Add Categorization for Suspicious Accounts

[BearHuddleston]

Currently, while GitHubWatchdog effectively flags accounts as suspicious, there is no categorization or tagging system to provide more details about why the account was flagged. This lack of categorization makes it difficult to assess the nature of the issue and take appropriate action.

Problem:

• Suspicious accounts are only flagged with a general status but are not categorized based on specific behaviors or triggers (e.g., mass repository creation, spammy behavior, automated activity patterns, etc.).

• This results in a lack of clarity for both users and administrators when reviewing flagged accounts.

Expected Behavior:

Suspicious accounts should be categorized into predefined types such as:

• Spam Behavior: Accounts exhibiting repetitive or low-quality content.
• Automated Activity: Accounts showing signs of bot-like behavior.
• Mass Repository Creation: Accounts creating multiple repositories with autogenerated or meaningless names.
• Other Suspicious Patterns: Any other patterns not falling under the above categories.
  Clear labels or tags should be visible to reviewers or administrators to provide additional context for flagged accounts.

Actual Behavior:

All flagged accounts are treated the same, with no differentiation between the reasons they were flagged.

Proposed Solution:

1. Implement a categorization system for suspicious accounts based on the detection algorithm's triggers.
2. Display the categorization labels in the account’s status or admin panel for better clarity.
3. Provide an explanation or guide for administrators and users on the different categories and what they signify.

Benefits:

• Improved transparency in the flagging process.
• Enhanced ability to prioritize and address issues based on the severity and nature of the suspicious activity.
• Reduction in false positives by enabling a more nuanced review process.

Additional Context

Categorizing suspicious accounts will help maintain the integrity of the platform and improve the user experience for both administrators and legitimate users.

[BearHuddleston]

We can add tags or meta tags. I'm thinking of adding barks into a SQL database associated with the heuristic method(s) it triggered.