Dataset Viewer
Auto-converted to Parquet Duplicate
triplet_id
string
entity
string
entity_type
string
relationship
string
target
string
target_type
string
start_date
string
end_date
string
extraction_type
string
ticker
string
year
int64
source_file
string
page_id
string
chunk_id
string
chunk_text
string
triplet_length
int64
chunk_text_length
int64
has_context
bool
ctas_2014_page_1_chunk_1_triplet_1
ctas
ORG
subject_to
sec regulation
REGULATORY_REQUIREMENT
January 2014
default_end_timestamp
default
CTAS
2,014
CTAS_10k_2014.pdf
page_1
chunk_1
## UNITED STATES SECURITIES AND EXCHANGE COMMISSION WASHINGTON, D.C. 20549 FORM 10-K X ANNUAL RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H E SE C U R I T I E S E X C H AN G E AC T O F 1934 For the Fiscal Year E nded M ay 31, 2014 TRANSITION RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H ...
30
3,578
true
ctas_2014_page_1_chunk_1_triplet_2
ctas
ORG
operates_in
united state
GPE
January 2014
default_end_timestamp
default
CTAS
2,014
CTAS_10k_2014.pdf
page_1
chunk_1
## UNITED STATES SECURITIES AND EXCHANGE COMMISSION WASHINGTON, D.C. 20549 FORM 10-K X ANNUAL RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H E SE C U R I T I E S E X C H AN G E AC T O F 1934 For the Fiscal Year E nded M ay 31, 2014 TRANSITION RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H ...
29
3,578
true
ctas_2014_page_1_chunk_1_triplet_3
ctas
ORG
discloses
common stock
FIN_INST
November 2013
default_end_timestamp
default
CTAS
2,014
CTAS_10k_2014.pdf
page_1
chunk_1
## UNITED STATES SECURITIES AND EXCHANGE COMMISSION WASHINGTON, D.C. 20549 FORM 10-K X ANNUAL RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H E SE C U R I T I E S E X C H AN G E AC T O F 1934 For the Fiscal Year E nded M ay 31, 2014 TRANSITION RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H ...
27
3,578
true
ctas_2014_page_1_chunk_1_triplet_4
ctas
ORG
discloses
market value
FIN_METRIC
November 2013
default_end_timestamp
default
CTAS
2,014
CTAS_10k_2014.pdf
page_1
chunk_1
## UNITED STATES SECURITIES AND EXCHANGE COMMISSION WASHINGTON, D.C. 20549 FORM 10-K X ANNUAL RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H E SE C U R I T I E S E X C H AN G E AC T O F 1934 For the Fiscal Year E nded M ay 31, 2014 TRANSITION RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H ...
27
3,578
true
ctas_2014_page_1_chunk_1_triplet_5
ctas
ORG
discloses
share outstanding
FIN_METRIC
June 2014
default_end_timestamp
default
CTAS
2,014
CTAS_10k_2014.pdf
page_1
chunk_1
## UNITED STATES SECURITIES AND EXCHANGE COMMISSION WASHINGTON, D.C. 20549 FORM 10-K X ANNUAL RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H E SE C U R I T I E S E X C H AN G E AC T O F 1934 For the Fiscal Year E nded M ay 31, 2014 TRANSITION RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H ...
32
3,578
true
ctas_2014_page_1_chunk_1_triplet_6
ctas
ORG
discloses
share issue
FIN_METRIC
June 2014
default_end_timestamp
default
CTAS
2,014
CTAS_10k_2014.pdf
page_1
chunk_1
## UNITED STATES SECURITIES AND EXCHANGE COMMISSION WASHINGTON, D.C. 20549 FORM 10-K X ANNUAL RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H E SE C U R I T I E S E X C H AN G E AC T O F 1934 For the Fiscal Year E nded M ay 31, 2014 TRANSITION RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H ...
26
3,578
true
ctas_2014_page_1_chunk_1_triplet_7
ctas
ORG
member_of
nasdaq global select market
FIN_MARKET
January 2014
default_end_timestamp
default
CTAS
2,014
CTAS_10k_2014.pdf
page_1
chunk_1
## UNITED STATES SECURITIES AND EXCHANGE COMMISSION WASHINGTON, D.C. 20549 FORM 10-K X ANNUAL RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H E SE C U R I T I E S E X C H AN G E AC T O F 1934 For the Fiscal Year E nded M ay 31, 2014 TRANSITION RE P OR T P U R SU AN T T O SE C T I O N 13 O R 15(d) O F T H ...
42
3,578
true
ctas_2014_page_3_chunk_2_triplet_1
ctas
ORG
discloses
rental uniform & ancillary prod .
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
48
509
true
ctas_2014_page_3_chunk_2_triplet_2
ctas
ORG
discloses
uniform direct sale
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
34
509
true
ctas_2014_page_3_chunk_2_triplet_3
ctas
ORG
discloses
first aid & safety serv .
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
40
509
true
ctas_2014_page_3_chunk_2_triplet_4
ctas
ORG
discloses
doc mgmt serv .
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
30
509
true
ctas_2014_page_3_chunk_2_triplet_5
ctas
ORG
discloses
total revenue
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
28
509
true
ctas_2014_page_3_chunk_2_triplet_6
ctas
ORG
increase
rental uniform & ancillary prod .
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
47
509
true
ctas_2014_page_3_chunk_2_triplet_7
ctas
ORG
decrease
uniform direct sale
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
33
509
true
ctas_2014_page_3_chunk_2_triplet_8
ctas
ORG
increase
first aid & safety serv .
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
39
509
true
ctas_2014_page_3_chunk_2_triplet_9
ctas
ORG
increase
doc mgmt serv .
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
29
509
true
ctas_2014_page_3_chunk_2_triplet_10
ctas
ORG
increase
total revenue
FIN_METRIC
May 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_2
| Fiscal Year EndedMay 31, (in thousands) | 2014 | 2013 | 2012 | |------------------------------------------------|-------------|-------------|-------------| | Rental Uniforms and Ancillary Products | $ 3,223,930 | $ 3,044,587 | $ 2,912,261 | | Uniform Direct Sales | 455,485 | 461,328 | 433,994 | | First Aid, Safety an...
27
509
true
ctas_2014_page_8_chunk_1_triplet_1
ctas
ORG
impacted_by
cybersecurity breach
RISK_FACTOR
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
37
5,206
true
ctas_2014_page_8_chunk_1_triplet_2
ctas
ORG
negatively_impacts
financial loss
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
38
5,206
true
ctas_2014_page_8_chunk_1_triplet_3
ctas
