{
  "$version": "0.1.0",
  "origin": "R/clean.R",
  "learn_more": "http://data-dict.tidyverse.org/",
  "tables": [
    {
      "name": "account",
      "description": "<p>Static characteristics of bank accounts. Each account belongs to a branch in a district and has a statement issuance frequency.</p>",
      "source": {
        "parquet": "data/account.parquet"
      },
      "columns": [
        {
          "name": "account_id",
          "description": "<p>Unique identifier for the account.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "referenced_by": [
            {
              "table": "disp",
              "column": "account_id"
            },
            {
              "table": "order",
              "column": "account_id"
            },
            {
              "table": "trans",
              "column": "account_id"
            },
            {
              "table": "loan",
              "column": "account_id"
            }
          ],
          "examples": [
            1,
            1183,
            2367,
            3552,
            11382
          ]
        },
        {
          "name": "district_id",
          "description": "<p>District where the account's branch is located.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "required"
          ],
          "references": {
            "table": "district",
            "column": "district_id"
          },
          "examples": [
            1,
            20,
            39,
            58,
            77
          ]
        },
        {
          "name": "frequency",
          "description": "<p>Frequency of statement issuance.</p>",
          "type": "enum",
          "constraints": [
            "required"
          ],
          "values": [
            "POPLATEK MESICNE",
            "POPLATEK TYDNE",
            "POPLATEK PO OBRATU"
          ],
          "value_labels": {
            "POPLATEK MESICNE": "Monthly",
            "POPLATEK PO OBRATU": "After each transaction",
            "POPLATEK TYDNE": "Weekly"
          }
        },
        {
          "name": "date",
          "description": "<p>Date the account was created.</p>",
          "type": "date",
          "constraints": [
            "required"
          ],
          "range": {
            "min": "1993-01-01",
            "max": "1997-12-29"
          }
        }
      ]
    },
    {
      "name": "client",
      "description": "<p>Characteristics of bank clients. Gender and birth date were originally encoded together in a \"birth number\"; they have been split into separate fields in this version.</p>",
      "source": {
        "parquet": "data/client.parquet"
      },
      "columns": [
        {
          "name": "client_id",
          "description": "<p>Unique identifier for the client.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "referenced_by": [
            {
              "table": "disp",
              "column": "client_id"
            }
          ],
          "examples": [
            1,
            1418,
            2839,
            4257,
            13998
          ]
        },
        {
          "name": "gender",
          "description": "<p>Client's gender.</p>",
          "type": "enum",
          "constraints": [
            "required"
          ],
          "values": [
            "F",
            "M"
          ]
        },
        {
          "name": "birth_date",
          "description": "<p>Client's date of birth.</p>",
          "type": "date",
          "constraints": [
            "required"
          ],
          "range": {
            "min": "1911-08-20",
            "max": "1987-09-27"
          }
        },
        {
          "name": "district_id",
          "description": "<p>District where the client lives.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "required"
          ],
          "references": {
            "table": "district",
            "column": "district_id"
          },
          "examples": [
            1,
            20,
            39,
            58,
            77
          ]
        }
      ]
    },
    {
      "name": "disp",
      "description": "<p>Dispositions linking clients to accounts. Describes the rights of clients to operate accounts. Only owners can issue permanent orders and apply for loans.</p>",
      "source": {
        "parquet": "data/disp.parquet"
      },
      "columns": [
        {
          "name": "disp_id",
          "description": "<p>Unique identifier for the disposition.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "referenced_by": [
            {
              "table": "card",
              "column": "disp_id"
            }
          ],
          "examples": [
            1,
            1418,
            2839,
            4257,
            13690
          ]
        },
        {
          "name": "client_id",
          "description": "<p>The client who has access to the account.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "required"
          ],
          "references": {
            "table": "client",
            "column": "client_id"
          },
          "examples": [
            1,
            1418,
            2839,
            4257,
            13998
          ]
        },
        {
          "name": "account_id",
          "description": "<p>The account the client has access to.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "required"
          ],
          "references": {
            "table": "account",
            "column": "account_id"
          },
          "examples": [
            1,
            1183,
            2367,
            3552,
            11382
          ]
        },
        {
          "name": "type",
