Efficient query clustering using multi-partite graphs

A computer-implemented method, comprising:

• receiving a corresponding subgraph of a multi-partite graph that encompasses search queries at each of a plurality computing devices, each corresponding subgraph including a different group of search queries from the plurality of search queries;
• receiving a copy of a global center vector table that includes cluster center entries for query clusters at each of the plurality of computing devices;
• filtering a corresponding copy of a global center vector table on each of a plurality of computing device to eliminate one or more cluster center entries that are irrelevant to a corresponding group of search queries;
• and
• clustering the plurality of search queries into the query clusters by at least comparing each group of search queries to a corresponding filtered copy of the global center vector table using a clustering algorithm that is implemented by the plurality of computing devices in a distributed manner.

A computing device, comprising:

• one or more processors;
• and
• a memory that includes a plurality of computer-executable components, the plurality of computer-executable components comprising:
• a graph filtration module that filters a global center vector table to eliminate one or more cluster center entries that are irrelevant to search queries encompassed in a subgraph of a multi-partite graph;
• and
• a query cluster module that clusters the search queries encompassed in the subgraph of the multi-partite graph by at least comparing each of the search queries to at least some of the cluster center entries in the global center vector table.