Alerting on Anomalous Authorization Requests Derived with Proximity Graph

A system for automated information security authorization that facilitates the approval or denial of a user's request for access to a module via an authorization request tool comprising:

• components for receiving the user request for access to the module from the authorization request tool over a computer network;
• a computer machine comprising a processor;
• and
• computer-executable instructions stored on a computer-readable medium configured to cause the processor to perform steps comprising:
• initiating, by the computer machine from authorization validation tool, a user authorization access profile verification scan;
• selecting, by the processor, a user authorization access profile;
• loading, from the computer-readable medium by the processor into a first sector of the computer-readable medium, the user profile;
• generating, by the processor, a peer group profile for the selected user profile;
• loading, from the computer-readable medium by the processor into a second sector of the computer-readable medium, the peer group profile;
• calculating, by the processor, a peer probability distribution;
• loading, from the computer-readable medium by the processor into a third sector of the computer-readable medium, the peer probability distribution;
• calculating, by the processor, a user profile probability distribution;
• loading, from the computer-readable medium by the processor into a fourth sector of the computer-readable medium, the user probability distribution;
• calculating, by the processor, a normalized divergence value using the peer probability distribution and the user probability distribution;
• loading, from the computer-readable medium by the processor into a fifth sector of the computer-readable medium, the normalized divergence value;
• determining, by the processor, if the normalized divergence value is less than an enterprise threshold;
• generating, by the processor, a validated user profile if the normalized divergence value is less than the enterprise threshold;
• or
• generating, by the processor, an invalidated user profile if the normalized divergence value is more than the enterprise threshold;
• transmitting, by the processor, the validated or invalidated user profile;
• and
• storing, by the processor, the validated or invalidated user profile.

A non-transitory machine-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform steps comprising:

• receiving, by a computer machine from a authorization request tool, a first user request for access to a first module;
• storing, by a processor into a first sector of the machine-readable medium, the first user request;
• generating, by the processor, a peer group dataset for the first user request;
• loading, from the machine-readable medium by the processor into a second sector of the machine-readable medium, the peer group dataset for the first user request;
• generating, by the processor, a user dataset for the first user request;
• loading, from the machine-readable medium by the processor into a third sector of the machine-readable medium, the user dataset for the first user request;
• calculating, by the processor, a critical score for the first user request;
• loading, from the machine-readable medium by the processor into a fourth sector of the machine-readable medium, the critical score for the first user request;
• determining, by the processor, if the critical score for the first user request is less than an enterprise threshold;
• generating, by the processor, an approval of the first user request if the critical score for the first user request is less than an enterprise threshold;
• or generating, by the processor, a denial of the first user request if the critical score for the first user request is more than an enterprise threshold;
• storing, by the processor into a sixth sector of the machine-readable medium, the approval or the denial of the first user request;
• and
• executing, by the processor, machine learning instructions using the peer group dataset, the user dataset, the critical score, the determination if the critical score for the user request is less than the enterprise threshold, and the approval or denial of the user request to generate a second approval or a second denial of a second user request.