ORG
negatively_impacts
brand reputation
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
40
5,206
true
ctas_2014_page_8_chunk_1_triplet_4
ctas
ORG
negatively_impacts
business performance
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
44
5,206
true
ctas_2014_page_8_chunk_1_triplet_5
ctas
ORG
negatively_impacts
liability
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
33
5,206
true
ctas_2014_page_8_chunk_1_triplet_6
ctas
ORG
depends_on
internal control
ACCOUNTING_POLICY
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
32
5,206
true
ctas_2014_page_8_chunk_1_triplet_7
ctas
ORG
negatively_impacts
reliable financial report
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
49
5,206
true
ctas_2014_page_8_chunk_1_triplet_8
ctas
ORG
negatively_impacts
stock price
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
35
5,206
true
ctas_2014_page_8_chunk_1_triplet_9
ctas
ORG
depends_on
key employee
PERSON
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
28
5,206
true
ctas_2014_page_8_chunk_1_triplet_10
ctas
ORG
negatively_impacts
strategic objective
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
43
5,206
true
ctas_2014_page_8_chunk_1_triplet_11
ctas
ORG
impacted_by
unexpected event
RISK_FACTOR
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
33
5,206
true
ctas_2014_page_8_chunk_1_triplet_12
ctas
ORG
negatively_impacts
operational efficiency
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
46
5,206
true
ctas_2014_page_8_chunk_1_triplet_13
ctas
ORG
negatively_impacts
result of operation
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
43
5,206
true
ctas_2014_page_8_chunk_1_triplet_14
ctas
ORG
recognizes
impairment charge
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
33
5,206
true
ctas_2014_page_8_chunk_1_triplet_15
ctas
ORG
negatively_impacts
financial condition
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
43
5,206
true
ctas_2014_page_8_chunk_1_triplet_16
ctas
ORG
negatively_impacts
operating cash flow
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
43
5,206
true
ctas_2014_page_8_chunk_1_triplet_17
ctas
ORG
depends_on
economic condition
MACRO_CONDITION
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
34
5,206
true
ctas_2014_page_8_chunk_1_triplet_18
ctas
ORG
negatively_impacts
revenue
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
31
5,206
true
ctas_2014_page_8_chunk_1_triplet_19
ctas
ORG
negatively_impacts
capacity utilization
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
44
5,206
true
ctas_2014_page_8_chunk_1_triplet_20
ctas
ORG
negatively_impacts
borrowing rate
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
38
5,206
true
ctas_2014_page_8_chunk_1_triplet_21
ctas
ORG
depends_on
real estate value
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
33
5,206
true
ctas_2014_page_8_chunk_1_triplet_22
ctas
ORG
depends_on
confidential information
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
40
5,206
true
ctas_2014_page_8_chunk_1_triplet_23
ctas
ORG
negatively_impacts
customer trust
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
38
5,206
true
ctas_2014_page_8_chunk_1_triplet_24
ctas
ORG
negatively_impacts
legal liability
FIN_METRIC
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_8
chunk_1
that could lead to disruptions in systems, unauthorized release of confidential or otherwise protected information and corruption of data. If the network of security controls, policy enforcement mechanisms and monitoring systems to address these threats to our technology fails, the compromising of confidential or other...
39
5,206
true
ctas_2014_page_3_chunk_3_triplet_1
ctas
ORG
discloses
shred revenue
FIN_METRIC
November 2013
October 2014
extracted
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
28
624
true
ctas_2014_page_3_chunk_3_triplet_2
ctas
ORG
operates_in
new york city
GPE
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
30
624
true
ctas_2014_page_3_chunk_3_triplet_3
ctas
ORG
competes_with
abc corp
COMP
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
27
624
true
ctas_2014_page_3_chunk_3_triplet_4
ctas
ORG
competes_with
xyz inc
COMP
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
26
624
true
ctas_2014_page_3_chunk_3_triplet_5
ctas
ORG
competes_with
def ltd
COMP
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
26
624
true
ctas_2014_page_3_chunk_3_triplet_6
ctas
ORG
depends_on
high-quality material
RAW_MATERIAL
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
37
624
true
ctas_2014_page_3_chunk_3_triplet_7
ctas
ORG
depends_on
innovative design
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
33
624
true
ctas_2014_page_3_chunk_3_triplet_8
ctas
ORG
depends_on
competitive pricing
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
35
624
true
ctas_2014_page_3_chunk_3_triplet_9
ctas
ORG
depends_on
customer service
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
32
624
true
ctas_2014_page_3_chunk_3_triplet_10
ctas
ORG
depends_on
user-friendly interface
CONCEPT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_3
chunk_3
(1) Fiscal year 2014 includes only eleven m onths of shredding revenue. Additional information regarding each operating segment is also included in Note 15 entitled Operating Segment Information of "Notes to Consolidated Financial Statements." The primary markets serv ed by all Cintas operating segments are local in na...
39
624
true
ctas_2014_page_7_chunk_1_triplet_1
ctas
ORG
subject_to
usdot regulation
REGULATORY_REQUIREMENT
January 2014
December 2014
default
CTAS
2,014
CTAS_10k_2014.pdf
page_7
chunk_1
Failure to comply w ith federal and state regulations to w hich w e are subject could result in penalties or costs that could adversely affect our results of operations. Our business is subject to complex and stringent state and federal regulations, including employment laws and regulations, minimum wage requirements, ...
32
6,159
true
End of preview. Expand in Data Studio

