Hugging Face's logo

Datasets:
Pranshurs
/
ylemis-india-pii-benchmark

Tasks:
Token Classification
Languages:
English
Hindi
Tamil
Size:
10K<n<100K
Tags:
pii
privacy
india
hinglish
Synthetic
License:
Dataset card Data Studio Files
xet
 Community
Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
benchmark: struct<name: string, version: string, cases: int64, sha256: string, synthetic_only: bool, contains_r (... 14 chars omitted)
  child 0, name: string
  child 1, version: string
  child 2, cases: int64
  child 3, sha256: string
  child 4, synthetic_only: bool
  child 5, contains_real_pii: bool
scoring: struct<span_offsets: string, primary: string, secondary: string, negative_metrics: list<item: string (... 2 chars omitted)
  child 0, span_offsets: string
  child 1, primary: string
  child 2, secondary: string
  child 3, negative_metrics: list<item: string>
      child 0, item: string
results: list<item: struct<system: string, configuration: string, overlap_precision: double, overlap_recall:  (... 119 chars omitted)
  child 0, item: struct<system: string, configuration: string, overlap_precision: double, overlap_recall: double, ove (... 107 chars omitted)
      child 0, system: string
      child 1, configuration: string
      child 2, overlap_precision: double
      child 3, overlap_recall: double
      child 4, overlap_f1: double
      child 5, exact_f1: double
      child 6, clean_benign_fpr: double
      child 7, hard_negative_fpr: double
      child 8, raw_result: string
limitations: list<item: string>
  child 0, item: string
published_at: timestamp[s]
reproduce_offline: string
text: string
category: string
language: string
entities: list<item: struct<type: string, start: int64, end: int64>>
  child 0, item: struct<type: string, start: int64, end: int64>
      child 0, type: string
      child 1, start: int64
      child 2, end: int64
id: string
to
{'id': Value('string'), 'text': Value('string'), 'entities': List({'type': Value('string'), 'start': Value('int64'), 'end': Value('int64')}), 'language': Value('string'), 'category': Value('string')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1816, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
                  for item in generator(*args, **kwargs):
                              ~~~~~~~~~^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              benchmark: struct<name: string, version: string, cases: int64, sha256: string, synthetic_only: bool, contains_r (... 14 chars omitted)
                child 0, name: string
                child 1, version: string
                child 2, cases: int64
                child 3, sha256: string
                child 4, synthetic_only: bool
                child 5, contains_real_pii: bool
              scoring: struct<span_offsets: string, primary: string, secondary: string, negative_metrics: list<item: string (... 2 chars omitted)
                child 0, span_offsets: string
                child 1, primary: string
                child 2, secondary: string
                child 3, negative_metrics: list<item: string>
                    child 0, item: string
              results: list<item: struct<system: string, configuration: string, overlap_precision: double, overlap_recall:  (... 119 chars omitted)
                child 0, item: struct<system: string, configuration: string, overlap_precision: double, overlap_recall: double, ove (... 107 chars omitted)
                    child 0, system: string
                    child 1, configuration: string
                    child 2, overlap_precision: double
                    child 3, overlap_recall: double
                    child 4, overlap_f1: double
                    child 5, exact_f1: double
                    child 6, clean_benign_fpr: double
                    child 7, hard_negative_fpr: double
                    child 8, raw_result: string
              limitations: list<item: string>
                child 0, item: string
              published_at: timestamp[s]
              reproduce_offline: string
              text: string
              category: string
              language: string
              entities: list<item: struct<type: string, start: int64, end: int64>>
                child 0, item: struct<type: string, start: int64, end: int64>
                    child 0, type: string
                    child 1, start: int64
                    child 2, end: int64
              id: string
              to
              {'id': Value('string'), 'text': Value('string'), 'entities': List({'type': Value('string'), 'start': Value('int64'), 'end': Value('int64')}), 'language': Value('string'), 'category': Value('string')}
              because column names don't match
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1683, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1869, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

