Detection and prevention of unwanted content on cloud-hosted services
AssigneeGoogle Inc.
InventorsMark Risher, Vishwanath Ramarao, Naveen Jamal, Youngnam Maeng, Cory O'Connor, Tejaswi Nadahalli, Sonal Rajan, Parag Namjoshi
Life of this patent
Estimating the term…
Abstract
The problem of abuse of privileges at cloud-hosted service sites is addressed by associating each user, preferably across multiple cloud-hosted service sites, with an individual “suspiciousness” score that may vary over time as additional user actions are detected and evaluated. Knowledge of the user is employed to better analyze the appropriateness or acceptability of user actions on the site.
Claims20
A method of moderating activities on cloud-hosted service sites, the method comprising the steps of:
- creating a database record for a first user when the first user registers at a first cloud-hosted service site, the database record including a plurality of profile fields and data in at least some of the profile fields;
- based on data in the profile fields of the first user's database record, assigning an initial risk score to the first user;
- identifying a cluster of database records associated with other users based on similarities in the profile fields thereof with respect to the profile fields of the first user, the record cluster having an associated composite risk score based on risk scores associated with the other users;
- assigning the composite risk score to the first user;
- and
- based on the first user's resultant risk score, taking an action with respect to a transaction conducted by the first user on the first cloud-hosted service site.
A server for moderating activities on cloud-hosted service sites, the server comprising:
- a master database comprising electronically stored records for users registered at a plurality of cloud-hosted service sites, each database record including a plurality of electronically stored profile fields and data electronically stored in at least some of the profile fields;
- a communications module for communicating bidirectionally with cloud-hosted service sites, the communications module being configured to receive, from the cloud-hosted service sites, messages posted by users thereon and data associated with newly registering users;
- and
- an analysis engine for managing the master database and assessing messages posted by users on cloud-hosted service sites and received via the communications module, the analysis engine being configured to:
- create a database record for a first user when the first user registers at a first cloud-hosted service site, the database record including a plurality of profile fields and data, received via the communications module from the first cloud-hosted service site pursuant to the first user's registration thereon, in at least some of the profile fields;
- compute an initial risk score for the first user based on data in the profile fields of the first user's database record and assign the initial risk score to the first user;
- identify a cluster of records in the master database associated with other users based on similarities in the profile fields of the corresponding database records, the cluster having an associated composite risk score based on risk scores associated with the other users;
- assign the composite risk score to the first user;
- and
- based on the first user's composite risk score, take an action with respect to a message posted by the first user on the first cloud-hosted service site.
Dependencies are inferred from the claim text.