domyn/FinReflectKG(複製)

データの内容

S&P 500 企業の 2014〜2024 年の 10-K から、LLM による抽出で作った金融の知識グラフのデータ(トリプレット形式)。1 行が 1 件のトリプレットで、entity、relationship、target に加えて、開始日と終了日(start_date、end_date)、ティッカー(ticker)、元の文脈(chunk_text)が入る。関係の語彙は多く、抽出の精度は論文の評価に依る。ライセンスは CC-BY-NC-4.0 で、非営利に限られる。

データの概要

項目 値
行数 17,513,372(1 構成・1 分割の合計)
列数 18
容量 1.67 GB(ファイル 105 本)
ファイルの形式 parquet 103
言語 en
公開 public

構成と分割

構成(config) 分割(split)ごとの行数 列数
default train 17,513,372 18

列(構成 default・分割 train)

列 型 先頭の行の値
triplet_id string ctas_2014_page_1_chunk_1_triplet_1
entity string ctas
entity_type string ORG
relationship string subject_to
target string sec regulation
target_type string REGULATORY_REQUIREMENT
start_date string January 2014
end_date string default_end_timestamp
extraction_type string default
ticker string CTAS
year int64 2014
source_file string CTAS_10k_2014.pdf
page_id string page_1
chunk_id string chunk_1
chunk_text string ## UNITED STATES SECURITIES AND EXCHANGE COMMISS…
triplet_length int64 30
chunk_text_length int64 3578
has_context bool True

行数と列は HF のデータセットビューアの集計(元のデータセットの現在の版で数えたもの)。全データセットの一覧は HYUKI5/dataset-catalog にある。

複製について

domyn/FinReflectKG をリビジョン 81819dfb4c95a96963c9a1848466cdf9612d47c0 の時点でそのまま複製したもの。中身は変えていない。判断モデル Laya(convaiinnovations/laya)の微調整に使う候補として、調査メモ(Laya 学習用データセット調査)に載せたデータを集めている。

出典

論文

アノテーションの方法

データセットの作成

データセットの作成過程、方法論、評価の詳細は、以下の論文を参照すること: FinReflectKG: Agentic Construction and Evaluation of Financial Knowledge Graphs。

微調整に使うときの注意

  • 調査メモの台帳での区分: C(NC・ND・研究専用・評価専用など、商用の利用を妨げうる記載がある)。区分は記載の文言で機械的に付けたもので、法的な判定ではない。学習に入れる前に元のカードとライセンスの原文で確かめる。
  • Laya の入力は 2,048 token まで(ModernBERT-large / mmBERT)。長い本文は先頭から黙って切られるので、切れる行は事前に除くか、要約・切り出しをしてから入れる。
  • 分割は記事・文書単位で行い、同じ本文から作った別の問いを学習側と較正側に分けない。温度と閾値は学習に使っていない分割で別々に当てる。

調査メモの記述

メモ 節 ライセンス(記載) 確認 注記
ontology_kg_entity.md Q4. サプライチェーン・企業間関係の知識グラフ(公開のもの)にどんなデータがあるか (L126) CC-BY-NC-4.0 確認済 確認印なし(各ファイルの凡例により確認済扱い) / 原文注記: NC。LLM による抽出(agentic 構築)で、精度は論文の評価に依る(原典未読)。関係名の語彙が大きく、供給・競合・買収へ畳む対応表が要る

元のカード(日本語訳)

以下は元のデータセットのカードの本文を日本語に訳したもの(機械翻訳)。訳に疑問があるときは、末尾の原文を確かめる。


FinReflectKG のデータセットカード

S&P 500 企業の 2014〜2024 年の 10-K(SEC 提出書類)から抽出した、包括的な金融知識グラフのデータセット。全文の文脈を伴う 17.51 million 件の正規化済みトリプレット を含む。

