STARXQ/FinTransEvalQA · Datasets at Hugging Face

id int64 1 92	question stringlengths 30 96	expected_answer stringlengths 1 60	code stringlengths 7 155	category stringclasses 15 values	difficulty stringclasses 3 values
1	What is the total number of transactions in the dataset?	38000	len(df)	basic_statistics	easy
2	How many unique clients are in the dataset?	1218	df['client_id'].nunique()	basic_statistics	easy
3	How many unique cards are in the dataset?	3838	df['card_id'].nunique()	basic_statistics	easy
4	What is the average transaction amount?	43.07	df['amount'].mean()	basic_statistics	easy
5	What is the median transaction amount?	29.15	df['amount'].median()	basic_statistics	easy
6	What is the maximum transaction amount?	1773.35	df['amount'].max()	basic_statistics	easy
7	What is the minimum transaction amount?	-498.00	df['amount'].min()	basic_statistics	easy
8	What is the standard deviation of transaction amounts?	81.05	df['amount'].std()	basic_statistics	medium
9	How many transactions are made with Visa cards?	14249	len(df[df['card_brand'] == 'Visa'])	card_analysis	easy
10	How many transactions are made with Mastercard cards?	20405	len(df[df['card_brand'] == 'Mastercard'])	card_analysis	easy
11	Which card brand has the most transactions?	Mastercard	df['card_brand'].value_counts().index[0]	card_analysis	easy
12	What percentage of transactions use Swipe Transactions?	52.50%	(len(df[df['use_chip'] == 'Swipe Transaction']) / len(df) * 100)	card_analysis	medium
13	How many transactions are made with Amex cards?	2409	len(df[df['card_brand'] == 'Amex'])	card_analysis	easy
14	How many unique merchant cities are in the dataset?	3459	df['merchant_city'].nunique()	geographic	easy
15	Which merchant state has the most transactions?	CA	df['merchant_state'].value_counts().index[0]	geographic	easy
16	How many transactions have missing merchant_state information?	4390	df['merchant_state'].isna().sum()	geographic	medium
17	What is the most common merchant city?	ONLINE	df['merchant_city'].value_counts().index[0]	geographic	easy
18	How many transactions are labeled as fraudulent?	27	len(df[df['fraud_label'] == 'Yes'])	fraud_analysis	easy
19	How many transactions are not fraudulent?	25408	len(df[df['fraud_label'] == 'No'])	fraud_analysis	easy
20	What percentage of transactions are fraudulent?	0.11%	(len(df[df['fraud_label'] == 'Yes']) / len(df[df['fraud_label'].notna()]) * 100)	fraud_analysis	medium
21	How many transactions have missing fraud labels?	12565	df['fraud_label'].isna().sum()	fraud_analysis	easy
22	What is the average credit score in the dataset?	713.26	df['credit_score'].mean()	credit_analysis	easy
23	What is the maximum credit score?	850	int(df['credit_score'].max())	credit_analysis	easy
24	What is the minimum credit score?	488	int(df['credit_score'].min())	credit_analysis	easy
25	How many clients have a credit score above 750?	10492	len(df[df['credit_score'] > 750])	credit_analysis	medium
26	What is the average yearly income in the dataset?	46717.33	df['yearly_income'].mean()	income_analysis	easy
27	What is the average per capita income?	24003.13	df['per_capita_income'].mean()	income_analysis	easy
28	What is the maximum yearly income?	280199.00	df['yearly_income'].max()	income_analysis	easy
29	How many clients have yearly income greater than 50000?	11949	len(df[df['yearly_income'] > 50000])	income_analysis	medium
30	What is the average total debt in the dataset?	58032.68	df['total_debt'].mean()	debt_analysis	easy
31	What is the maximum total debt?	461854.00	df['total_debt'].max()	debt_analysis	easy
32	How many clients have total debt greater than 100000?	6866	len(df[df['total_debt'] > 100000])	debt_analysis	medium
33	What is the average credit limit?	15620.43	df['credit_limit'].mean()	credit_limit	easy
34	What is the maximum credit limit?	141391.00	df['credit_limit'].max()	credit_limit	easy
35	How many cards have credit limit of 0?	158	len(df[df['credit_limit'] == 0])	credit_limit	medium