id
string
text
string
entities
list
language
string
category
string
ylemis-india-pii-00000
இந்த ஆவணத்தில் தனிப்பட்ட தகவல் எதுவும் இல்லை; பொதுவான வழிகாட்டுதல்கள் மட்டுமே.
[]
ta
benign
ylemis-india-pii-00001
The quarterly report reviews market trends and approves the updated logistics schedule.
[]
en-IN
benign
ylemis-india-pii-00002
Delivery to Sandeep Pillai, 210 Hauz Khas, Ahmedabad 309693; driver assigned vehicle KL49Q2612.
[ { "type": "person", "start": 12, "end": 26 }, { "type": "address", "start": 28, "end": 59 }, { "type": "vehicle_registration", "start": 85, "end": 94 } ]
en-IN
logistics
ylemis-india-pii-00003
केवाईसी में Kabir Saxena, नियोक्ता Deccan Enterprises, पता 776 Gomti Nagar, Ludhiana 125287 और पैन RTWCU4617O दर्ज है।
[ { "type": "person", "start": 12, "end": 24 }, { "type": "organization", "start": 35, "end": 53 }, { "type": "address", "start": 59, "end": 91 }, { "type": "pan", "start": 99, "end": 109 } ]
hi
kyc
ylemis-india-pii-00004
Ayaan Naidu ka naya address 88 Alkapuri, Varanasi 126369 hai, gaadi RJ01WT4862 bhi register karni hai
[ { "type": "person", "start": 0, "end": 11 }, { "type": "address", "start": 28, "end": 56 }, { "type": "vehicle_registration", "start": 68, "end": 78 } ]
hi-Latn
chat
ylemis-india-pii-00005
Account holder Zoya Saxena (gurpreet.rao93@example.com) linked UPI vikram35@oksbi and branch IFSC HDFC0072020.
[ { "type": "person", "start": 15, "end": 26 }, { "type": "email", "start": 28, "end": 54 }, { "type": "upi_id", "start": 67, "end": 81 }, { "type": "ifsc", "start": 98, "end": 109 } ]
en-IN
banking
ylemis-india-pii-00006
Conference room 304 booked 10:00-11:30 on 22/09/2025; agenda item 7, doc ref AB-2231.
[]
en-IN
hard_negative
ylemis-india-pii-00007
The team finalized the roadmap and scheduled the next sprint review for the platform.
[]
en-IN
benign
ylemis-india-pii-00008
GST filing for Pragati Infotech under GSTIN 23JIHAM6145G4ZW; signatory Mary Banerjee, PAN GBIJE2624N.
[ { "type": "organization", "start": 15, "end": 31 }, { "type": "gstin", "start": 44, "end": 59 }, { "type": "person", "start": 71, "end": 84 }, { "type": "pan", "start": 90, "end": 100 } ]
en-IN
tax
ylemis-india-pii-00009
Context: Manish Pillai at Greenfield Technologies, Aadhaar 8311 9898 7066, account policy_5628203_824 — redact before answering.
[ { "type": "person", "start": 9, "end": 22 }, { "type": "organization", "start": 26, "end": 49 }, { "type": "aadhaar", "start": 59, "end": 73 }, { "type": "account_id", "start": 83, "end": 101 } ]
llm-prompt-IN
rag
ylemis-india-pii-00010
Electricity connection for Gurpreet Verma at 383 MG Road, Kolkata 394772; consumer no cust_1795603_166, mobile +91 71474 04791.
[ { "type": "person", "start": 27, "end": 41 }, { "type": "address", "start": 45, "end": 72 }, { "type": "account_id", "start": 86, "end": 102 }, { "type": "indian_phone", "start": 111, "end": 126 } ]
en-IN
utilities
ylemis-india-pii-00011
Applicant Anita Desai submitted Aadhaar 2724 2732 5795 and PAN RSZBA4890J from Karnataka.
[ { "type": "person", "start": 10, "end": 21 }, { "type": "aadhaar", "start": 40, "end": 54 }, { "type": "pan", "start": 63, "end": 73 }, { "type": "location", "start": 79, "end": 88 } ]
en-IN
identity
ylemis-india-pii-00012
কেওয়াইসিতে Aditya Hegde, নিয়োগকর্তা Kaveri Infotech, ঠিকানা 784 Gomti Nagar, Bhopal 476812, প্যান RRRHF7242M।
[ { "type": "person", "start": 12, "end": 24 }, { "type": "organization", "start": 38, "end": 53 }, { "type": "address", "start": 62, "end": 92 }, { "type": "pan", "start": 100, "end": 110 } ]
bn
kyc
ylemis-india-pii-00013
सदस्य Anjali Desai अब Chennai में रहते हैं; खाता profile_3527959_845, ईमेल ayaan.kaur69@example.com।
[ { "type": "person", "start": 6, "end": 18 }, { "type": "location", "start": 22, "end": 29 }, { "type": "account_id", "start": 49, "end": 68 }, { "type": "email", "start": 75, "end": 99 } ]
hi
support
ylemis-india-pii-00014
KYC review for Zoya Sengupta: employer Vajra Private Limited, residence 650 Civil Lines, Lucknow 451015, PAN HVPPO1042L.
[ { "type": "person", "start": 15, "end": 28 }, { "type": "organization", "start": 39, "end": 60 }, { "type": "address", "start": 72, "end": 103 }, { "type": "pan", "start": 109, "end": 119 } ]
en-IN
kyc
ylemis-india-pii-00015
yaar Sunita Bhat ka PAN JHWCH7041B aur UPI abdul89@oksbi galat aa raha hai, theek karo
[ { "type": "person", "start": 5, "end": 16 }, { "type": "pan", "start": 24, "end": 34 }, { "type": "upi_id", "start": 43, "end": 56 } ]
hi-Latn
chat
ylemis-india-pii-00016
मरीज Anjali Gupta का चार्ट MRN-475721-12 और पता 604 Malviya Nagar, Mysuru 271952 दर्ज है; फोन +91 83329 93256।
[ { "type": "person", "start": 5, "end": 17 }, { "type": "medical_id", "start": 27, "end": 40 }, { "type": "address", "start": 48, "end": 80 }, { "type": "indian_phone", "start": 94, "end": 109 } ]
