{% extends "base.html" %} {% set active_page = "false_positives" %} {% block title %}False Positive Profile — IETF Draft Analyzer{% endblock %} {% block extra_head %}{% endblock %} {% block content %}

False Positive Profile

Analysis of {{ data.count }} drafts flagged as false positives — documents that matched AI/agent search keywords but were determined not to be genuinely about AI agent infrastructure.

False Positives
{{ data.count }}
% of All Drafts
{{ data.pct_of_total }}%
% of Rated
{{ data.pct_of_rated }}%
Total Rated
{{ data.total_rated }}
Total Drafts
{{ data.total_drafts }}

Rating Distributions: FP vs Non-FP

Box plots comparing each rating dimension between false positives (red) and genuine AI/agent drafts (blue). Shows what rating patterns distinguish false positives.

False Positives by Source

Which standards bodies produce the most false positives in our search results.

Categories Assigned to False Positives

Categories that the classifier assigned to false positive drafts before they were flagged. Shows which categories are most prone to false matches.

Top Terms in FP Abstracts

Most frequent words in false positive titles and abstracts (stop words excluded). These terms trigger AI/agent keyword matches but appear in unrelated contexts.

All False Positives

Complete list of flagged drafts. Click a name to view details.

{% for fp in data.fp_list %} {% endfor %}
# Draft Title Source Relevance Categories
{{ loop.index }} {{ fp.name | replace('draft-', '') | truncate(40) }} {{ fp.title }} {{ fp.source }} {{ fp.relevance }} {% for cat in fp.categories[:2] %} {{ cat }} {% endfor %}
{% endblock %} {% block extra_scripts %} {% endblock %}