36	What is the average age of clients?	54.09	df['current_age'].mean()	demographics	easy
37	What is the oldest client age?	101	int(df['current_age'].max())	demographics	easy
38	What is the youngest client age?	23	int(df['current_age'].min())	demographics	easy
39	How many clients are over 60 years old?	11986	len(df[df['current_age'] > 60])	demographics	medium
40	How many male clients are in the dataset?	18544	len(df[df['gender'] == 'Male'])	demographics	easy
41	How many female clients are in the dataset?	19456	len(df[df['gender'] == 'Female'])	demographics	easy
42	What is the gender ratio (Male:Female)?	18544:19456	f"{len(df[df['gender'] == 'Male'])}:{len(df[df['gender'] == 'Female'])}"	demographics	medium
43	How many unique merchant categories are in the dataset?	104	df['mcc_description'].nunique()	merchant	easy
44	What is the most common merchant category?	Grocery Stores, Supermarkets	df['mcc_description'].value_counts().index[0]	merchant	easy
45	How many transactions are for Eating Places and Restaurants?	2972	len(df[df['mcc_description'] == 'Eating Places and Restaurants'])	merchant	medium
46	How many Debit card transactions are there?	23789	len(df[df['card_type'] == 'Debit'])	card_analysis	easy
47	How many Credit card transactions are there?	11658	len(df[df['card_type'] == 'Credit'])	card_analysis	easy
48	What is the most common card type?	Debit	df['card_type'].value_counts().index[0]	card_analysis	easy
49	How many Online transactions are there?	4371	len(df[df['use_chip'] == 'Online Transaction'])	transaction_type	easy
50	How many Chip transactions are there?	13680	len(df[df['use_chip'] == 'Chip Transaction'])	transaction_type	easy
51	What percentage of transactions are Online?	11.50%	(len(df[df['use_chip'] == 'Online Transaction']) / len(df) * 100)	transaction_type	medium
52	How many transactions have errors?	614	df['errors'].notna().sum()	error_analysis	easy
53	What percentage of transactions have errors?	1.62%	(df['errors'].notna().sum() / len(df) * 100)	error_analysis	medium
54	What is the most common error type?	Insufficient Balance	df['errors'].value_counts().index[0]	error_analysis	medium
55	What is the earliest transaction date?	2010-01-01	df['transaction_date'].min().split()[0]	temporal	easy
56	What is the latest transaction date?	2019-10-31	df['transaction_date'].max().split()[0]	temporal	easy
57	How many Visa card transactions are fraudulent?	13	len(df[(df['card_brand'] == 'Visa') & (df['fraud_label'] == 'Yes')])	complex_query	medium
58	How many clients have more than 3 credit cards?	21850	len(df[df['num_credit_cards'] > 3])	complex_query	medium
59	How many transactions are made with cards that have chips?	34242	len(df[df['has_chip'] == 'YES'])	card_analysis	easy
60	What percentage of cards have EMV chips?	90.11%	(len(df[df['has_chip'] == 'YES']) / len(df) * 100)	card_analysis	medium
61	How many transactions occurred in Texas (TX)?	2841	len(df[df['merchant_state'] == 'TX'])	geographic	medium
62	How many retired clients (age > 65) are in the dataset?	8485	len(df[df['current_age'] > 65])	demographics	medium
63	How many clients are issued exactly 2 cards?	19330	len(df[df['num_cards_issued'] == 2])	card_analysis	medium
64	What is the most common number of credit cards clients have?	4	int(df['num_credit_cards'].mode()[0])	demographics	medium
65	How many unique merchant IDs are in the dataset?	6183	df['merchant_id'].nunique()	merchant	easy
66	What is the average number of cards issued per client?	1.52	df['num_cards_issued'].mean()	card_analysis	medium
67	How many transactions have negative amounts?	1925	len(df[df['amount'] < 0])	data_quality	medium
68	How many transactions are from clients born in the 1960s?	8823	len(df[(df['birth_year'] >= 1960) & (df['birth_year'] < 1970)])	demographics	medium
69	What is the latitude range for client addresses?	21.30 to 48.53	f"{df['latitude'].min():.2f} to {df['latitude'].max():.2f}"	geographic	medium
70	What is the longitude range for client addresses?	-158.18 to -68.67	f"{df['longitude'].min():.2f} to {df['longitude'].max():.2f}"	geographic	medium