データセットの詳細

データセットの説明

キュレーター: Domyn

言語: 英語

ライセンス: CC-BY-NC-4.0

FinReflectKG は、規制当局への提出書類から抽出した、金融上の関係、実体、時間情報の構造化された表現を提供する大規模な金融知識グラフのデータセットである。各トリプレットは、時間的な範囲と豊富な文脈情報を伴う、(entity, relationship, target) の形式の構造化された事実を表す。

データセットの出典

論文: FinReflectKG: Agentic Construction and Evaluation of Financial Knowledge Graphs

主な特徴

  • 743 社の S&P 500 企業にわたる 17.51M 件のトリプレット
  • 10 年以上の時間的カバレッジ(2014〜2024 年)
  • 各トリプレットの全文の文脈
  • 正規化された実体と関係
  • 開始日・終了日を伴う時間情報
  • 出典文書の情報を含む豊富なメタデータ

データセットの構造

各行は 1 つのトリプレットを表し、以下のフィールドを持つ。

トリプレットの中核要素

  • triplet_id: 各トリプレットの一意の識別子
  • entity: 固有表現(正規化済み。例: "aapl")
  • entity_type: 実体のカテゴリ(ORG、PERSON、GPE など)
  • relationship: 関係の種類(正規化済み。例: "discloses"、"operates_in")
  • target: 対象の実体(正規化済み)
  • target_type: 対象実体のカテゴリ
  • start_date: 関係の開始日("Month YYYY" 形式)
  • end_date: 関係の終了日("Month YYYY" または "default_end_timestamp")
  • extraction_type: 抽出方法("default" または "extracted")

文書のメタデータ

  • ticker: 企業のティッカーシンボル(例: "aapl"、"msft")
  • year: 提出年
  • source_file: 元の PDF のファイル名
  • page_id: PDF のページ識別子
  • chunk_id: テキストチャンクの識別子

文脈と特徴

  • chunk_text: トリプレットを取り囲む全文の文脈
  • triplet_length: トリプレットのテキスト表現の長さ
  • chunk_text_length: 文脈テキストの長さ
  • has_context: 文脈テキストが利用可能かどうか

時間情報

時間フィールドの説明

FinReflectKG は、関係が生じた時点や有効な時点を捉えるために、高度な時間抽出を用いる。時間情報は次のもので構成される:

開始日と終了日

  • 形式: "Month YYYY"(例: "January 2024"、"March 2023")
  • 目的: 各関係の時間的な有効期間を捉える

既定の時刻

明示的な時間情報を文脈のテキストから確実に抽出できない場合:

  • default_start_timestamp: 明示的な開始日が記述されておらず、推定もできない場合に使う
  • default_end_timestamp: 明示的な終了日が記述されておらず、推定もできない場合に使う

例:

  • 抽出: "Environmental spending was approximately $21 million in fiscal 2014" → start_date: "January 2014", end_date: "December 2014"
  • 既定値: "Cintas depends on multiple suppliers" → start_date: "January 2014"(提出年), end_date: "default_end_timestamp"

抽出種別の分類

extraction_type フィールドは、時間情報の信頼性と出所を示す:

"extracted"

  • 条件: start_date と end_date の両方が、テキストから正しく抽出された場合
  • 例: 特定の会計期間を持つ財務指標、日付の付いた発表、期間が限定された出来事

"default"

  • 条件: start_date または end_date(あるいは両方)が既定値を使っている場合
  • 例: 期間が明示されていない一般的な業務上の関係、継続中の事業活動

使用例

データセットの読み込み

from datasets import load_dataset

# Load the complete dataset
dataset = load_dataset("domyn/FinReflectKG")

# Access the data
triplets = dataset["train"]
print(f"Dataset size: {len(triplets):,} triplets")

# Example triplet
example = triplets[0]
print(f"Entity: {example['entity']}")
print(f"Relationship: {example['relationship']}")  
print(f"Target: {example['target']}")
print(f"Context: {example['chunk_text'][:200]}...")

企業での絞り込み

# Filter for Apple (AAPL) triplets
apple_data = dataset["train"].filter(lambda x: x["ticker"] == "aapl")
print(f"Apple triplets: {len(apple_data):,}")

期間での絞り込み

# Filter for recent years (2022-2024)
recent_data = dataset["train"].filter(lambda x: x["year"] >= 2022)
print(f"Recent triplets: {len(recent_data):,}")

関係の種類での絞り込み

# Find all "discloses" relationships
disclosure_data = dataset["train"].filter(lambda x: x["relationship"] == "discloses")
print(f"Disclosure triplets: {len(disclosure_data):,}")

実体の種類

本データセットには、金融文書に関連する様々な実体の種類が含まれる:

