Admin: Document OCR review UI and docs

2026-09-10T12:42:40Z

Document OCR review UI and docs

← Older revision		Revision as of 12:42, 10 September 2026
Line 2:		Line 2:
	'''Facebook document OCR review queue'''		'''Facebook document OCR review queue'''

	<div class="hatnote">'''''Editors only.''''' This is an internal review aid, not a public directory. It does not change soldier profiles by itself. Prefer [[Contribute documents]] for new uploads.</div>		<div class="hatnote">'''''Editors only.''''' This is an internal review aid, not a public directory. It does not change soldier profiles by itself. Prefer [[Contribute documents]] for new uploads and for seeded 70–79% review jobs.</div>

	This page expands the review queue across progressively lower confidence bands. It is a working aid for human review, not a source of automatically accepted facts.		This page expands the review queue across progressively lower confidence bands. It is a working aid for human review, not a source of automatically accepted facts.
Line 9:		Line 9:
	* Every candidate has a soldier profile link, an exact File link, confidence, a short document/summary cue, and concise evidence where available.		* Every candidate has a soldier profile link, an exact File link, confidence, a short document/summary cue, and concise evidence where available.
	* '''Nothing in the queue below changes a soldier profile.''' Verify the scan and surrounding profile context before making any separate edit.		* '''Nothing in the queue below changes a soldier profile.''' Verify the scan and surrounding profile context before making any separate edit.
			* '''70–79% bridge:''' maintainers can seed that band into the [[Contribute documents]] OCR queue with <code>python scripts/seed_fb_ocr_review_jobs.py</code> (optional <code>--dry-run</code> / <code>--limit</code>). Seeded jobs are review-only until Apply.
	* Generated from <code>import\facebook_posts\enrichments\FB_OCR_REVIEW.json</code> and the preserved applied report <code>import\facebook_posts\enrichments\FB_OCR_APPLIED.json</code>.		* Generated from <code>import\facebook_posts\enrichments\FB_OCR_REVIEW.json</code> and the preserved applied report <code>import\facebook_posts\enrichments\FB_OCR_APPLIED.json</code>.

Admin: Editors-only hatnote for OCR queue

2026-09-02T17:53:52Z

Editors-only hatnote for OCR queue

← Older revision		Revision as of 17:53, 2 September 2026
Line 1:		Line 1:
	{{DISPLAYTITLE:Facebook OCR review}}		{{DISPLAYTITLE:Facebook OCR review}}
	'''Facebook document OCR review queue'''		'''Facebook document OCR review queue'''

			<div class="hatnote">'''''Editors only.''''' This is an internal review aid, not a public directory. It does not change soldier profiles by itself. Prefer [[Contribute documents]] for new uploads.</div>

	This page expands the review queue across progressively lower confidence bands. It is a working aid for human review, not a source of automatically accepted facts.		This page expands the review queue across progressively lower confidence bands. It is a working aid for human review, not a source of automatically accepted facts.

Admin: Expand Facebook OCR review queue with confidence bands

2026-08-28T12:39:15Z

Expand Facebook OCR review queue with confidence bands

← Older revision		Revision as of 12:39, 28 August 2026
Line 606:		Line 606:
	* Review-only candidates include roster-like and weakly attributed documents. Treat lower bands as leads to inspect, not facts to publish.		* Review-only candidates include roster-like and weakly attributed documents. Treat lower bands as leads to inspect, not facts to publish.

	Generated: 2026-08-28 08:37		Generated: 2026-08-28 08:39

Admin: Expand Facebook OCR review queue with confidence bands

2026-08-28T12:37:46Z

Expand Facebook OCR review queue with confidence bands

Show changes

Admin: Expand Facebook OCR review queue with confidence bands

2026-08-28T12:37:05Z

Expand Facebook OCR review queue with confidence bands

Show changes

Admin: OCR apply report and review summary

2026-08-27T03:43:03Z

OCR apply report and review summary

New page

{{DISPLAYTITLE:Facebook OCR review}}
This page summarizes the Facebook gallery '''document OCR''' pass and what was automatically incorporated into soldier profiles.