hi
health
ylemis-india-pii-00017
এই প্রতিবেদনে শুধুমাত্র সাধারণ নির্দেশিকা রয়েছে, কোনো ব্যক্তিগত তথ্য নেই।
[]
bn
benign
ylemis-india-pii-00018
{{"naam":"Naina Sharma","shahar":"Hyderabad","pan":"SGKAT0792T","ref":"member_9369831_613"}}
[ { "type": "person", "start": 10, "end": 22 }, { "type": "location", "start": 34, "end": 43 }, { "type": "pan", "start": 52, "end": 62 }, { "type": "account_id", "start": 71, "end": 89 } ]
code-mixed-IN
json
ylemis-india-pii-00019
Shipment tracking 7788123456 cleared customs; container MSKU-9981, seal 44120.
[]
en-IN
hard_negative
ylemis-india-pii-00020
Subsidy for Kabir Mukherjee of Karnataka: Aadhaar 4384 8473 9376, account member_7726279_549, IFSC CNRB0209179.
[ { "type": "person", "start": 12, "end": 27 }, { "type": "location", "start": 31, "end": 40 }, { "type": "aadhaar", "start": 50, "end": 64 }, { "type": "account_id", "start": 74, "end": 92 }, { "type": "ifsc", "start": 99, "end": 110 } ]
en-IN
scheme
ylemis-india-pii-00021
Student Tanvi Sengupta from Chennai enrolled at Yamuna Healthcare; guardian phone +91 76901 81614.
[ { "type": "person", "start": 8, "end": 22 }, { "type": "location", "start": 28, "end": 35 }, { "type": "organization", "start": 48, "end": 65 }, { "type": "indian_phone", "start": 82, "end": 97 } ]
en-IN
education
ylemis-india-pii-00022
Return pickup for Sandeep Singh, 108 Gandhi Nagar, Mumbai 138241; refund to UPI manish14@ybl, phone +91 78581 26290.
[ { "type": "person", "start": 18, "end": 31 }, { "type": "address", "start": 33, "end": 64 }, { "type": "upi_id", "start": 80, "end": 92 }, { "type": "indian_phone", "start": 100, "end": 115 } ]
en-IN
returns
ylemis-india-pii-00023
Aravalli Logistics का जीएसटीआईएन 13IIPFH6716Q5ZY है; हस्ताक्षरकर्ता Ayaan Chowdhury, पैन HYNPQ9228H।
[ { "type": "organization", "start": 0, "end": 18 }, { "type": "gstin", "start": 33, "end": 48 }, { "type": "person", "start": 68, "end": 83 }, { "type": "pan", "start": 89, "end": 99 } ]
hi
tax
ylemis-india-pii-00024
DISCHARGE SUMMARY :: patient Suresh Joshi :: MRN MRN-658814-41 :: addr 2 Sector 17, Lucknow 520166
[ { "type": "person", "start": 29, "end": 41 }, { "type": "medical_id", "start": 49, "end": 62 }, { "type": "address", "start": 71, "end": 98 } ]
ocr-en-IN
ocr
ylemis-india-pii-00025
Order for Sunita Kaur to 760 Sector 17, Lucknow 899385; contact +91 68882 48641, account policy_3108950_458.
[ { "type": "person", "start": 10, "end": 21 }, { "type": "address", "start": 25, "end": 54 }, { "type": "indian_phone", "start": 64, "end": 79 }, { "type": "account_id", "start": 89, "end": 107 } ]
en-IN
ecommerce
ylemis-india-pii-00026
Zoya Sharma का वाहन GJ10C5681 बीमा दावा; पॉलिसी policy_1891050_432, संपर्क +91 85847 97362।
[ { "type": "person", "start": 0, "end": 11 }, { "type": "vehicle_registration", "start": 20, "end": 29 }, { "type": "account_id", "start": 48, "end": 66 }, { "type": "indian_phone", "start": 75, "end": 90 } ]
hi
insurance
ylemis-india-pii-00027
நோயாளி Aman Nair, பதிவு MRN-520261-34, முகவரி 137 Anna Nagar, Lucknow 396324; தொலைபேசி +91 98983 50601.
[ { "type": "person", "start": 7, "end": 16 }, { "type": "medical_id", "start": 24, "end": 37 }, { "type": "address", "start": 46, "end": 76 }, { "type": "indian_phone", "start": 87, "end": 102 } ]
ta
health
ylemis-india-pii-00028
ਨਾਮ: Shreya Patel ਪਤਾ: 446 Shanti Vihar, Tiruchirappalli 277726 ਫ਼ੋਨ: +91 91955 72987
[ { "type": "person", "start": 5, "end": 17 }, { "type": "address", "start": 23, "end": 63 }, { "type": "indian_phone", "start": 70, "end": 85 } ]
pa
form
ylemis-india-pii-00029
નામ: Imran Sinha સરનામું: 847 Salt Lake, Mumbai 678818 ફોન: +91 83303 56844 પાન: HFNPR4448X
[ { "type": "person", "start": 5, "end": 16 }, { "type": "address", "start": 26, "end": 54 }, { "type": "indian_phone", "start": 60, "end": 75 }, { "type": "pan", "start": 81, "end": 91 } ]
gu
form
ylemis-india-pii-00030
SIM activation for Devang Gupta at 907 Anna Nagar, Thiruvananthapuram 745992; alt number +91 63997 93389, email arjun.hegde68@example.com.
[ { "type": "person", "start": 19, "end": 31 }, { "type": "address", "start": 35, "end": 76 }, { "type": "indian_phone", "start": 89, "end": 104 }, { "type": "email", "start": 112, "end": 137 } ]
en-IN
telecom
ylemis-india-pii-00031
Lease for Nikhil Nair at 760 Park Street, Madurai 330550; employer Orchid Ventures, PAN NNZHM7172M.
[ { "type": "person", "start": 10, "end": 21 }, { "type": "address", "start": 25, "end": 56 }, { "type": "organization", "start": 67, "end": 82 }, { "type": "pan", "start": 88, "end": 98 } ]
en-IN
realestate
ylemis-india-pii-00032
bhai Ayaan Shetty ka account policy_668360_373 hai, city Punjab, phone +91 97551 19151 update kardo