71	How many transactions involve clients with credit score above 700 AND yearly income above 50000?	7322	len(df[(df['credit_score'] > 700) & (df['yearly_income'] > 50000)])	complex_query	hard
72	What is the average transaction amount for fraudulent transactions?	80.78	df[df['fraud_label'] == 'Yes']['amount'].mean()	complex_query	hard
73	What is the average transaction amount for non-fraudulent transactions?	42.94	df[df['fraud_label'] == 'No']['amount'].mean()	complex_query	hard
74	What is the average credit limit for clients with high debt (>100000)?	22458.09	df[df['total_debt'] > 100000]['credit_limit'].mean()	complex_query	hard
75	What percentage of the dataset has missing values?	3.98%	(df.isna().sum().sum() / (len(df) * len(df.columns)) * 100)	data_quality	hard
76	What is the average amount for transactions in Texas?	45.15	df[df['merchant_state'] == 'TX']['amount'].mean()	geographic	hard
77	What is the debt-to-income ratio for the average client?	1.24	df['total_debt'].mean() / df['yearly_income'].mean()	complex_query	hard
78	How many transactions were made by clients older than the median age?	18121	len(df[df['current_age'] > df['current_age'].median()])	demographics	hard
79	What is the correlation between credit score and yearly income?	-0.0329	df['credit_score'].corr(df['yearly_income'])	complex_query	hard
80	How many transactions involve amounts greater than 2 standard deviations from the mean?	1094	len(df[np.abs(df['amount'] - df['amount'].mean()) > 2 * df['amount'].std()])	complex_query	hard
81	What is the fraud rate for clients with credit score below 650?	0.04%	(len(df[(df['credit_score'] < 650) & (df['fraud_label'] == 'Yes')]) / len(df[df['credit_score'] < 650]) * 100)	fraud_analysis	hard
82	How many transactions are from the top 10 merchant cities by volume?	6563	len(df[df['merchant_city'].isin(df['merchant_city'].value_counts().head(10).index)])	merchant	hard
83	What is the average transaction amount for each card brand?	Visa: 41.17, Mastercard: 43.79, Amex: 46.48, Discover: 44.14	df.groupby('card_brand')['amount'].mean().round(2).to_dict()	complex_query	hard
84	What percentage of fraudulent transactions use Online payment method?	40.74%	(len(df[(df['fraud_label'] == 'Yes') & (df['use_chip'] == 'Online Transaction')]) / len(df[df['fraud_label'] == 'Yes']) * 100)	fraud_analysis	hard
85	What is the average credit score for clients with fraudulent transactions?	718.04	df[df['fraud_label'] == 'Yes']['credit_score'].mean()	fraud_analysis	hard
86	How does average transaction amount vary by card type?	Debit: 42.51, Credit: 43.91, Debit (Prepaid): 45.32	df.groupby('card_type')['amount'].mean().round(2).to_dict()	complex_query	hard
87	What percentage of clients with total debt > yearly income exist?	1825.86%	len(df[df['total_debt'] > df['yearly_income']]) / len(df.drop_duplicates('client_id')) * 100	complex_query	hard
88	What is the median transaction amount by fraud status?	Fraudulent: 27.50, Non-Fraudulent: 29.28	df.groupby('fraud_label')['amount'].median().round(2).to_dict()	fraud_analysis	hard
89	How many transactions exceed the typical transaction amount by more than 3 standard deviations?	465	len(df[df['amount'] > df['amount'].mean() + 3 * df['amount'].std()])	data_quality	hard
90	What is the relationship between number of credit cards and fraud rate?	Requires groupby analysis	df.groupby('num_credit_cards').apply(lambda x: (x['fraud_label'] == 'Yes').sum() / x['fraud_label'].notna().sum() * 100)	complex_query	hard
91	Which states have the highest average transaction amount?	Requires top states analysis	df.groupby('merchant_state')['amount'].mean().nlargest(5)	geographic	hard
92	What is the average age difference between clients with high vs low debt?	-19.85	df[df['total_debt'] > df['total_debt'].quantile(0.75)]['current_age'].mean() - df[df['total_debt'] < df['total_debt'].quantile(0.25)]['current_age'].mean()	demographics	hard