== Why 472 vs 233? ==
* About '''472''' document-like gallery images were actually run through OCR (earlier estimate was ~462).
* Only '''233''' of those cleared both filters for the review list: document score ≥ 2 '''and''' confidence ≥ 25%.
* Portraits, weak OCR, and low-signal scans stay out of the checklist so review stays focused.
* Full numbered checklist (all 233): kept in the repo as <code>import/facebook_posts/enrichments/FB_OCR_REVIEW.md</code> (also CSV/JSON alongside it).

== Applied this session (confidence ≥ 80%) ==
Updated '''33''' soldier pages. Empty infobox fields only; document dates/cues go under Research notes. Roster-like multi-ASN images were skipped.

{| class="wikitable sortable"
! Soldier !! Fields filled !! Research notes !! Review #s
|-
| [[Atchley, Frank M. (0306244)]] || — || OCR (85%) of `5758974407457766.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie). || 48
|-
| [[Biel, Frederick (32526380)]] || — || OCR (100%) of `5781899001831973.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie). || 1
|-
| [[Bloch, Fritz W. (32967700)]] || — || OCR (90%) of `1646020365419878.jpg` (gallery document). OCR (87%) of `1646017645420150.jpg` (gallery document) — dated 21 November 1945; 28 May 1947 — rank mentioned: Brigadier General. OCR mentioned officer rank Brigadier General (not auto-filled — verify). || 42, 47
|-
| [[Ciluzzi, John J. (12204941)]] || — || OCR (100%) of `5800255109996362.jpg` (DISCHARGE, GOOD CONDUCT) — dated 11 MAY 1943; 25 SEP 1923. OCR (95%) of `5800255119996361.jpg` (gallery document) — dated 15 July 1944. OCR (84%) of `5800255116663028.jpg` (Discharge) — dated 10 JANUARY 1946 — rank mentioned: Major. OCR mentioned officer rank Major (not auto-filled — verify). || 2, 24, 49
|-
| [[Cucuzzella, Adam M. (32621831)]] || — || OCR (94%) of `1756685527686694.jpg` (gallery document). OCR (80%) of `1756685631020017.jpg` (DISCHARGE, GOOD CONDUCT) — dated 1 November 1944. || 31, 67
|-
| [[Eells, Charles B. Jr (01180175)]] || — || OCR (92%) of `5810812955607244.jpg` (Discharge) — rank mentioned: Colonel. OCR mentioned officer rank Colonel (not auto-filled — verify). || 39
|-
| [[Englander, John Louis (0555160)]] || — || OCR (83%) of `968424883179433.jpg` (Military Intelligence Training Center, discharge) — dated 16 June 1944. || 51
|-
| [[Finger, Aaron (32881200)]] || — || OCR (80%) of `5812455455442994.jpg` (gallery document) — dated 5 February 1946. OCR mentioned officer rank Second Lieutenant (not auto-filled — verify). || 68
|-
| [[Freudmann, Felix R. (01184825)]] || — || OCR (80%) of `5704387439583130.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER). || 69
|-
| [[Graf, Ernst (32908365)]] || rank_final=Sergeant || OCR (97%) of `5860556947299511.jpg` (CERTIFICATE OF) — rank mentioned: Sergeant. OCR (92%) of `5860557423966130.jpg` (gallery document). || 19, 40
|-
| [[Hauboldt, Walter H. (36252690)]] || rank_final=Staff Sergeant || OCR (92%) of `5721135984574942.jpg` (MILITARY RECORD, Camp Ritchie, Military Intelligence Training Center) — dated 24 July 1942 — rank mentioned: Staff Sergeant. || 41
|-
| [[Hauser, Frank C. (32417389)]] || — || OCR (98%) of `1906949665993612.jpg` (gallery document) — dated 16 AUG 1944; 4 SEP 1944; 18 SEP 1944. OCR (96%) of `1775827892439124.jpg` (NATURALIZATION, naturalization). OCR (94%) of `1907097249312187.jpg` (gallery document). OCR (94%) of `1907097262645519.jpg` (gallery document). || 17, 22, 32, 33