[ { "type": "person", "start": 5, "end": 17 }, { "type": "account_id", "start": 29, "end": 46 }, { "type": "location", "start": 57, "end": 63 }, { "type": "indian_phone", "start": 71, "end": 86 } ]
hi-Latn
chat
ylemis-india-pii-00033
नाव: Suresh Mukherjee पत्ता: 487 Park Street, Varanasi 309564 फोन: +91 76843 77589 पॅन: TDGJE3476O
[ { "type": "person", "start": 5, "end": 21 }, { "type": "address", "start": 29, "end": 61 }, { "type": "indian_phone", "start": 67, "end": 82 }, { "type": "pan", "start": 88, "end": 98 } ]
mr
form
ylemis-india-pii-00034
Petitioner Yusuf Iyer, resident of 788 Gandhi Nagar, Pune 785328, holding passport P2921304.
[ { "type": "person", "start": 11, "end": 21 }, { "type": "address", "start": 35, "end": 64 }, { "type": "indian_passport", "start": 83, "end": 91 } ]
en-IN
legal
ylemis-india-pii-00035
Passenger Anita Singh, passport H7278470, contact sunita.nair16@example.com and +91 60284 12929.
[ { "type": "person", "start": 10, "end": 21 }, { "type": "indian_passport", "start": 32, "end": 40 }, { "type": "email", "start": 50, "end": 75 }, { "type": "indian_phone", "start": 80, "end": 95 } ]
en-IN
travel
ylemis-india-pii-00036
नाम/name: Ananya Saxena पता/addr: 723 Viman Nagar, Vadodara 682159 GSTIN: 15IPYJX3078H6ZZ email: mary.mishra93@example.com
[ { "type": "person", "start": 10, "end": 23 }, { "type": "address", "start": 34, "end": 66 }, { "type": "gstin", "start": 74, "end": 89 }, { "type": "email", "start": 97, "end": 122 } ]
mixed-script-IN
form
ylemis-india-pii-00037
গ্রাহক Deepa Gupta এর UPI krishna25@oksbi এবং ফোন +91 92917 57156 যাচাই করা হয়েছে।
[ { "type": "person", "start": 7, "end": 18 }, { "type": "upi_id", "start": 26, "end": 41 }, { "type": "indian_phone", "start": 50, "end": 65 } ]
bn
banking
ylemis-india-pii-00038
வாடிக்கையாளர் Tara Patel இன் UPI sandeep60@okicici, IFSC PUNB0534754 மற்றும் தொலைபேசி +91 79898 40012.
[ { "type": "person", "start": 14, "end": 24 }, { "type": "upi_id", "start": 33, "end": 50 }, { "type": "ifsc", "start": 57, "end": 68 }, { "type": "indian_phone", "start": 86, "end": 101 } ]
ta
banking
ylemis-india-pii-00039
Refund for Arnav Reddy credited; PAN XVHCD9897C, account member_2962842_901, IFSC PUNB0318638.
[ { "type": "person", "start": 11, "end": 22 }, { "type": "pan", "start": 37, "end": 47 }, { "type": "account_id", "start": 57, "end": 75 }, { "type": "ifsc", "start": 82, "end": 93 } ]
en-IN
tax
ylemis-india-pii-00040
Admission of Ishaan Rao to Vajra Logistics, Indore; guardian Aadhaar 4806 0711 8240.
[ { "type": "person", "start": 13, "end": 23 }, { "type": "organization", "start": 27, "end": 42 }, { "type": "location", "start": 44, "end": 50 }, { "type": "aadhaar", "start": 69, "end": 83 } ]
en-IN
education
ylemis-india-pii-00041
Punjab இன் Shreya Sinha க்கு மானியம்; ஆதார் 8303 9195 2148, கணக்கு acct_7161343_768.
[ { "type": "location", "start": 0, "end": 6 }, { "type": "person", "start": 11, "end": 23 }, { "type": "aadhaar", "start": 44, "end": 58 }, { "type": "account_id", "start": 67, "end": 83 } ]
ta
scheme
ylemis-india-pii-00042
Please review the attached policy summary; it contains only general guidelines, no personal details.
[]
en-IN
benign
ylemis-india-pii-00043
Product SKU TZ-99812 and batch B-4471 dispatched; warehouse bay 12, lot 3308.
[]
en-IN
hard_negative
ylemis-india-pii-00044
KYC பதிவில் Lakshmi Pandey, பணியிடம் Sahyadri Infotech, முகவரி 530 Sector 17, Guwahati 428707, பான் KLXTH7160C.
[ { "type": "person", "start": 12, "end": 26 }, { "type": "organization", "start": 37, "end": 54 }, { "type": "address", "start": 63, "end": 93 }, { "type": "pan", "start": 100, "end": 110 } ]
ta
kyc
ylemis-india-pii-00045
আবেদনকারী Kabir Mehta Kerala থেকে আধার 2537 5257 0798 ও ভোটার আইডি MDU6387510 জমা দিয়েছেন।
[ { "type": "person", "start": 10, "end": 21 }, { "type": "location", "start": 22, "end": 28 }, { "type": "aadhaar", "start": 39, "end": 53 }, { "type": "voter_id", "start": 67, "end": 77 } ]
bn
identity
ylemis-india-pii-00046
Joseph Saxena এর ঠিকানা 428 Anna Nagar, Patna 490114 এ বিদ্যুৎ সংযোগ; গ্রাহক নম্বর profile_7437540_897, মোবাইল +91 70569 09482।
[ { "type": "person", "start": 0, "end": 13 }, { "type": "address", "start": 24, "end": 52 }, { "type": "account_id", "start": 83, "end": 102 }, { "type": "indian_phone", "start": 111, "end": 126 } ]
bn
utilities
ylemis-india-pii-00047
ऑर्डर संख्या 5566778 भेज दिया गया; चालान 99120, बॉक्स 14 में पैक किया गया।
[]
hi
hard_negative
ylemis-india-pii-00048
పేరు: Pooja Saxena చిరునామా: 684 Alkapuri, Lucknow 513631 ఫోన్: +91 90780 38347 పాన్: UHJCG8469Q
[ { "type": "person", "start": 6, "end": 18 }, { "type": "address", "start": 29, "end": 57 }, { "type": "indian_phone", "start": 64, "end": 79 }, { "type": "pan", "start": 86, "end": 96 } ]
te
form
ylemis-india-pii-00049