          "description": "<p>Type of disposition.</p>",
          "type": "enum",
          "constraints": [
            "required"
          ],
          "values": [
            "OWNER",
            "DISPONENT"
          ],
          "value_labels": {
            "DISPONENT": "Limited access",
            "OWNER": "Full rights (can issue orders and apply for loans)"
          }
        }
      ]
    },
    {
      "name": "order",
      "description": "<p>Permanent (standing) payment orders on accounts. These are recurring debits set up by account owners.</p>",
      "source": {
        "parquet": "data/order.parquet"
      },
      "columns": [
        {
          "name": "order_id",
          "description": "<p>Unique identifier for the permanent order.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "examples": [
            29401,
            31187,
            32988,
            34786,
            46338
          ]
        },
        {
          "name": "account_id",
          "description": "<p>Account the order is issued for.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "required"
          ],
          "references": {
            "table": "account",
            "column": "account_id"
          },
          "examples": [
            1,
            1174,
            2342,
            3502,
            11362
          ]
        },
        {
          "name": "bank_to",
          "description": "<p>Two-letter code identifying the recipient's bank.</p>",
          "type": "string",
          "constraints": [
            "required"
          ],
          "examples": [
            "AB",
            "GH",
            "MN",
            "ST",
            "YZ"
          ]
        },
        {
          "name": "account_to",
          "description": "<p>Account number of the recipient.</p>",
          "type": "number(id)",
          "constraints": [
            "required"
          ],
          "examples": [
            399,
            24175644,
            49756062,
            73948222,
            99994199
          ]
        },
        {
          "name": "amount",
          "description": "<p>Amount debited per payment.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 1,
            "max": 14882
          }
        },
        {
          "name": "k_symbol",
          "description": "<p>Characterization of the payment. Never null; unspecified payments carry an empty string rather than a null (1379 of 6471 rows).</p>",
          "type": "enum",
          "values": [
            "POJISTNE",
            "SIPO",
            "LEASING",
            "UVER",
            ""
          ],
          "value_labels": {
            "": "Not specified",
            "LEASING": "Leasing",
            "POJISTNE": "Insurance payment",
            "SIPO": "Household payment",
            "UVER": "Loan payment"
          }
        }
      ]
    },
    {
      "name": "trans",
      "description": "<p>Individual transactions on accounts, including credits and debits from various sources.</p>",
      "source": {
        "parquet": "data/trans.parquet"
      },
      "columns": [
        {
          "name": "trans_id",
          "description": "<p>Unique identifier for the transaction.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "examples": [
            1,
            430263,
            858506,
            2060979,
            3682987
          ]
        },
        {
          "name": "account_id",
          "description": "<p>Account the transaction belongs to.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "required"
          ],
          "references": {
            "table": "account",
            "column": "account_id"
          },
          "examples": [
            1,
            1183,
            2367,
            3552,
            11382
          ]
        },
        {
          "name": "date",
          "description": "<p>Date of the transaction.</p>",
          "type": "date",
          "constraints": [
            "required"
          ],
          "range": {
            "min": "1993-01-01",
            "max": "1998-12-31"
          }
        },
        {
          "name": "type",
          "description": "<p>Transaction direction.</p>",
          "type": "enum",
          "constraints": [
            "required"
          ],
          "values": [
            "PRIJEM",
            "VYDAJ",
            "VYBER"
          ],
          "value_labels": {
            "PRIJEM": "Credit",
            "VYBER": "Withdrawal (cash)",
            "VYDAJ": "Withdrawal (debit)"
          }
        },
        {
          "name": "operation",
          "description": "<p>Mode of transaction.</p>",
          "type": "enum",
          "values": [
            "VYBER KARTOU",
            "VKLAD",
            "PREVOD Z UCTU",
            "VYBER",
            "PREVOD NA UCET"
          ],
          "value_labels": {
            "PREVOD NA UCET": "Remittance to another bank",
            "PREVOD Z UCTU": "Collection from another bank",
            "VKLAD": "Credit in cash",
            "VYBER": "Cash withdrawal",
            "VYBER KARTOU": "Credit card withdrawal"
          }
        },
        {
          "name": "amount",
          "description": "<p>Amount of money in the transaction.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 0,
            "max": 87400
          }
        },
        {
          "name": "balance",
          "description": "<p>Account balance after the transaction.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": -41126,
            "max": 209637
          }
        },
        {
          "name": "k_symbol",