1

What is the total number of transactions in the dataset?

38000

len(df)

basic_statistics

easy

2

How many unique clients are in the dataset?

1218

df['client_id'].nunique()

basic_statistics

easy

3

How many unique cards are in the dataset?

3838

df['card_id'].nunique()

basic_statistics

easy

4

What is the average transaction amount?

43.07

df['amount'].mean()

basic_statistics

easy

5

What is the median transaction amount?

29.15

df['amount'].median()

basic_statistics

easy

6

What is the maximum transaction amount?

1773.35

df['amount'].max()

basic_statistics

easy

7

What is the minimum transaction amount?

-498.00

df['amount'].min()

basic_statistics

easy

8

What is the standard deviation of transaction amounts?

81.05

df['amount'].std()

basic_statistics

medium

9

How many transactions are made with Visa cards?

14249

len(df[df['card_brand'] == 'Visa'])

card_analysis

easy

10

How many transactions are made with Mastercard cards?

20405

len(df[df['card_brand'] == 'Mastercard'])

card_analysis

easy

11

Which card brand has the most transactions?

Mastercard

df['card_brand'].value_counts().index[0]

card_analysis

easy

12

What percentage of transactions use Swipe Transactions?

52.50%

(len(df[df['use_chip'] == 'Swipe Transaction']) / len(df) * 100)

card_analysis

medium

13

How many transactions are made with Amex cards?

2409

len(df[df['card_brand'] == 'Amex'])

card_analysis

easy

14

How many unique merchant cities are in the dataset?

3459

df['merchant_city'].nunique()

geographic

easy

15

Which merchant state has the most transactions?

CA

df['merchant_state'].value_counts().index[0]

geographic

easy

16

How many transactions have missing merchant_state information?

4390

df['merchant_state'].isna().sum()

geographic

medium

17

What is the most common merchant city?

ONLINE

df['merchant_city'].value_counts().index[0]

geographic

easy

18

How many transactions are labeled as fraudulent?

27

len(df[df['fraud_label'] == 'Yes'])

fraud_analysis

easy

19

How many transactions are not fraudulent?

25408

len(df[df['fraud_label'] == 'No'])

fraud_analysis

easy

20

What percentage of transactions are fraudulent?

0.11%

(len(df[df['fraud_label'] == 'Yes']) / len(df[df['fraud_label'].notna()]) * 100)

fraud_analysis

medium

21

How many transactions have missing fraud labels?

12565

df['fraud_label'].isna().sum()

fraud_analysis

easy

22

What is the average credit score in the dataset?

713.26

df['credit_score'].mean()

credit_analysis

easy

23

What is the maximum credit score?

850

int(df['credit_score'].max())

credit_analysis

easy

24

What is the minimum credit score?

488

int(df['credit_score'].min())

credit_analysis

easy

25

How many clients have a credit score above 750?

10492

len(df[df['credit_score'] > 750])

credit_analysis

medium

26

What is the average yearly income in the dataset?

46717.33

df['yearly_income'].mean()

income_analysis

easy

27

What is the average per capita income?

24003.13

df['per_capita_income'].mean()

income_analysis

easy

28

What is the maximum yearly income?

280199.00

df['yearly_income'].max()

income_analysis

easy

29

How many clients have yearly income greater than 50000?

11949

len(df[df['yearly_income'] > 50000])

income_analysis

medium

30

What is the average total debt in the dataset?

58032.68

df['total_debt'].mean()

debt_analysis

easy

31

What is the maximum total debt?

461854.00

df['total_debt'].max()

debt_analysis

easy

32

How many clients have total debt greater than 100000?

6866

len(df[df['total_debt'] > 100000])

debt_analysis

medium

33

What is the average credit limit?

15620.43

df['credit_limit'].mean()

credit_limit

easy

34

What is the maximum credit limit?

141391.00

df['credit_limit'].max()

credit_limit

easy

35

How many cards have credit limit of 0?

158

len(df[df['credit_limit'] == 0])

credit_limit

medium

36

What is the average age of clients?

54.09

df['current_age'].mean()

demographics

easy

37

What is the oldest client age?