Karnataka के Kavya Saxena को सब्सिडी; आधार 9600 3883 2945, खाता member_6596446_527, आईएफएससी UTIB0813071।
[ { "type": "location", "start": 0, "end": 9 }, { "type": "person", "start": 13, "end": 25 }, { "type": "aadhaar", "start": 43, "end": 57 }, { "type": "account_id", "start": 64, "end": 82 }, { "type": "ifsc", "start": 93, "end": 104 } ]
hi
scheme
ylemis-india-pii-00050
Exit settlement for Yusuf Trivedi at Ganga Solutions: PAN ASSJM6432J, payout to UPI rahul49@paytm.
[ { "type": "person", "start": 20, "end": 33 }, { "type": "organization", "start": 37, "end": 52 }, { "type": "pan", "start": 58, "end": 68 }, { "type": "upi_id", "start": 84, "end": 97 } ]
en-IN
hr_exit
ylemis-india-pii-00051
ಹೆಸರು: Divya Bhat ವಿಳಾಸ: 103 Rajouri Garden, Mumbai 686191 ಫೋನ್: +91 84858 28615
[ { "type": "person", "start": 7, "end": 17 }, { "type": "address", "start": 25, "end": 58 }, { "type": "indian_phone", "start": 65, "end": 80 } ]
kn
form
ylemis-india-pii-00052
{{"customer":"Deepa Jain","aadhaar":"2497 1336 8050","phone":"+91 93446 12178"}}
[ { "type": "person", "start": 14, "end": 24 }, { "type": "aadhaar", "start": 37, "end": 51 }, { "type": "indian_phone", "start": 62, "end": 77 } ]
code-mixed-IN
json
ylemis-india-pii-00053
ग्राहक Lakshmi Bhat का यूपीआई karthik81@oksbi, आईएफएससी BARB0002299 और फोन +91 79303 42149 सत्यापित किया गया।
[ { "type": "person", "start": 7, "end": 19 }, { "type": "upi_id", "start": 30, "end": 45 }, { "type": "ifsc", "start": 56, "end": 67 }, { "type": "indian_phone", "start": 75, "end": 90 } ]
hi
banking
ylemis-india-pii-00054
Loan file policy_1314081_318 for Neha Chowdhury: phone +91 94339 45757, PAN JJVLK7909X, address 795 Brigade Road, Tiruchirappalli 857444.
[ { "type": "account_id", "start": 10, "end": 28 }, { "type": "person", "start": 33, "end": 47 }, { "type": "indian_phone", "start": 55, "end": 70 }, { "type": "pan", "start": 76, "end": 86 }, { "type": "address", "start": 96, "end": 136 } ...
en-IN
banking
ylemis-india-pii-00055
விண்ணப்பதாரர் Neha Mehta Uttarakhand இலிருந்து ஆதார் 9113 5314 4222 சமர்ப்பித்தார்.
[ { "type": "person", "start": 14, "end": 24 }, { "type": "location", "start": 25, "end": 36 }, { "type": "aadhaar", "start": 53, "end": 67 } ]
ta
identity
ylemis-india-pii-00056
आवेदक Tara Das ने Lucknow से आधार 5619 9873 8230 और वोटर आईडी QMF1202883 जमा किया।
[ { "type": "person", "start": 6, "end": 14 }, { "type": "location", "start": 18, "end": 25 }, { "type": "aadhaar", "start": 34, "end": 48 }, { "type": "voter_id", "start": 62, "end": 72 } ]
hi
identity
ylemis-india-pii-00057
Divya Bose के पते 825 Indiranagar, Pune 615967 पर बिजली कनेक्शन; उपभोक्ता संख्या member_4351387_921, मोबाइल +91 81254 13107।
[ { "type": "person", "start": 0, "end": 10 }, { "type": "address", "start": 18, "end": 46 }, { "type": "account_id", "start": 81, "end": 99 }, { "type": "indian_phone", "start": 108, "end": 123 } ]
hi
utilities
ylemis-india-pii-00058
Onboarding Sneha Patil of Kerala; submitted Aadhaar 4348 7463 3824 and voter ID VDP7720518.
[ { "type": "person", "start": 11, "end": 22 }, { "type": "location", "start": 26, "end": 32 }, { "type": "aadhaar", "start": 52, "end": 66 }, { "type": "voter_id", "start": 80, "end": 90 } ]
en-IN
kyc
ylemis-india-pii-00059
Dispute by Pooja Iyer on account profile_5935366_906; PAN ZRKBO6814F, registered email suresh.sharma89@example.com.
[ { "type": "person", "start": 11, "end": 21 }, { "type": "account_id", "start": 33, "end": 52 }, { "type": "pan", "start": 58, "end": 68 }, { "type": "email", "start": 87, "end": 114 } ]
en-IN
dispute
ylemis-india-pii-00060
Payroll: Neha Naidu at Greenfield Enterprises, PAN LJZBP9438O, bank UPI anjali42@oksbi, IFSC IDIB0550779.
[ { "type": "person", "start": 9, "end": 19 }, { "type": "organization", "start": 23, "end": 45 }, { "type": "pan", "start": 51, "end": 61 }, { "type": "upi_id", "start": 72, "end": 86 }, { "type": "ifsc", "start": 93, "end": 104 } ]
en-IN
hr
ylemis-india-pii-00061
Tool result tuple (Ayesha Trivedi, 618 Rajouri Garden, Nashik 333485, RJIHC7974J); never expose in the response.
[ { "type": "person", "start": 19, "end": 33 }, { "type": "address", "start": 35, "end": 68 }, { "type": "pan", "start": 70, "end": 80 } ]
llm-prompt-IN
rag
ylemis-india-pii-00062
Beneficiary Sandeep Saxena of Chennai: Aadhaar 4385 3971 9985, voter ID FDV0073896.
[ { "type": "person", "start": 12, "end": 26 }, { "type": "location", "start": 30, "end": 37 }, { "type": "aadhaar", "start": 47, "end": 61 }, { "type": "voter_id", "start": 72, "end": 82 } ]
en-IN
government
ylemis-india-pii-00063
ID CARD || Kabir Sinha || Dehradun || Aadhaar 2276 3703 9345
[ { "type": "person", "start": 11, "end": 22 }, { "type": "location", "start": 26, "end": 34 }, { "type": "aadhaar", "start": 46, "end": 60 } ]