          "description": "<p>Characterization of the transaction. Has two distinct kinds of missingness: a single space <code>\" \"</code> (53,433 rows) and a true null (481,881 rows). Treat both as \"not specified\".</p>",
          "type": "enum",
          "values": [
            "POJISTNE",
            "SLUZBY",
            "UROK",
            "SANKC. UROK",
            "SIPO",
            "DUCHOD",
            "UVER",
            " "
          ],
          "value_labels": {
            " ": "Not specified",
            "DUCHOD": "Old-age pension",
            "POJISTNE": "Insurance payment",
            "SANKC. UROK": "Sanction interest (negative balance)",
            "SIPO": "Household payment",
            "SLUZBY": "Payment for statement",
            "UROK": "Interest credited",
            "UVER": "Loan payment"
          }
        },
        {
          "name": "bank",
          "description": "<p>Two-letter code identifying the partner's bank.</p>",
          "type": "string",
          "examples": [
            "AB",
            "GH",
            "MN",
            "ST",
            "YZ"
          ]
        },
        {
          "name": "account",
          "description": "<p>Account number of the transaction partner.</p>",
          "type": "number(id)",
          "examples": [
            0,
            24118912,
            49954766,
            74343323,
            99994199
          ]
        }
      ]
    },
    {
      "name": "loan",
      "description": "<p>Loans granted for accounts. At most one loan can be granted per account. This is the primary target table for the PKDD'99 discovery challenge. 682 rows.</p>",
      "details": "<p>The raw export carries a 683rd row that is null in every column; <code>R/clean.R</code> drops it. See the dataset <code>origin</code> for details.</p>",
      "source": {
        "parquet": "data/loan.parquet"
      },
      "columns": [
        {
          "name": "loan_id",
          "description": "<p>Unique identifier for the loan.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "examples": [
            4959,
            5576,
            6178,
            6753,
            7308
          ]
        },
        {
          "name": "account_id",
          "description": "<p>Account the loan was granted for.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "unique",
            "required"
          ],
          "references": {
            "table": "account",
            "column": "account_id"
          },
          "examples": [
            2,
            2962,
            5742,
            8688,
            11362
          ]
        },
        {
          "name": "date",
          "description": "<p>Date the loan was granted.</p>",
          "type": "date",
          "constraints": [
            "required"
          ],
          "range": {
            "min": "1993-07-05",
            "max": "1998-12-08"
          }
        },
        {
          "name": "amount",
          "description": "<p>Total amount of the loan.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 4980,
            "max": 590820
          }
        },
        {
          "name": "duration",
          "description": "<p>Duration of the loan in months.</p>",
          "type": "number(ordinal)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 12,
            "max": 60
          }
        },
        {
          "name": "payments",
          "description": "<p>Monthly payment amount.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 304,
            "max": 9910
          }
        },
        {
          "name": "status",
          "description": "<p>Status of the loan.</p>",
          "type": "enum",
          "constraints": [
            "required"
          ],
          "values": [
            "A",
            "B",
            "C",
            "D"
          ],
          "value_labels": {
            "A": "Contract finished, no problems",
            "B": "Contract finished, loan not paid",
            "C": "Running contract, OK so far",
            "D": "Running contract, client in debt"
          }
        }
      ]
    },
    {
      "name": "card",
      "description": "<p>Credit cards issued to account holders via their dispositions. Multiple cards can be issued per account.</p>",
      "source": {
        "parquet": "data/card.parquet"
      },
      "columns": [
        {
          "name": "card_id",
          "description": "<p>Unique identifier for the card.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "examples": [
            1,
            230,
            456,
            684,
            1247
          ]
        },
        {
          "name": "disp_id",
          "description": "<p>Disposition linking the card to a client and account.</p>",
          "type": "number(id)",
          "constraints": [
            "foreign_key",
            "required"
          ],
          "references": {
            "table": "disp",
            "column": "disp_id"
          },
          "examples": [
            9,
            1388,
            2929,
            4457,
            13660
          ]
        },
        {
          "name": "type",
          "description": "<p>Type of credit card.</p>",
          "type": "enum",
          "constraints": [
            "required"
          ],
          "values": [
            "junior",
            "classic",
            "gold"
          ]
        },
        {
          "name": "issued",
          "description": "<p>Date the card was issued.</p>",
          "type": "date",
          "constraints": [
            "required"
          ],
          "range": {
            "min": "1993-11-07",
            "max": "1998-12-29"
          }
        }
      ]
    },
    {
