Method and apparatus for clustering platform sessions and user accounts associated with the platform sessions

A system comprising processing circuitry, a data repository, and a communications network, wherein the system is configured to:

• access, from the data repository, sessions data associated with the platform sessions, the sessions data comprising a plurality of session records, each session record comprising a session identifier, a first-level platform engagement count, a second-level platform engagement count, a platform query count, and a platform output status indicator;
• retrieve, from the data repository and for each session record, the first-level platform engagement count to generate an intermediate first-level platform engagement count, the intermediate first-level platform engagement count being at least partially determined by dividing the first-level platform engagement count by a first-level platform engagement normalization factor, wherein the first-level platform engagement normalization factor is common among all session records;
• store, in the data repository and for each session record, the intermediate first-level platform engagement count and associate the intermediate first-level platform engagement count with a session identifier;
• retrieve, from the data repository and for each session record, the second-level platform engagement count, the platform query count, and the intermediate first-level platform engagement count to generate a transformed session records data, the transformed session records data comprising a transformed second-level platform engagement count, a transformed platform query count, and a transformed first-level platform engagement count,
• wherein the transformed second-level platform engagement count is at least partially determined by dividing the second-level platform engagement count by a session-specific normalization factor, the transformed platform query count is at least partially determined by dividing the second-level platform engagement count by the session-specific normalization factor, and the transformed first-level platform engagement count is at least partially determined by dividing the intermediate first-level platform engagement count by the session-specific normalization factor, and
• wherein the session-specific normalization factor is at least partially determined, for each session record, by adding the second-level platform engagement count, the platform query count, and the intermediate first-level platform engagement count;
• and
• identify a plurality of clusters, wherein each cluster is associated with a subset of the platform sessions, wherein the identifying is based on the transformed second-level platform engagement count, the transformed platform query count, and the transformed first-level platform engagement count, and wherein each platform session of the plurality of platform sessions is associated with one cluster.

A method for clustering platform sessions and user accounts associated with the platform sessions comprising:

• accessing, from a data repository, sessions data associated with the platform sessions, the sessions data comprising a plurality of session records, each session record comprising a session identifier, a first-level platform engagement count, a second-level platform engagement count, a platform query count, and a platform output status indicator;
• retrieving, from the data repository and for each session record, the first-level platform engagement count to generate an intermediate first-level platform engagement count, the intermediate first-level platform engagement count being at least partially determined by dividing the first-level platform engagement count by a first-level platform engagement normalization factor, wherein the first-level platform engagement normalization factor is common among all session records;
• storing, in the data repository and for each session record, the intermediate first-level platform engagement count and associate the intermediate first-level platform engagement count with a session identifier;
• retrieving, from the data repository and for each session record, the second-level platform engagement count, the platform query count, and the intermediate first-level platform engagement count to generate a transformed session records data, the transformed session records data comprising a transformed second-level platform engagement count, a transformed platform query count, and a transformed first-level platform engagement count,
• wherein the transformed second-level platform engagement count is at least partially determined by dividing the second-level platform engagement count by a session-specific normalization factor, the transformed platform query count is at least partially determined by dividing the second-level platform engagement count by the session-specific normalization factor, and the transformed first-level platform engagement count is at least partially determined by dividing the intermediate first-level platform engagement count by the session-specific normalization factor, and
• wherein the session-specific normalization factor is at least partially determined, for each session record, by adding the second-level platform engagement count, the platform query count, and the intermediate first-level platform engagement count;
• and
• identifying a plurality of clusters, wherein each cluster is associated with a subset of the platform sessions, wherein the identifying is based on the transformed second-level platform engagement count, the transformed platform query count, and the transformed first-level platform engagement count, and wherein each platform session of the plurality of platform sessions is associated with one cluster.