  • ORG: 提出企業(10-K 提出の対象となる上場企業)
  • ORG_GOV: 政府機関(例: United States Government)
  • ORG_REG: 国内外の規制当局(例: SEC, Federal Reserve, ECB)
  • GPE: 地理的な事業やリスクで言及される国、州、都市
  • PERSON: 主要な個人(例: CEO, CFO)
  • COMP: 競合、サプライヤー、顧客、パートナーを含む、提出書類で言及される外部企業
  • PRODUCT: 製品またはサービス(例: iPhone, AWS)
  • EVENT: パンデミック、自然災害、M&A などの重要な出来事
  • SECTOR: 提出企業(ORG)または COMP の実体に関連する分野や業界(例: Technology, Healthcare, Financials)
  • ECON_IND: 経済の動向や状況を反映する定量的な指標(例: Inflation Rate, GDP Growth, Unemployment Rate, Interest Rate, Consumer Confidence Index)
  • FIN_INST: 取引可能な金融資産または負債(例: bonds, derivatives, options)
  • FIN_MARKET: 金融指数と市場の動向(例: S&P 500, Dow Jones)
  • CONCEPT: 抽象的な概念や理論(例: Artificial Intelligence, Digital Transformation, Circular Economy)
  • RAW_MATERIAL: 必須の原材料(例: Lithium)
  • LOGISTICS: サプライチェーンと物流の実体(例: Ports)
  • ACCOUNTING_POLICY: 主要な会計方針(例: revenue recognition, lease accounting, goodwill impairment)
  • RISK_FACTOR: 文書化されたリスク(例: market risk, supply chain risk, cybersecurity risk, geopolitical risk)
  • LITIGATION: 10-K に開示された訴訟、規制当局の調査、和解を含む、法的紛争や手続き
  • SEGMENT: 提出企業(ORG)の内部部門または事業セグメント(例: Cloud segment, North America retail)
  • FIN_METRIC: 財務指標や数値(例: Net Income, EBITDA, Long-Term Debt, CapEx, R&D Expense)
  • ESG_TOPIC: 環境・社会・ガバナンスのテーマ(例: Carbon Emissions, DEI, Renewable Energy, Climate Risk)
  • MACRO_CONDITION: 企業や業界に影響する、定性的またはテーマ的なマクロ経済の動向(例: Recession, Inflationary Pressures, Tightening Monetary Policy, Economic Uncertainty, Labor Shortages)
  • REGULATORY_REQUIREMENT: 特定の規制や法的枠組み(例: Basel III, SEC rules, GDPR)
  • COMMENTARY: 経営陣による声明や開示(例: outlooks, explanations, guidance)

関係の種類

本データセットには、金融知識グラフのための包括的な関係の種類が含まれる:

  • Has_Stake_In: 全部または一部の所有や持分を示す(例: ORG owns SEGMENT or has stake in COMP)
  • Announces: 公に開示または発表する(例: ORG announces PRODUCT launch or ESG_TOPIC initiative)
  • Operates_In: 事業の地理的範囲または市場での存在を示す(例: ORG operates in GPE)
  • Introduces: 新しい製品、方針、セグメントを展開または導入する(例: ORG introduces PRODUCT or ACCOUNTING_POLICY)
  • Produces: 製品またはサービスを製造または開発する(例: ORG produces PRODUCT)
  • Regulates: 管理または規制上の監督を行使する(例: ORG_REG regulates ORG)
  • Involved_In: 合併、買収、訴訟などの出来事への直接の関与を示す
  • Impacted_By: 主要な出来事によってエンティティが重大な影響を受けたことを示す(例: Amazon impacted_by COVID-19)
  • Impacts: 財務業績、市場動向、その他の主要な成果に対して、エンティティや出来事が与える広範な影響や効果を示す
  • Positively_Impacts: 好ましい成果への寄与を示す(例: ESG_TOPIC positively impacts FIN_METRIC)
  • Negatively_Impacts: 好ましくない成果への寄与を示す(例: RISK_FACTOR negatively impacts FIN_METRIC)
  • Related_To: 関連や結びつきを示す
  • Member_Of: 正式な所属やグループへの加入を示す(例: COMP is member of FIN_MARKET index)
  • Invests_In: 金融資本または戦略資本を配分する(例: ORG invests in COMP or ESG_TOPIC)
  • Increases: 価値や活動の増加を示す(例: ORG increases CapEx or debt)
  • Decreases: 価値や活動の減少を示す(例: ORG decreases headcount or R&D spend)
  • Depends_On: 別のエンティティへの支えや依存を必要とすることを示す(例: ORG depends on RAW_MATERIAL or COMP)
  • Causes_Shortage_Of: 出来事や状況によって引き起こされる供給不足を示す(例: EVENT causes shortage of RAW_MATERIAL)
  • Affects_Stock: 株価や評価額への直接的な影響を示す(例: RISK_FACTOR affects_stock of ORG)
  • Stock_Decline_Due_To: 株価下落の原因となる要因を示す(例: MACRO_CONDITION stock_decline_due_to ORG)
  • Stock_Rise_Due_To: 株価上昇の原因となる要因を示す(例: ESG_TOPIC stock_rise_due_to ORG)
  • Market_Reacts_To: 外部の出来事や開示に対する市場の反応を示す(例: FIN_MARKET market_react_to EVENT)
  • Discloses: 明らかにする、または報告する(例: ORG discloses FIN_METRIC, ESG_TOPIC, or RISK_FACTOR)
  • Faces: 法的または規制上の課題に直面する(例: ORG faces LITIGATION or REGULATORY_REQUIREMENT)
  • Guides_On: 経営陣によるコメントや見通しを提供する(例: COMMENTARY guides_on FIN_METRIC or MACRO_CONDITION)
  • Complies_With: 規制や方針の要件を満たす(例: ORG complies_with REGULATORY_REQUIREMENT)
  • Subject_To: エンティティが統治される、または影響を受けることを示す(例: ORG subject_to ACCOUNTING_POLICY or REGULATORY_REQUIREMENT)
  • Supplies: ベンダーまたはサプライヤーの関係を示す(例: COMP supplies RAW_MATERIAL to ORG)
  • Partners_With: 正式または戦略的な協業を示す(例: ORG partners_with COMP for PRODUCT)

