WIMS Entity Match Report — TM → Company

Total Matches

2,008,094

across all confidence levels

Unique Holders Matched

814,945

17.9% of 4,554,581 total holders

Companies Reached

10,173

74% of 13,748 companies

High-Confidence

274,835

conf ≥ 0.90 · 13.7% of matches

Confidence Distribution

1.0 (exact)

186,141

0.9

89,657

0.8

51,216

0.7

63,465

0.6

619,303

0.5

998,312

⚠ 80.6% of matches sit at conf 0.5–0.6 (fuzzy floor). These are candidates, not confirmed links — AI scoring or manual review required before use.

Match Method

Fuzzy (token sort)

1,737,472

Exact (normalized)

264,818

DBA / Alias

5,804

Matches per Holder

1 company

386,229

2 companies

131,791

3 companies

50,334

4 companies

40,019

5 companies (cap)

205,820

205K holders hit the topN=5 cap — likely over-matching from short/generic names.

Top Countries (Matched Holders)

🇮🇳 India (IN)

444,104

🇺🇸 United States (US)

373,228

🇨🇭 Switzerland (CH)

149,270

🇬🇧 United Kingdom (GB)

138,919

🇨🇳 China (CN)

100,008

🇩🇪 Germany (DE)

98,158

🇦🇺 Australia (AU)

68,379

🇯🇵 Japan (JP)

61,867

Top 20 Companies by Match Count

#	Company	Total Matches	Avg Confidence	Exact	Fuzzy	Quality
1	Izun Pharmaceuticals Corp.	36,656	0.52	5	36,651	⚠ review
2	TFF Pharmaceuticals, Inc.	23,948	0.53	0	23,948	⚠ review
3	Avery Pharmaceuticals Pvt. Ltd.	23,697	0.59	0	23,697	⚠ review
4	Dycine Pharmaceuticals, Limited	20,327	0.58	25	20,302	⚠ review
5	CURE Pharmaceutical Holding Corp.	18,669	0.50	0	18,669	⚠ review
6	Novartis AG	16,913	0.99	16,687	226	✓ high
7	Vivera Pharmaceuticals, Inc.	16,540	0.53	1	16,539	⚠ review
8	UNIVO Pharmaceuticals Ltd.	16,445	0.51	2	16,443	⚠ review
9	Bacfo Pharmaceuticals (India), Ltd.	15,615	0.58	6	15,609	⚠ review
10	Société des Produits Nestlé S.A.	15,480	0.88	14,063	1,417	✓ high
11	Alexza Pharmaceuticals, Inc.	14,880	0.55	23	14,857	⚠ review
12	Midas Care Pharmaceuticals Pvt. Ltd.	14,867	0.58	70	14,797	⚠ review
13	NVE Pharmaceuticals, Inc.	13,303	0.59	0	13,303	⚠ review
14	Axcentria Pharmaceuticals LLC	11,673	0.54	13	11,660	⚠ review
15	Johnson & Johnson, Inc.	11,423	0.81	8,685	2,738	✓ high
16	Love Pharma Inc.	10,743	0.50	0	10,743	⚠ review
17	Ilera Healthcare, LLC	10,727	0.50	0	10,727	⚠ review
18	Wuhan Ruisheng Pharmaceutical Co., Ltd.	10,213	0.56	0	10,213	⚠ review
19	Philip Morris Products S.A.	10,076	0.99	9,764	310	✓ high

⚠ Notable pattern: Most top-ranked companies are pharma-related with avg confidence ~0.50–0.59 and near-zero exact matches. The token "pharmaceuticals" is likely acting as a false-positive magnet in fuzzy matching — consider adding it to a stopword list before the next run.

Sample High-Confidence Matches (conf = 1.0)

Holder Name	Company	Country	Method	Conf
Vitamin Shoppe Industries Inc.	Vitamin Shoppe, Inc.	US	exact	1.00
SANOFI	Sanofi S.A.	FR	exact	1.00
Boehringer Ingelheim International GmbH	Boehringer Ingelheim International GmbH.	DE	exact	1.00
SWISHER INTERNATIONAL, INC.	Swisher International, Inc.	US	exact	1.00
WM. WRIGLEY JR. COMPANY	WM. Wrigley Jr. Company	US	exact	1.00
Actavis Group PTC ehf.	Actavis Group Ptc. Ehf.	IS	exact	1.00
Aspeya Switzerland SA	Aspeya Switzerland, SA	CH	exact	1.00
Reign Beverage Company LLC	Reign Beverage Company, LLC	US	exact	1.00
Stokely-Van Camp Inc.	Stokely-Van Camp, Inc.	US	exact	1.00

Recommended Next Steps

Priority	Action	Impact
🔴 High	Add industry-generic tokens ("pharmaceuticals", "healthcare", "industries") to fuzzy stopwords to eliminate false-positive inflation	Removes ~1M low-quality matches
🔴 High	Raise fuzzy minimum ratio from 70 → 85 for the next pass; run AI scoring (Ollama) on the 0.60–0.74 band	Improves precision of 619K matches
🟡 Medium	Run `match:patents` — same pipeline against 1.5M patent applicants	Extends company linkage coverage
🟡 Medium	Export high-confidence matches (conf ≥ 0.90) for immediate use	274,835 ready-to-use links
🟢 Low	Build review queue UI for the 51K medium-confidence (0.75–0.89) matches	Human validation of borderline cases