|-
| [[Hedley, Debonsall (6997907)]] || — || OCR (100%) of `6021299114558626.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie). OCR (99%) of `6021299107891960.jpg` (gallery document) — dated 2 November 1945 — rank mentioned: Sergeant. OCR (97%) of `6021299111225293.jpg` (DISCHARGE, bronze star, Good Conduct) — rank mentioned: Captain. || 3, 15, 20
|-
| [[Hellman, Fred W. (0907226)]] || — || OCR (83%) of `1706678459354068.jpg` (MILITARY INTELLIGENCE TRAINING CENTER) — dated 16 Aug 1944. || 52
|-
| [[Heyl, William C. S. (12135845)]] || — || OCR (94%) of `5750524371636103.jpg` (DISCHARGE). OCR (90%) of `5750524374969436.jpg` (gallery document). || 34, 43
|-
| [[Kaneb, Kenneth A. (31025646)]] || — || OCR (100%) of `6113386158683254.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie). || 4
|-
| [[Kay, John Lewis (01584444)]] || — || OCR (97%) of `3332775190077712.jpg` (Bronze Star) — dated 2 March 1944 — rank mentioned: First Lieutenant. OCR mentioned officer rank First Lieutenant (not auto-filled — verify). || 21
|-
| [[Kessler, Ernest (32611173)]] || — || OCR (83%) of `4386135078075046.jpg` (gallery document) — dated 18 JULY 1944; 6 JUNE 1944 — rank mentioned: Major General. OCR mentioned officer rank Major General (not auto-filled — verify). || 53
|-
| [[Koehnke, Hans F. W. (32002994)]] || rank_final=Technician Fourth Grade || OCR (95%) of `4243732358981986.jpg` (DISCHARGE). OCR (84%) of `4243758552312700.jpg` (gallery document) — dated 4 DECEMBER 1945 — rank mentioned: Technician Fourth Grade. || 25, 50
|-
| [[Krakowski, Meyer (0582821)]] || — || OCR (100%) of `5707496559272218.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie) — dated 18 Sept 1944. OCR (83%) of `5707496525938888.jpg` (gallery document). || 5, 54
|-
| [[Krass, Boris (32542552)]] || — || OCR (95%) of `5744108042277736.jpg` (Discharge) — dated 28 September 1945 — rank mentioned: Colonel. OCR (95%) of `5744120795609794.jpg` (Discharge) — dated 28 September 1945 — rank mentioned: Colonel. OCR (83%) of `5744108048944402.jpg` (DISCHARGE, Good Conduct, CERTIFICATE OF) — dated 1/3/24; 10 Nov 1942. OCR (83%) of `5744120802276460.jpg` (DISCHARGE, Good Conduct, CERTIFICATE OF) — dated 1/3/24; 10 Nov 1942. OCR (82%) of `5744108045611069.jpg` (gallery document) — dated 22 Sep 1945. OCR (82%) of `5744120805609793.jpg` (gallery document) — dated 22 Sep 1945. || 26, 27, 55, 56, 59, 60
|-
| [[Marsh, Louis V. (01056160)]] || — || OCR (88%) of `6072873556067848.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie) — dated 3 March 1944. || 46
|-
| [[Nagel, William (32108533)]] || — || OCR (99%) of `6014047925283745.jpg` (Discharge). OCR (98%) of `6014047935283744.jpg` (DISCHARGE, Camp Ritchie, Good Conduct) — dated 1 November 1944. OCR birth/death cue: 3100. || 16, 18
|-
| [[Neustadt, Hermann (0555150)]] || — || OCR (93%) of `5784574961564377.jpg` (gallery document) — dated 18 Juno 1944. OCR (83%) of `5784574941564379.jpg` (gallery document) — dated 29 JANUAPy 1944. OCR (82%) of `5784574968231043.jpg` (gallery document) — rank mentioned: Private First Class. OCR (81%) of `5784574964897710.jpg` (CERTIFICATE OF) — dated 12 JuL 1944; 1 November 1944. || 38, 57, 61, 63
|-
