feat: stage2/3 pipeline, taxonomy-free cascade, E1.8/E4 experiments, corpus migration state

2026-04-30 04:04:31 +00:00
parent 62b5b5453a
commit 2b9a1782c1
14 changed files with 6145 additions and 5 deletions
@@ -0,0 +1,177 @@
+{
+  "metadata": {
+    "purpose": "E1.4 cascade re-extraction replication (n=30)",
+    "exclusions": "E1's 10 sources",
+    "stratification": {
+      "high": 8,
+      "mid": 8,
+      "low": 8,
+      "document": 6,
+      "document_subtypes": {
+        "academic": 2,
+        "creative": 2,
+        "reference": 2
+      }
+    },
+    "quartile_top": 19,
+    "quartile_bottom": 5
+  },
+  "selected": [
+    {
+      "name": "Claude: Finding ideal rural housing near University of Utah",
+      "entities": 40,
+      "bucket": "high"
+    },
+    {
+      "name": "ChatGPT: Rhino 3D object flow",
+      "entities": 39,
+      "bucket": "high"
+    },
+    {
+      "name": "Claude: Evaluating tenure prospects at R1 universities",
+      "entities": 34,
+      "bucket": "high"
+    },
+    {
+      "name": "Claude: Law enforcement career options",
+      "entities": 34,
+      "bucket": "high"
+    },
+    {
+      "name": "ChatGPT: Resume formatting and review",
+      "entities": 28,
+      "bucket": "high"
+    },
+    {
+      "name": "ChatGPT: Tulsa Concept Album Guide",
+      "entities": 28,
+      "bucket": "high"
+    },
+    {
+      "name": "ChatGPT: SEC coaches with OSU ties",
+      "entities": 27,
+      "bucket": "high"
+    },
+    {
+      "name": "Claude: Lubbock on everything album lyrics",
+      "entities": 26,
+      "bucket": "high"
+    },
+    {
+      "name": "Claude: Bonding ASA 3D printed parts",
+      "entities": 11,
+      "bucket": "mid"
+    },
+    {
+      "name": "Claude: SUNY faculty conflict of interest policies",
+      "entities": 11,
+      "bucket": "mid"
+    },
+    {
+      "name": "Claude: Interview presentation research and preparation",
+      "entities": 11,
+      "bucket": "mid"
+    },
+    {
+      "name": "ChatGPT: Remington 700 5R Gen 1",
+      "entities": 11,
+      "bucket": "mid"
+    },
+    {
+      "name": "Claude: I filling out my annual report...",
+      "entities": 9,
+      "bucket": "mid"
+    },
+    {
+      "name": "Aaron AI: So, I've been working on the RNAI project, and the way I've ...",
+      "entities": 9,
+      "bucket": "mid"
+    },
+    {
+      "name": "Claude: Internship agreement writing help",
+      "entities": 8,
+      "bucket": "mid"
+    },
+    {
+      "name": "ChatGPT: Research Statement Restructure",
+      "entities": 8,
+      "bucket": "mid"
+    },
+    {
+      "name": "ChatGPT: Sink Sprayer Fitting Name",
+      "entities": 5,
+      "bucket": "low"
+    },
+    {
+      "name": "ChatGPT: Title: User request summary.",
+      "entities": 5,
+      "bucket": "low"
+    },
+    {
+      "name": "ChatGPT: Regex for inserting letters",
+      "entities": 5,
+      "bucket": "low"
+    },
+    {
+      "name": "ChatGPT: Provide DXF file",
+      "entities": 5,
+      "bucket": "low"
+    },
+    {
+      "name": "ChatGPT: Rectangle Edge Offset Algorithm",
+      "entities": 3,
+      "bucket": "low"
+    },
+    {
+      "name": "ChatGPT: Testing Vector Alignment",
+      "entities": 3,
+      "bucket": "low"
+    },
+    {
+      "name": "ChatGPT: Scholarship Recommendation Letter Tips",
+      "entities": 3,
+      "bucket": "low"
+    },
+    {
+      "name": "ChatGPT: Respect Individual Interests for Christmas",
+      "entities": 2,
+      "bucket": "low"
+    },
+    {
+      "name": "Nic Oconnor Ind Study S2024 Syllabus.docx",
+      "entities": 14,
+      "subtype": "academic",
+      "bucket": "document"
+    },
+    {
+      "name": "Nic Oconnor Field Work F2023 Syllabus.pdf",
+      "entities": 13,
+      "subtype": "academic",
+      "bucket": "document"
+    },
+    {
+      "name": "Circuits II.pptx",
+      "entities": 11,
+      "subtype": "creative",
+      "bucket": "document"
+    },
+    {
+      "name": "Polar Coordinates.pptx",
+      "entities": 10,
+      "subtype": "creative",
+      "bucket": "document"
+    },
+    {
+      "name": "University of North Texas Cover letter.pdf",
+      "entities": 14,
+      "subtype": "reference",
+      "bucket": "document"
+    },
+    {
+      "name": "New Mexico Cover Letter.docx",
+      "entities": 13,
+      "subtype": "reference",
+      "bucket": "document"
+    }
+  ]
+}