101

int(df['current_age'].max())

demographics

easy

38

What is the youngest client age?

23

int(df['current_age'].min())

demographics

easy

39

How many clients are over 60 years old?

11986

len(df[df['current_age'] > 60])

demographics

medium

40

How many male clients are in the dataset?

18544

len(df[df['gender'] == 'Male'])

demographics

easy

41

How many female clients are in the dataset?

19456

len(df[df['gender'] == 'Female'])

demographics

easy

42

What is the gender ratio (Male:Female)?

18544:19456

f"{len(df[df['gender'] == 'Male'])}:{len(df[df['gender'] == 'Female'])}"

demographics

medium

43

How many unique merchant categories are in the dataset?

104

df['mcc_description'].nunique()

merchant

easy

44

What is the most common merchant category?

Grocery Stores, Supermarkets

df['mcc_description'].value_counts().index[0]

merchant

easy

45

How many transactions are for Eating Places and Restaurants?

2972

len(df[df['mcc_description'] == 'Eating Places and Restaurants'])

merchant

medium

46

How many Debit card transactions are there?

23789

len(df[df['card_type'] == 'Debit'])

card_analysis

easy

47

How many Credit card transactions are there?

11658

len(df[df['card_type'] == 'Credit'])

card_analysis

easy

48

What is the most common card type?

Debit

df['card_type'].value_counts().index[0]

card_analysis

easy

49

How many Online transactions are there?

4371

len(df[df['use_chip'] == 'Online Transaction'])

transaction_type

easy

50

How many Chip transactions are there?

13680

len(df[df['use_chip'] == 'Chip Transaction'])

transaction_type

easy

51

What percentage of transactions are Online?

11.50%

(len(df[df['use_chip'] == 'Online Transaction']) / len(df) * 100)

transaction_type

medium

52

How many transactions have errors?

614

df['errors'].notna().sum()

error_analysis

easy

53

What percentage of transactions have errors?

1.62%

(df['errors'].notna().sum() / len(df) * 100)

error_analysis

medium

54

What is the most common error type?

Insufficient Balance

df['errors'].value_counts().index[0]

error_analysis

medium

55

What is the earliest transaction date?

2010-01-01

df['transaction_date'].min().split()[0]

temporal

easy

56

What is the latest transaction date?

2019-10-31

df['transaction_date'].max().split()[0]

temporal

easy

57

How many Visa card transactions are fraudulent?

13

len(df[(df['card_brand'] == 'Visa') & (df['fraud_label'] == 'Yes')])

complex_query

medium

58

How many clients have more than 3 credit cards?

21850

len(df[df['num_credit_cards'] > 3])

complex_query

medium

59

How many transactions are made with cards that have chips?

34242

len(df[df['has_chip'] == 'YES'])

card_analysis

easy

60

What percentage of cards have EMV chips?

90.11%

(len(df[df['has_chip'] == 'YES']) / len(df) * 100)

card_analysis

medium

61

How many transactions occurred in Texas (TX)?

2841

len(df[df['merchant_state'] == 'TX'])

geographic

medium

62

How many retired clients (age > 65) are in the dataset?

8485

len(df[df['current_age'] > 65])

demographics

medium

63

How many clients are issued exactly 2 cards?

19330

len(df[df['num_cards_issued'] == 2])

card_analysis

medium

64

What is the most common number of credit cards clients have?

4

int(df['num_credit_cards'].mode()[0])

demographics

medium

65

How many unique merchant IDs are in the dataset?

6183

df['merchant_id'].nunique()

merchant

easy

66

What is the average number of cards issued per client?

1.52

df['num_cards_issued'].mean()

card_analysis

medium

67

How many transactions have negative amounts?

1925

len(df[df['amount'] < 0])

data_quality

medium

68

How many transactions are from clients born in the 1960s?

8823

len(df[(df['birth_year'] >= 1960) & (df['birth_year'] < 1970)])

demographics

medium

69

What is the latitude range for client addresses?

21.30 to 48.53

f"{df['latitude'].min():.2f} to {df['latitude'].max():.2f}"

geographic

medium

70

What is the longitude range for client addresses?

