based multi-threading group detection

Methods and systems are presented for retrieving data associated with one or more portions of a graph by a computer system. Multiple graph indices are generated based on the graph. Each of the multiple graph indices stores records of data representing the graph in different formats. Upon receiving a request for accessing a sub-graph of the graph, edge-based attributes of the sub-graph are analyzed. The sub-graph may be divided into a first portion and a second portion of the sub-graph. A first graph index may be selected for retrieving records associated with the first portion of the sub-graph based on the edge-based attributes of the first portion of the sub-graph. A second graph index may be selected for retrieving records associated with the second portion of the sub-graph based on the edge-based attributes of the second portion of the sub-graph.

Statistics-aware sub-graph query engine

Account information for a plurality of accounts and transaction information for a plurality of transactions conducted among the plurality of accounts are analyzed. A transaction volume threshold is determined based on the analysis. A first subset of the accounts is identified. The first subset of account each corresponds to an amount of conducted transactions below the transaction volume threshold. A second subset of the accounts is identified. The second subset of the accounts each corresponds to an amount of conducted transactions above the transaction volume threshold. In response to the identification of the first subset and the second subset of the accounts, the account information and the transaction information are stored into an electronic database collectively for the first subset of the accounts, and the account information and the transaction information are stored into the electronic database separately for the second subset of the accounts.

Growing community of inventors

Shanghai, China

Yunjing Dai