データ品質

本データセットには、広範なクリーニングと検証を施している:

  • 正しい "Month YYYY" 形式の整形済みの日付が 99.08%
  • 見出し語化(lemmatization)を用いて実体と関係を正規化
  • 重複を除去したトリプレット
  • 無効なデータを除外
  • データ構造と整合性の包括的な検証

時間的カバレッジ

📊 総トリプレット数: 17,513,372
📅 年の範囲: 2014-2024
🏢 企業数: 743

年の範囲 トリプレット数 正確な件数 企業数
2014-2018 7.55M 7,549,552 743
2019-2021 5.04M 5,043,004 743
2022-2024 4.92M 4,920,816 743
合計 17.51M 17,513,372 743

用途

本データセットは次の用途に適している:

研究用途

  • 金融 NLP: 固有表現認識、関係抽出
  • 知識グラフの構築: 金融の知識ベースの構築
  • 時系列分析: 金融上の関係を時間とともに研究すること
  • リスク評価: リスク要因とその推移の分析
  • コンプライアンス研究: 規制上の関係の理解

産業界での用途

  • 金融インテリジェンス: SEC 提出書類の自動分析
  • デューデリジェンス: 企業間の関係の包括的なマッピング
  • ESG 分析: 環境、社会、ガバナンスに関する知見
  • 市場調査: 競争環境の把握
  • レグテック: コンプライアンスとリスクのモニタリング

データの出典

本データセットは、SEC に提出された S&P 500 企業の 10-K 年次報告書に由来する。

倫理上の配慮

  • 公開データ: 元データはすべて公開されている SEC 提出書類である
  • 個人情報なし: 企業と金融の実体に焦点を当てる
  • 規制への準拠: SEC の開示要件を尊重する
  • 研究利用: 学術・研究目的を想定している

引用

研究で本データセットを使う場合は、次を引用すること。

@article{arun2025finreflectkg,
  title={FinReflectKG: Agentic Construction and Evaluation of Financial Knowledge Graphs},
  author={Arun, Abhinav and Dimino, Fabrizio and Agarwal, Tejas Prakash and Sarmah, Bhaskarjit and Pasquali, Stefano},
  journal={arXiv preprint arXiv:2508.17906},
  year={2025},
  url={https://arxiv.org/pdf/2508.17906}
}

データセットの作成

データセットの作成過程、方法論、評価の詳細は、以下の論文を参照すること: FinReflectKG: Agentic Construction and Evaluation of Financial Knowledge Graphs。

データセットカードの連絡先

Reetu Raj Harsh (reeturaj.harsh@domyn.com)

原文(英語)

Dataset Card for FinReflectKG

A comprehensive financial knowledge graph dataset extracted from S&P 500 companies' 10-K SEC filings spanning 2014-2024, containing 17.51 million normalized triplets with full textual context.

Dataset Details

Dataset Description

Curated by: Domyn

Language(s): English

License: CC-BY-NC-4.0

FinReflectKG is a large-scale financial knowledge graph dataset that provides structured representations of financial relationships, entities, and temporal information extracted from regulatory filings. Each triplet represents a structured fact in the format (entity, relationship, target) with temporal bounds and rich contextual information.

Dataset Sources

Paper: FinReflectKG: Agentic Construction and Evaluation of Financial Knowledge Graphs

Key Features

  • 17.51M triplets across 743 S&P 500 companies
  • 10+ years of temporal coverage (2014-2024)
  • Full text context for each triplet
  • Normalized entities and relationships
  • Temporal information with start/end dates
  • Rich metadata including source document information

Dataset Structure

Each row contains a single triplet with the following fields:

Core Triplet Components

  • triplet_id: Unique identifier for each triplet
  • entity: Named entity (normalized, e.g., "aapl")
  • entity_type: Entity category (ORG, PERSON, GPE, etc.)
  • relationship: Relationship type (normalized, e.g., "discloses", "operates_in")
  • target: Target entity (normalized)
  • target_type: Target entity category
  • start_date: Relationship start date (Month YYYY format)
  • end_date: Relationship end date (Month YYYY or "default_end_timestamp")
  • extraction_type: Extraction method ("default" or "extracted")

Document Metadata

  • ticker: Company ticker symbol (e.g., "aapl", "msft")
  • year: Filing year
  • source_file: Original PDF filename
  • page_id: PDF page identifier
  • chunk_id: Text chunk identifier

Context & Features

  • chunk_text: Full text context surrounding the triplet
  • triplet_length: Length of triplet text representation
  • chunk_text_length: Length of context text
  • has_context: Whether contextual text is available

Temporal Information

Temporal Fields Explained

FinReflectKG employs sophisticated temporal extraction to capture when relationships occur or are valid. The temporal information consists of:

Start Date & End Date

  • Format: "Month YYYY" (e.g., "January 2024", "March 2023")
  • Purpose: Captures the temporal validity period of each relationship

Default Timestamps

When explicit temporal information cannot be reliably extracted from the text context:

  • default_start_timestamp: Used when no explicit start date is mentioned or inferable
  • default_end_timestamp: Used when no explicit end date is mentioned or inferable