-158.18 to -68.67

f"{df['longitude'].min():.2f} to {df['longitude'].max():.2f}"

geographic

medium

71

How many transactions involve clients with credit score above 700 AND yearly income above 50000?

7322

len(df[(df['credit_score'] > 700) & (df['yearly_income'] > 50000)])

complex_query

hard

72

What is the average transaction amount for fraudulent transactions?

80.78

df[df['fraud_label'] == 'Yes']['amount'].mean()

complex_query

hard

73

What is the average transaction amount for non-fraudulent transactions?

42.94

df[df['fraud_label'] == 'No']['amount'].mean()

complex_query

hard

74

What is the average credit limit for clients with high debt (>100000)?

22458.09

df[df['total_debt'] > 100000]['credit_limit'].mean()

complex_query

hard

75

What percentage of the dataset has missing values?

3.98%

(df.isna().sum().sum() / (len(df) * len(df.columns)) * 100)

data_quality

hard

76

What is the average amount for transactions in Texas?

45.15

df[df['merchant_state'] == 'TX']['amount'].mean()

geographic

hard

77

What is the debt-to-income ratio for the average client?

1.24

df['total_debt'].mean() / df['yearly_income'].mean()

complex_query

hard

78

How many transactions were made by clients older than the median age?

18121

len(df[df['current_age'] > df['current_age'].median()])

demographics

hard

79

What is the correlation between credit score and yearly income?

-0.0329

df['credit_score'].corr(df['yearly_income'])

complex_query

hard

80

How many transactions involve amounts greater than 2 standard deviations from the mean?

1094

len(df[np.abs(df['amount'] - df['amount'].mean()) > 2 * df['amount'].std()])

complex_query

hard

81

What is the fraud rate for clients with credit score below 650?

0.04%

(len(df[(df['credit_score'] < 650) & (df['fraud_label'] == 'Yes')]) / len(df[df['credit_score'] < 650]) * 100)

fraud_analysis

hard

82

How many transactions are from the top 10 merchant cities by volume?

6563

len(df[df['merchant_city'].isin(df['merchant_city'].value_counts().head(10).index)])

merchant

hard

83

What is the average transaction amount for each card brand?

Visa: 41.17, Mastercard: 43.79, Amex: 46.48, Discover: 44.14

df.groupby('card_brand')['amount'].mean().round(2).to_dict()

complex_query

hard

84

What percentage of fraudulent transactions use Online payment method?

40.74%

(len(df[(df['fraud_label'] == 'Yes') & (df['use_chip'] == 'Online Transaction')]) / len(df[df['fraud_label'] == 'Yes']) * 100)

fraud_analysis

hard

85

What is the average credit score for clients with fraudulent transactions?

718.04

df[df['fraud_label'] == 'Yes']['credit_score'].mean()

fraud_analysis

hard

86

How does average transaction amount vary by card type?

Debit: 42.51, Credit: 43.91, Debit (Prepaid): 45.32

df.groupby('card_type')['amount'].mean().round(2).to_dict()

complex_query

hard

87

What percentage of clients with total debt > yearly income exist?

1825.86%

len(df[df['total_debt'] > df['yearly_income']]) / len(df.drop_duplicates('client_id')) * 100

complex_query

hard

88

What is the median transaction amount by fraud status?

Fraudulent: 27.50, Non-Fraudulent: 29.28

df.groupby('fraud_label')['amount'].median().round(2).to_dict()

fraud_analysis

hard

89

How many transactions exceed the typical transaction amount by more than 3 standard deviations?

465

len(df[df['amount'] > df['amount'].mean() + 3 * df['amount'].std()])

data_quality

hard

90

What is the relationship between number of credit cards and fraud rate?

Requires groupby analysis

df.groupby('num_credit_cards').apply(lambda x: (x['fraud_label'] == 'Yes').sum() / x['fraud_label'].notna().sum() * 100)

complex_query

hard

91

Which states have the highest average transaction amount?

Requires top states analysis

df.groupby('merchant_state')['amount'].mean().nlargest(5)

geographic

hard

92

What is the average age difference between clients with high vs low debt?

-19.85

df[df['total_debt'] > df['total_debt'].quantile(0.75)]['current_age'].mean() - df[df['total_debt'] < df['total_debt'].quantile(0.25)]['current_age'].mean()

demographics

hard