ocr-en-IN
ocr
ylemis-india-pii-00064
Order #INV-2024-88410 shipped via route 47; support ticket 5523 closed after 3 days.
[]
en-IN
hard_negative
ylemis-india-pii-00065
Lab report for Imran Jain (MRN MRN-485991-42) sent to manish.chauhan45@example.com; clinic in Bihar.
[ { "type": "person", "start": 15, "end": 25 }, { "type": "medical_id", "start": 31, "end": 44 }, { "type": "email", "start": 54, "end": 82 }, { "type": "location", "start": 94, "end": 99 } ]
en-IN
health
ylemis-india-pii-00066
Member Fatima Yadav moved to Surat; profile cust_5760577_161, phone +91 97846 77789.
[ { "type": "person", "start": 7, "end": 19 }, { "type": "location", "start": 29, "end": 34 }, { "type": "account_id", "start": 44, "end": 60 }, { "type": "indian_phone", "start": 68, "end": 83 } ]
en-IN
support
ylemis-india-pii-00067
Claim by Rajesh Sharma for vehicle AP21Y5527; policy policy_9414427_492, contact +91 74836 26569.
[ { "type": "person", "start": 9, "end": 22 }, { "type": "vehicle_registration", "start": 35, "end": 44 }, { "type": "account_id", "start": 53, "end": 71 }, { "type": "indian_phone", "start": 81, "end": 96 } ]
en-IN
insurance
ylemis-india-pii-00068
ग्राहक: Kavya Chowdhury passport: T8102764 phone: +91 99902 82971 shahar: Kolkata
[ { "type": "person", "start": 8, "end": 23 }, { "type": "indian_passport", "start": 34, "end": 42 }, { "type": "indian_phone", "start": 50, "end": 65 }, { "type": "location", "start": 74, "end": 81 } ]
mixed-script-IN
form
ylemis-india-pii-00069
Discharge note: patient Rahul Banerjee, chart MRN-582809-97, residence 170 Indira Colony, Tiruchirappalli 324558, phone +91 64974 66264.
[ { "type": "person", "start": 24, "end": 38 }, { "type": "medical_id", "start": 46, "end": 59 }, { "type": "address", "start": 71, "end": 112 }, { "type": "indian_phone", "start": 120, "end": 135 } ]
en-IN
health
ylemis-india-pii-00070
बैठक दोपहर 3:45 बजे कमरा 204 में है; क्रम संख्या 7788 और चालान 55120 दर्ज करें।
[]
hi
hard_negative
ylemis-india-pii-00071
इस दस्तावेज़ में कोई व्यक्तिगत जानकारी नहीं है, केवल सामान्य नीति विवरण हैं।
[]
hi
benign
ylemis-india-pii-00072
Postpaid plan for Aman Joshi at 499 MG Road, Delhi 853789; number +91 97121 31395, email krishna.pillai69@example.com, account policy_6520056_711.
[ { "type": "person", "start": 18, "end": 28 }, { "type": "address", "start": 32, "end": 57 }, { "type": "indian_phone", "start": 66, "end": 81 }, { "type": "email", "start": 89, "end": 117 }, { "type": "account_id", "start": 127, "end": 145 ...
en-IN
telecom
ylemis-india-pii-00073
Invoice total was Rs 1,24,500 due on 15/08/2025, reference number 9087654321, PO 220045.
[]
en-IN
hard_negative
ylemis-india-pii-00074
इस दस्तावेज़ में कोई व्यक्तिगत जानकारी नहीं है, केवल सामान्य नीति विवरण हैं।
[]
hi
benign
ylemis-india-pii-00075
Arnav Gupta का वाहन TN42J4866 बीमा दावा; पॉलिसी policy_673551_801, संपर्क +91 83296 16977।
[ { "type": "person", "start": 0, "end": 11 }, { "type": "vehicle_registration", "start": 20, "end": 29 }, { "type": "account_id", "start": 48, "end": 65 }, { "type": "indian_phone", "start": 74, "end": 89 } ]
hi
insurance
ylemis-india-pii-00076
Lease for Vikram Mukherjee at 293 Rajouri Garden, Rajkot 852594; employer Vajra Agro, PAN SBMJR4994M.
[ { "type": "person", "start": 10, "end": 26 }, { "type": "address", "start": 30, "end": 63 }, { "type": "organization", "start": 74, "end": 84 }, { "type": "pan", "start": 90, "end": 100 } ]
en-IN
realestate
ylemis-india-pii-00077
வாடிக்கையாளர் Meera Das இன் UPI joseph3@okicici, IFSC BARB0821759 மற்றும் தொலைபேசி +91 73027 93146.
[ { "type": "person", "start": 14, "end": 23 }, { "type": "upi_id", "start": 32, "end": 47 }, { "type": "ifsc", "start": 54, "end": 65 }, { "type": "indian_phone", "start": 83, "end": 98 } ]
ta
banking
ylemis-india-pii-00078
Electricity connection for Sandeep Sengupta at 266 Civil Lines, Patna 686805; consumer no acct_4453390_159, mobile +91 64019 50809.
[ { "type": "person", "start": 27, "end": 43 }, { "type": "address", "start": 47, "end": 76 }, { "type": "account_id", "start": 90, "end": 106 }, { "type": "indian_phone", "start": 115, "end": 130 } ]
en-IN
utilities
ylemis-india-pii-00079
Context: Priya Verma at Ganga Healthcare, Aadhaar 4189 4052 5081, account acct_422409_302 — redact before answering.
[ { "type": "person", "start": 9, "end": 20 }, { "type": "organization", "start": 24, "end": 40 }, { "type": "aadhaar", "start": 50, "end": 64 }, { "type": "account_id", "start": 74, "end": 89 } ]
llm-prompt-IN
rag
ylemis-india-pii-00080
Product SKU TZ-99812 and batch B-4471 dispatched; warehouse bay 12, lot 3308.
[]
en-IN
hard_negative
ylemis-india-pii-00081
Shipment tracking 7788123456 cleared customs; container MSKU-9981, seal 44120.
[]
en-IN
hard_negative
ylemis-india-pii-00082