Examples:

  • Extracted: "Environmental spending was approximately $21 million in fiscal 2014" → start_date: "January 2014", end_date: "December 2014"
  • Default: "Cintas depends on multiple suppliers" → start_date: "January 2014" (filing year), end_date: "default_end_timestamp"

Extraction Type Classification

The extraction_type field indicates the reliability and source of temporal information:

"extracted"

  • When: Both start_date AND end_date are successfully extracted from the text
  • Example: Financial metrics with specific fiscal periods, dated announcements, time-bounded events

"default"

  • When: Either start_date OR end_date (or both) uses default values
  • Example: General operational relationships, ongoing business activities without specified timeframes

Example Usage

Loading the Dataset

from datasets import load_dataset

# Load the complete dataset
dataset = load_dataset("domyn/FinReflectKG")

# Access the data
triplets = dataset["train"]
print(f"Dataset size: {len(triplets):,} triplets")

# Example triplet
example = triplets[0]
print(f"Entity: {example['entity']}")
print(f"Relationship: {example['relationship']}")  
print(f"Target: {example['target']}")
print(f"Context: {example['chunk_text'][:200]}...")

Filtering by Company

# Filter for Apple (AAPL) triplets
apple_data = dataset["train"].filter(lambda x: x["ticker"] == "aapl")
print(f"Apple triplets: {len(apple_data):,}")

Filtering by Time Period

# Filter for recent years (2022-2024)
recent_data = dataset["train"].filter(lambda x: x["year"] >= 2022)
print(f"Recent triplets: {len(recent_data):,}")

Filtering by Relationship Type

# Find all "discloses" relationships
disclosure_data = dataset["train"].filter(lambda x: x["relationship"] == "discloses")
print(f"Disclosure triplets: {len(disclosure_data):,}")

Entity Types

The dataset includes various entity types relevant to financial documents:

  • ORG: Filing Company (The public company that is the subject of the 10-K filing)
  • ORG_GOV: Government bodies (e.g., United States Government)
  • ORG_REG: Domestic or international regulatory bodies (e.g., SEC, Federal Reserve, ECB)
  • GPE: Countries, states, or cities mentioned in geographic operations or risks
  • PERSON: Key individuals (e.g., CEO, CFO)
  • COMP: External companies referenced in the filing, including competitors, suppliers, customers, or partners
  • PRODUCT: Products or services (e.g., iPhone, AWS)
  • EVENT: Material events such as pandemics, natural disasters, M&A events
  • SECTOR: Sectors or industries relevant to the filer ORG or COMP entities (e.g., Technology, Healthcare, Financials)
  • ECON_IND: Quantitative metrics that reflect economic trends or conditions (e.g., Inflation Rate, GDP Growth, Unemployment Rate, Interest Rate, Consumer Confidence Index)
  • FIN_INST: Tradable financial assets or liabilities (e.g., bonds, derivatives, options)
  • FIN_MARKET: Financial indices and market dynamics (e.g., S&P 500, Dow Jones)
  • CONCEPT: Abstract concepts or theories (e.g., Artificial Intelligence, Digital Transformation, Circular Economy)
  • RAW_MATERIAL: Essential raw materials (e.g., Lithium)
  • LOGISTICS: Supply chain and logistics entities (e.g., Ports)
  • ACCOUNTING_POLICY: Key accounting policies (e.g., revenue recognition, lease accounting, goodwill impairment)
  • RISK_FACTOR: Documented risks (e.g., market risk, supply chain risk, cybersecurity risk, geopolitical risk)
  • LITIGATION: Legal disputes or proceedings including lawsuits, regulatory investigations, or settlements disclosed in the 10-K
  • SEGMENT: Internal divisions or business segments of the filer ORG (e.g., Cloud segment, North America retail)
  • FIN_METRIC: Financial metrics or values (e.g., Net Income, EBITDA, Long-Term Debt, CapEx, R&D Expense)
  • ESG_TOPIC: Environmental, Social, and Governance themes (e.g., Carbon Emissions, DEI, Renewable Energy, Climate Risk)
  • MACRO_CONDITION: Qualitative or thematic macroeconomic trends that affect the company or industry (e.g., Recession, Inflationary Pressures, Tightening Monetary Policy, Economic Uncertainty, Labor Shortages)
  • REGULATORY_REQUIREMENT: Specific regulations or legal frameworks (e.g., Basel III, SEC rules, GDPR)
  • COMMENTARY: Statements or disclosures from company management (e.g., outlooks, explanations, guidance)

Relationship Types

The dataset includes comprehensive relationship types for financial knowledge graphs:

  • Has_Stake_In: Indicates full or partial ownership or equity interest (e.g., ORG owns SEGMENT or has stake in COMP)
  • Announces: Publicly discloses or communicates (e.g., ORG announces PRODUCT launch or ESG_TOPIC initiative)
  • Operates_In: Indicates operational geography or market presence (e.g., ORG operates in GPE)
  • Introduces: Rolls out or implements a new product, policy, or segment (e.g., ORG introduces PRODUCT or ACCOUNTING_POLICY)
  • Produces: Manufactures or develops a product or service (e.g., ORG produces PRODUCT)
  • Regulates: Exerts control or regulatory oversight (e.g., ORG_REG regulates ORG)
  • Involved_In: Specifies direct involvement in an event such as a merger, acquisition, or litigation
  • Impacted_By: Indicates that the entity was materially affected by a major event (e.g., Amazon impacted_by COVID-19)
  • Impacts: Specifies the broad influence or effect an entity or event has on financial performance, market trends, or other key outcomes
  • Positively_Impacts: Contributes to positive outcomes (e.g., ESG_TOPIC positively impacts FIN_METRIC)
  • Negatively_Impacts: Contributes to adverse outcomes (e.g., RISK_FACTOR negatively impacts FIN_METRIC)
  • Related_To: Indicates a connection or relationship
  • Member_Of: Indicates formal affiliation or group membership (e.g., COMP is member of FIN_MARKET index)
  • Invests_In: Allocates financial or strategic capital (e.g., ORG invests in COMP or ESG_TOPIC)
  • Increases: Denotes a growth or rise in value or activity (e.g., ORG increases CapEx or debt)
  • Decreases: Denotes a decline in value or activity (e.g., ORG decreases headcount or R&D spend)
  • Depends_On: Requires support or shows reliance on another entity (e.g., ORG depends on RAW_MATERIAL or COMP)
  • Causes_Shortage_Of: Indicates supply constraint driven by an event or condition (e.g., EVENT causes shortage of RAW_MATERIAL)
  • Affects_Stock: Indicates direct influence on stock price or valuation (e.g., RISK_FACTOR affects_stock of ORG)
  • Stock_Decline_Due_To: Specifies factor causing drop in stock price (e.g., MACRO_CONDITION stock_decline_due_to ORG)
  • Stock_Rise_Due_To: Specifies factor causing increase in stock price (e.g., ESG_TOPIC stock_rise_due_to ORG)
  • Market_Reacts_To: Indicates market response to external events or disclosures (e.g., FIN_MARKET market_react_to EVENT)
  • Discloses: Reveals or reports (e.g., ORG discloses FIN_METRIC, ESG_TOPIC, or RISK_FACTOR)
  • Faces: Encounters legal or regulatory challenges (e.g., ORG faces LITIGATION or REGULATORY_REQUIREMENT)
  • Guides_On: Provides management commentary or forecast (e.g., COMMENTARY guides_on FIN_METRIC or MACRO_CONDITION)
  • Complies_With: Meets regulatory or policy requirements (e.g., ORG complies_with REGULATORY_REQUIREMENT)
  • Subject_To: Indicates entity is governed or affected by (e.g., ORG subject_to ACCOUNTING_POLICY or REGULATORY_REQUIREMENT)
  • Supplies: Indicates vendor or supplier relationship (e.g., COMP supplies RAW_MATERIAL to ORG)
  • Partners_With: Indicates formal or strategic collaboration (e.g., ORG partners_with COMP for PRODUCT)

Data Quality

The dataset has undergone extensive cleaning and validation:

  • 99.08% clean dates in proper "Month YYYY" format
  • Normalized entities and relationships using lemmatization
  • Deduplicated triplets
  • Filtered invalid data
  • Comprehensive validation of data structure and integrity

Temporal Coverage

📊 TOTAL TRIPLETS: 17,513,372
📅 YEAR RANGE: 2014-2024
🏢 COMPANIES: 743

Year Range Triplet Count Exact Count Companies
2014-2018 7.55M 7,549,552 743
2019-2021 5.04M 5,043,004 743
2022-2024 4.92M 4,920,816 743
TOTAL 17.51M 17,513,372 743

Applications

This dataset is suitable for:

Research Applications

  • Financial NLP: Named entity recognition, relation extraction
  • Knowledge Graph Construction: Building financial knowledge bases
  • Temporal Analysis: Studying financial relationships over time
  • Risk Assessment: Analyzing risk factors and their evolution
  • Compliance Research: Understanding regulatory relationships

Industry Applications

  • Financial Intelligence: Automated analysis of SEC filings
  • Due Diligence: Comprehensive company relationship mapping
  • ESG Analysis: Environmental, social, governance insights
  • Market Research: Understanding competitive landscapes
  • Regulatory Technology: Compliance and risk monitoring

Data Source

The dataset is derived from S&P 500 companies' 10-K annual reports filed with the SEC.

Ethical Considerations

  • Public data: All source data is publicly available SEC filings
  • No personal information: Focus on corporate and financial entities
  • Regulatory compliance: Respects SEC disclosure requirements
  • Research use: Intended for academic and research purposes

Citation

If you use this dataset in your research, please cite:

@article{arun2025finreflectkg,
  title={FinReflectKG: Agentic Construction and Evaluation of Financial Knowledge Graphs},
  author={Arun, Abhinav and Dimino, Fabrizio and Agarwal, Tejas Prakash and Sarmah, Bhaskarjit and Pasquali, Stefano},
  journal={arXiv preprint arXiv:2508.17906},
  year={2025},
  url={https://arxiv.org/pdf/2508.17906}
}

Dataset Creation

For detailed information about the dataset creation process, methodology, and evaluation, please refer to the attached paper: FinReflectKG: Agentic Construction and Evaluation of Financial Knowledge Graphs.

Dataset Card Contact

Reetu Raj Harsh (reeturaj.harsh@domyn.com)

Downloads last month
79

Paper for HYUKI5/domyn__FinReflectKG