| [[Perlman, Harry (32614939)]] || — || OCR (100%) of `5999223263432878.jpg` (gallery document) — dated 13 November 1944. OCR (100%) of `5999223276766210.jpg` (MILITARY INTELLIGENCE TRAINING CENTER, CAMP RITCHIE) — dated 6 March 1943; 6 Mar 1943 — rank mentioned: Colonel. OCR (94%) of `5999223273432877.jpg` (gallery document). OCR mentioned officer rank Colonel (not auto-filled — verify). || 6, 7, 35
|-
| [[Sabbadini, Alessandro (32626304)]] || — || OCR (100%) of `1459807756183480.jpg` (gallery document) — dated 28 August 1943 — rank mentioned: Major. OCR mentioned officer rank Major (not auto-filled — verify). || 9
|-
| [[Sajben, Charles Jr (01284221)]] || — || OCR (81%) of `5718812254807315.jpg` (Camp Ritchie). || 64
|-
| [[Shi, Joseph E. B. (34572899)]] || rank_final=Sergeant || OCR (100%) of `5715285408493333.jpg` (BRONZE STAR) — dated 4 February 1944; 30 June 1945; 7 July 1945 — rank mentioned: Sergeant. OCR (100%) of `5715285411826666.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie) — dated 8 May 1944. OCR (100%) of `5725797330775474.jpg` (CAMP RITCHIE, MILITARY INTELLIGENCE TRAINING CENTER, Camp Ritchie) — dated 8 May 1944. OCR (100%) of `5725797334108807.jpg` (BRONZE STAR) — dated 4 February 1944; 30 June 1945; 7 July 1945 — rank mentioned: Sergeant. || 10, 11, 12, 13
|-
| [[Siegel, Julius (36657381)]] || — || OCR (100%) of `4243861458969076.jpg` (BRONZE STAR) — dated 1 December 1944. || 14
|-
| [[Skipper, Harry G. (16051110)]] || — || OCR (95%) of `5747397751948765.jpg` (Discharge) — dated 19 JANUARY 1946 — rank mentioned: Major. OCR (95%) of `5749898245032049.jpg` (Discharge) — dated 19 JANUARY 1946 — rank mentioned: Major. OCR (94%) of `5747397748615432.jpg` (DISCHARGE, GOOD CONDUCT). OCR (94%) of `5749898238365383.jpg` (DISCHARGE, GOOD CONDUCT). OCR mentioned officer rank Major (not auto-filled — verify). || 28, 29, 36, 37
|-
| [[Stark, Werner E. (36598320)]] || — || OCR (89%) of `5797157686972771.jpg` (gallery document). || 44
|-
| [[Weisgall, Hugo D. (13072714)]] || rank_final=Sergeant || OCR (83%) of `5774657545889452.jpg` (Camp Ritchie) — rank mentioned: Sergeant. || 58
|-
| [[Wolf, Edwin II (33619146)]] || — || OCR (96%) of `6070697312952139.jpg` (gallery document) — rank mentioned: Staff Sergeant. OCR (95%) of `6070697292952141.jpg` (Camp Ritchie). OCR (89%) of `6070697302952140.jpg` (gallery document). OCR (81%) of `6070697316285472.jpg` (gallery document) — rank mentioned: Sergeant. OCR (81%) of `6070697332952137.jpg` (gallery document) — rank mentioned: Sergeant. OCR (80%) of `6070697319618805.jpg` (Camp Ritchie). || 23, 30, 45, 65, 66, 70
|}

=== Skipped (manual review) ===
* [[Fleck, Herbert (36048143)]] — roster-like document (many ASNs); left for manual review (items 62)
* [[Riehl, William K. (13043367)]] — roster-like document (many ASNs); left for manual review (items 8)

== How to keep reviewing ==
* Open any soldier link above and check '''Research notes''' / infobox fields.
* To apply more from the full 233-item list, tell the agent e.g. <code>incorporate 40-60</code> or <code>incorporate all ≥70%</code>.
* Default for unmarked checklist items remains '''skip'''.

Generated: 2026-08-26 23:43

Facebook OCR review - Revision history

Admin: Document OCR review UI and docs

Admin: Editors-only hatnote for OCR queue

Admin: Expand Facebook OCR review queue with confidence bands

Admin: Expand Facebook OCR review queue with confidence bands

Admin: Expand Facebook OCR review queue with confidence bands

Admin: OCR apply report and review summary