Delivery to Mary Das, 905 Anna Nagar, Raipur 246559; driver assigned vehicle RJ40U8401.
[ { "type": "person", "start": 12, "end": 20 }, { "type": "address", "start": 22, "end": 51 }, { "type": "vehicle_registration", "start": 77, "end": 86 } ]
en-IN
logistics
ylemis-india-pii-00083
Thiruvananthapuram के Arjun Naidu को सब्सिडी; आधार 7912 5306 4416, खाता policy_2956575_220, आईएफएससी UTIB0239662।
[ { "type": "location", "start": 0, "end": 18 }, { "type": "person", "start": 22, "end": 33 }, { "type": "aadhaar", "start": 51, "end": 65 }, { "type": "account_id", "start": 72, "end": 90 }, { "type": "ifsc", "start": 101, "end": 112 } ]
hi
scheme
ylemis-india-pii-00084
ਨਾਮ: Ritika Gill ਪਤਾ: 644 Gomti Nagar, Vadodara 632642 ਫ਼ੋਨ: +91 74446 92571
[ { "type": "person", "start": 5, "end": 16 }, { "type": "address", "start": 22, "end": 54 }, { "type": "indian_phone", "start": 61, "end": 76 } ]
pa
form
ylemis-india-pii-00085
Loan file member_9286358_460 for Rohit Rao: phone +91 71495 73043, PAN HMULA4911V, address 320 Park Street, Surat 118318.
[ { "type": "account_id", "start": 10, "end": 28 }, { "type": "person", "start": 33, "end": 42 }, { "type": "indian_phone", "start": 50, "end": 65 }, { "type": "pan", "start": 71, "end": 81 }, { "type": "address", "start": 91, "end": 120 } ...
en-IN
banking
ylemis-india-pii-00086
আবেদনকারী Ananya Hegde Maharashtra থেকে আধার 8127 6812 7600 ও ভোটার আইডি GYI5207663 জমা দিয়েছেন।
[ { "type": "person", "start": 10, "end": 22 }, { "type": "location", "start": 23, "end": 34 }, { "type": "aadhaar", "start": 45, "end": 59 }, { "type": "voter_id", "start": 73, "end": 83 } ]
bn
identity
ylemis-india-pii-00087
The team finalized the roadmap and scheduled the next sprint review for the platform.
[]
en-IN
benign
ylemis-india-pii-00088
मरीज Manish Desai का चार्ट MRN-878796-58 और पता 953 Gomti Nagar, Coimbatore 240535 दर्ज है; फोन +91 63472 35288।
[ { "type": "person", "start": 5, "end": 17 }, { "type": "medical_id", "start": 27, "end": 40 }, { "type": "address", "start": 48, "end": 82 }, { "type": "indian_phone", "start": 96, "end": 111 } ]
hi
health
ylemis-india-pii-00089
Pune இன் Rohan Kulkarni க்கு மானியம்; ஆதார் 4489 3902 7920, கணக்கு member_1775447_956.
[ { "type": "location", "start": 0, "end": 4 }, { "type": "person", "start": 9, "end": 23 }, { "type": "aadhaar", "start": 44, "end": 58 }, { "type": "account_id", "start": 67, "end": 85 } ]
ta
scheme
ylemis-india-pii-00090
গ্রাহক Aisha Bhattacharya এর UPI vikram33@paytm এবং ফোন +91 70773 08412 যাচাই করা হয়েছে।
[ { "type": "person", "start": 7, "end": 25 }, { "type": "upi_id", "start": 33, "end": 47 }, { "type": "indian_phone", "start": 56, "end": 71 } ]
bn
banking
ylemis-india-pii-00091
DISCHARGE SUMMARY :: patient Tanvi Naidu :: MRN MRN-330082-36 :: addr 728 Viman Nagar, Kolkata 404793
[ { "type": "person", "start": 29, "end": 40 }, { "type": "medical_id", "start": 48, "end": 61 }, { "type": "address", "start": 70, "end": 101 } ]
ocr-en-IN
ocr
ylemis-india-pii-00092
Order #INV-2024-88410 shipped via route 47; support ticket 5523 closed after 3 days.
[]
en-IN
hard_negative
ylemis-india-pii-00093
ID CARD || Ananya Jain || Madurai || Aadhaar 2243 0290 7546
[ { "type": "person", "start": 11, "end": 22 }, { "type": "location", "start": 26, "end": 33 }, { "type": "aadhaar", "start": 45, "end": 59 } ]
ocr-en-IN
ocr
ylemis-india-pii-00094
কেওয়াইসিতে Divya Trivedi, নিয়োগকর্তা Kaveri Healthcare, ঠিকানা 901 Brigade Road, Nagpur 264765, প্যান NEEHB8306S।
[ { "type": "person", "start": 12, "end": 25 }, { "type": "organization", "start": 39, "end": 56 }, { "type": "address", "start": 65, "end": 96 }, { "type": "pan", "start": 104, "end": 114 } ]
bn
kyc
ylemis-india-pii-00095
The quarterly report reviews market trends and approves the updated logistics schedule.
[]
en-IN
benign
ylemis-india-pii-00096
Please review the attached policy summary; it contains only general guidelines, no personal details.
[]
en-IN
benign
ylemis-india-pii-00097
Passenger Tara Patel, passport P3201219, contact devang.shetty69@example.com and +91 97582 22016.
[ { "type": "person", "start": 10, "end": 20 }, { "type": "indian_passport", "start": 31, "end": 39 }, { "type": "email", "start": 49, "end": 76 }, { "type": "indian_phone", "start": 81, "end": 96 } ]
en-IN
travel
ylemis-india-pii-00098
Exit settlement for Priya Kulkarni at Yamuna Technologies: PAN IJSCF3926Y, payout to UPI pallavi33@ybl.
[ { "type": "person", "start": 20, "end": 34 }, { "type": "organization", "start": 38, "end": 57 }, { "type": "pan", "start": 63, "end": 73 }, { "type": "upi_id", "start": 89, "end": 102 } ]
en-IN
hr_exit
ylemis-india-pii-00099
இந்த ஆவணத்தில் தனிப்பட்ட தகவல் எதுவும் இல்லை; பொதுவான வழிகாட்டுதல்கள் மட்டுமே.
[]
ta
benign
End of preview.

