# US 2013/0159227 A1 · Clustering cookies for identifying unique mobile devices > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2013/0159227 A1 - Document ID: US20130159227A1 - Status: Published application - Kind: A1 - Application number: 13/767,695 - Filed: 2013-02-14 - Published: 2013-06-20 - Earliest claimed: 2010-12-23 - Assignees: Yahoo! Inc. - Inventors: Anirban Dasgupta; Liang Zhang; Maxim Gurevich; Achint Oommen Thomas; Belle Tseng - Web page: https://patentgrep.com/patent/US20130159227A1 ## CPC - G06N 20/00: Machine learning - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Embodiments are directed towards clustering cookies for identifying unique mobile devices for associating activities over a network with a given mobile device. The cookies are clustered based on a Bayes Factor similarity model that is trained from cookie features of known mobile devices. The clusters may be used to determine the number of unique mobile devices that access a website. The clusters may also be used to provide targeted content to each unique mobile device. ## Claims ### Claim 1 A method operating on one or more network devices, the method comprising: receiving a plurality of cookies, wherein each cookie is associated with one or more Interne Protocol (IP) addresses; partitioning the plurality of cookies into one or more connected components based on the one or more IP addresses, each of the one or more connected components including one or more cookies from the plurality of cookies; and for each of the one or more connected components, generating one or more clusters of cookies based on a similarity between cookies, wherein each of the one or more clusters is associated with a unique mobile device. ### Claim 2 (depends on 1) The method of claim 1, wherein the method further comprising: sorting the one or more cookies based on a start-time associated with each of the one or more cookies. ### Claim 3 (depends on 1) The method of claim 1, wherein the method further comprising: employing the clusters to identify a plurality of unique mobile devices, wherein each cluster is determined to be statistically associated with a unique mobile device. ### Claim 4 (depends on 1) The method of claim 1, wherein the similarity further comprises a Bayes Factor similarity model based on training data obtained from a plurality of known mobile devices. ### Claim 5 A network device, comprising: a transceiver to send and receive data over a network; and a processor that is operative on the received data to perform actions, including: receiving a plurality of cookies; partitioning the plurality of cookies into one or more subsets, wherein each of the one or more subsets includes one or more cookies; and clustering the one or more cookies within each of the one or more subsets based on a similarity of feature sets between cookies, wherein each cluster is associated with a unique mobile device. ### Claim 6 (depends on 5) The network device of claim 5, wherein the partitioning of the plurality of cookies into one or more subsets further comprises partitioning the plurality of cookies into one or more connected components based on one or more cookie IP addresses, wherein each of the one or more connected components includes one or more cookies. ### Claim 7 (depends on 5) The network device of claim 5, wherein partitioning the plurality of cookies into one or more subsets further comprises partitioning the plurality of cookies into one or more subsets based on a start-time of each of the plurality of cookies. ### Claim 8 (depends on 5) The network device of claim 5, wherein partitioning the plurality of cookies into one or more subsets further comprises partitioning the plurality of cookies into one or more subsets based an access frequency of each of the plurality of cookies. ### Claim 9 (depends on 5) The network device of claim 5, wherein the processor is operative to perform actions, further comprising: sorting the one or more cookies within each of the one or more subsets by a start-time. ### Claim 10 (depends on 5) The network device of claim 5, wherein the feature sets includes at least one of a start-time or a lifetime length; and wherein the similarity is based on a combination of Bayes Factors for each of a plurality of features in the feature set. ### Claim 11 A system, comprising: at least one network device that manages a communications over a network; and one or more other network devices that are configured to perform actions, including: receiving a plurality of cookies; partitioning the plurality of cookies into one or more subsets, wherein each of the one or more subsets includes one or more cookies; and clustering the one or more cookies within each of the one or more connected components based on a similarity of feature sets between cookie, wherein each cluster is associated with a unique mobile device. ### Claim 12 (depends on 11) The system of claim 11, wherein the one or more other network devices are further configured to perform actions, comprising: partitioning the plurality of cookies into one or more connected components based on one or more cookie IP addresses, wherein each of the one or more connected components includes one or more cookies. ### Claim 13 (depends on 11) The system of claim 11, wherein the one or more other network devices are further configured to perform actions, comprising: partitioning the plurality of cookies into one or more subsets based on a start-time of each of the plurality of cookies. ### Claim 14 (depends on 13) The system of claim 13, wherein the one or more other network devices are further configured to perform actions, comprising: partitioning the plurality of cookies into one or more subsets based on an access frequency of each of the plurality of cookies. ### Claim 15 (depends on 13) The system of claim 13, wherein the one or more other network devices are further configured to perform actions, comprising: sorting the one or more cookies within each of the one or more subsets by a start-time. ### Claim 16 (depends on 13) The system of claim 13, wherein the similarity further includes a Bayes Factor similarity model based on training data obtained from a plurality of known mobile devices. ### Claim 17 A processor readable non-transitive storage media that includes instructions for identifying at least one mobile device, wherein execution of the instructions by a processor device enables actions, comprising: receiving a plurality of cookies, wherein each cookie is associated with one or more Interne Protocol (IP) addresses; partitioning the plurality of cookies into one or more connected components based on the one or more IP addresses, each of the one or more connected components including one or more cookies from the plurality of cookies; and for each of the one or more connected components, generating one or more clusters of cookies based on a similarity between cookies, wherein each of the one or more clusters is associated with a unique mobile device. ### Claim 18 (depends on 17) The media of claim 17, wherein the method further comprising: sorting the one or more cookies based on a start-time associated with each of the one or more cookies. ### Claim 19 (depends on 17) The media of claim 17, wherein the method further comprising: employing the clusters to identify a plurality of unique mobile devices, wherein each cluster is determined to be statistically associated with a unique mobile device. ### Claim 20 (depends on 17) The media of claim 17, wherein the similarity further comprises a Bayes Factor similarity model based on training data obtained from a plurality of known mobile devices.