Spam detection for user-generated multimedia items based on appearance in popular queries
AssigneeGoogle Inc.
InventorsVarun Kacholia, Ashutosh Garg, David Petrie Stoutamire
Life of this patent
Estimating the term…
Abstract
A system, a method, and various software tools enable a video hosting website to automatically identify posted video items that contain spam in the metadata associated with a respective video item. A spam detection tool for user-generated video items based on appearance in popular queries is provided that facilitates the detection of spam in the metadata associated with a video item.
Claims15
A computer-implemented method of identifying spam in a collection of multimedia items, comprising:
- storing the collection of multimedia items;
- establishing a set of top queries for the multimedia items in the collection, wherein the set of top queries comprises a plurality of frequently made queries for the multimedia items in the collection;
- identifying results of each query in the set of top queries, wherein the results comprise one or more of the multimedia items from the collection;
- for a multimedia item in the collection, counting a number of times that the item appears in the results of the set of top queries;
- and
- responsive to the number of times that the multimedia item appears in the results of the set of top queries exceeding a predetermined threshold, marking the multimedia item as spam.
A system, comprising:
- a non-transitory computer-readable storage medium storing computer instructions for:
- storing a collection of multimedia items;
- establishing a set of top queries for the multimedia items in the collection, wherein the set of top queries comprises a plurality of frequently made queries for the multimedia items in the collection;
- identifying results of each query in the set of top queries, wherein the results comprise one or more of the multimedia items from the collection;
- for a multimedia item in the collection, counting a number of times that the item appears in the results of the set of top queries;
- and
- responsive to the number of times that the multimedia item appears in the results of the set of top queries exceeding a predetermined threshold,
- marking the multimedia item as spam;
- and
- a processor for executing the computer instructions.
A non-transitory computer-readable storage medium storing computer instructions for:
- storing a collection of multimedia items;
- establishing a set of top queries for the multimedia items in the collection, wherein the set of top queries comprises a plurality of frequently made queries for the multimedia items in the collection;
- identifying results of each query in the set of top queries, wherein the results comprise one or more of the multimedia items from the collection;
- for a multimedia item in the collection, counting a number of times that the item appears in the results of the set of top queries;
- and
- responsive to the number of times that the multimedia item appears in the results of the set of top queries exceeding a predetermined threshold, marking the multimedia item as spam.
Dependencies are inferred from the claim text.