Ylemis India-PII Benchmark v1

An open, reproducible benchmark for span-based PII detection in Indian and multilingual business text. The dataset is fully synthetic and is intended for evaluation, regression testing, and transparent comparison—not model training and not as a substitute for a customer-specific pilot.

Release facts

  • 10,000 cases, all synthetic; no intended real-person PII.
  • 16 entity types and 14 language/format slices.
  • 1,756 negative cases: 810 clean benign and 946 hard negatives.
  • Fixed seed: 20260630.
  • Canonical benchmark.jsonl SHA-256: 7003719c62f451b68ec68280fbccaca12990718ad9ee7aacbcaba66c4c43f914.
  • The file hash was verified against the Kaggle input used by the published Ylemis v5, GLiNER, and Presidio evaluation runs.

Schema

Each JSONL row contains:

{
  "id": "ylemis-india-pii-00000",
  "text": "...",
  "entities": [
    {"type": "pan", "start": 20, "end": 30}
  ],
  "language": "en-IN",
  "category": "kyc"
}

Offsets are zero-based character offsets with an exclusive end.

Entity types

person, address, location, organization, account_id, medical_id, email, indian_phone, pan, aadhaar, gstin, ifsc, upi_id, voter_id, vehicle_registration, and indian_passport.

Language and format slices

English (India), Hindi, Tamil, Bengali, Marathi, Gujarati, Telugu, Punjabi, Kannada, Hinglish/Latin Hindi, mixed-script text, code-mixed JSON, LLM/RAG text, and OCR-style synthetic text.

Published comparison

All rows below were scored on the exact same 10,000-case file and span-scoring implementation. Higher F1 is better; lower false-positive rate is better.

System Overlap P Overlap R Overlap F1 Exact F1 Clean FPR Hard-negative FPR
Ylemis PII Shield v5 FP32, threshold 0.7 99.11% 98.24% 98.68% 96.24% 0.00% 0.00%
GLiNER urchade/gliner_multi_pii-v1, threshold 0.5 93.26% 87.13% 90.09% 82.21% 16.67% 42.81%
Microsoft Presidio default English analyzer 67.31% 43.22% 52.64% 48.40% 50.00% 28.54%

These are benchmark-specific results, not universal claims. The baseline run installed then-current package releases without pinning their versions, so the raw results are published here but should be rerun with pinned packages for a long-term archival leaderboard.

Voter-ID recall on this benchmark is now 100% after a recognizer fix (previously 49.9%), verified on a live-API run. The aggregate score must not be used to hide weak entity slices; full per-entity recall is in results/.

Production artifact qualification

The release-gated Ylemis artifact is FP32 ONNX. On the Kaggle CPU environment, using two threads, batch size one, and including tokenization, it measured:

  • p50: 22.74 ms
  • p95: 31.97 ms
  • p99: 36.91 ms
  • throughput: 41.57 texts/second

PyTorch-to-FP32 span-set and token-argmax agreement were both 100% on the parity set. INT8 was rejected: its span-set agreement against FP32 was 93.16%, below the 98% release gate. See qualification/v5-fp32.json for the raw report.

Reproduce

Offline, no API calls (recommended) — verify our exact score with zero load on the service and no quota used:

python run_benchmark.py --score-file ylemis_v5_live_outputs.jsonl --name ylemis-v5

ylemis_v5_live_outputs.jsonl holds our model's raw predictions on all 10,000 cases (one JSON array of [type,start,end] per line, aligned to benchmark.jsonl). To re-measure the live API yourself instead, call it sequentially (a single stream), not in large parallel bursts:

python run_benchmark.py --system api \
  --endpoint https://api.ylemis.com/pii-shield/v1/scan \
  --api-key "$YLEMIS_PII_SHIELD_KEY" \
  --name ylemis-v5

python run_benchmark.py --system gliner
python run_benchmark.py --system presidio

The API adapter sends {"text":"..."} and expects character spans under entities. Review run_benchmark.py before adapting it to another vendor.

Files

  • benchmark.jsonl: canonical evaluation data.
  • manifest.json: counts, seed, and canonical hash.
  • build_benchmark.py: deterministic generator.
  • run_benchmark.py: reference scoring and adapters.
  • results/: raw leaderboard outputs.
  • qualification/: raw deployment qualification report.
  • leaderboard.json: publication-ready result summary and provenance.

Intended use and limitations

  • Use this as a public, reproducible evaluation floor.
  • Keep a separate locked, audited benchmark for model selection so public test data does not become training data.
  • Synthetic data cannot prove performance on every real customer document.
  • OCR is not evaluated as an image-recognition pipeline. The OCR-style slice is text that simulates OCR noise.
  • Do not publish “best in the world” claims from this benchmark.

License

The benchmark data is released under CC0 1.0. The Ylemis model and weights are not part of this dataset release.
Downloads last month
63
Total file size:
4.52 MB