      "name": "district",
      "description": "<p>Demographic data for Czech districts. One row per district, 77 in all. Column names A2–A16 are the original anonymized names from the PKDD'99 challenge.</p>",
      "details": "<p>The raw export carries a 78th row whose <code>district_id</code> is the literal text <code>sdasdf</code> and whose every other column is null — junk written into the public source database. <code>R/clean.R</code> drops it. A12 and A15 each have one genuine null, which is why neither is marked <code>required</code>.</p>",
      "source": {
        "parquet": "data/district.parquet"
      },
      "columns": [
        {
          "name": "district_id",
          "description": "<p>Unique identifier for the district.</p>",
          "type": "number(id)",
          "constraints": [
            "primary_key",
            "unique",
            "required"
          ],
          "referenced_by": [
            {
              "table": "account",
              "column": "district_id"
            },
            {
              "table": "client",
              "column": "district_id"
            }
          ],
          "examples": [
            1,
            20,
            39,
            58,
            77
          ]
        },
        {
          "name": "A2",
          "description": "<p>District name.</p>",
          "type": "string",
          "constraints": [
            "unique",
            "required"
          ],
          "examples": [
            "Benesov",
            "Hradec Kralove",
            "Nachod",
            "Rokycany",
            "Znojmo"
          ]
        },
        {
          "name": "A3",
          "description": "<p>Region name.</p>",
          "type": "enum",
          "constraints": [
            "required"
          ],
          "values": [
            "Prague",
            "central Bohemia",
            "south Bohemia",
            "west Bohemia",
            "north Bohemia",
            "east Bohemia",
            "south Moravia",
            "north Moravia"
          ]
        },
        {
          "name": "A4",
          "description": "<p>Number of inhabitants.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 42821,
            "max": 1204953
          }
        },
        {
          "name": "A5",
          "description": "<p>Number of municipalities with fewer than 499 inhabitants.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 0,
            "max": 151
          }
        },
        {
          "name": "A6",
          "description": "<p>Number of municipalities with 500–1999 inhabitants.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 0,
            "max": 70
          }
        },
        {
          "name": "A7",
          "description": "<p>Number of municipalities with 2000–9999 inhabitants.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 0,
            "max": 20
          }
        },
        {
          "name": "A8",
          "description": "<p>Number of municipalities with 10,000 or more inhabitants.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 0,
            "max": 5
          }
        },
        {
          "name": "A9",
          "description": "<p>Number of cities.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 1,
            "max": 11
          }
        },
        {
          "name": "A10",
          "description": "<p>Ratio of urban inhabitants (percentage).</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 33.9,
            "max": 100
          }
        },
        {
          "name": "A11",
          "description": "<p>Average salary.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 8110,
            "max": 12541
          }
        },
        {
          "name": "A12",
          "description": "<p>Unemployment rate in 1995 (percentage).</p>",
          "type": "number(quantity)",
          "range": {
            "min": 0.2,
            "max": 7.3
          }
        },
        {
          "name": "A13",
          "description": "<p>Unemployment rate in 1996 (percentage).</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 0.43,
            "max": 9.4
          }
        },
        {
          "name": "A14",
          "description": "<p>Number of entrepreneurs per 1000 inhabitants.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 81,
            "max": 167
          }
        },
        {
          "name": "A15",
          "description": "<p>Number of committed crimes in 1995.</p>",
          "type": "number(quantity)",
          "range": {
            "min": 818,
            "max": 85677
          }
        },
        {
          "name": "A16",
          "description": "<p>Number of committed crimes in 1996.</p>",
          "type": "number(quantity)",
          "constraints": [
            "required"
          ],
          "range": {
            "min": 888,
            "max": 99107
          }
        }
      ]
    }
  ],
  "relationships": [
    {
      "description": "<p>Account is located in a district.</p>",
      "cardinality": "many-to-one",
      "declared_cardinality": "many-to-one",
      "pairs": [
        {
          "left": {
            "table": "account",
            "column": "district_id"
          },
          "right": {
            "table": "district",
            "column": "district_id"
          }
        }
      ],
      "join": "account.district_id = district.district_id"
    },
    {
      "description": "<p>Client lives in a district.</p>",
      "cardinality": "many-to-one",
      "declared_cardinality": "many-to-one",
      "pairs": [
        {
          "left": {
            "table": "client",
            "column": "district_id"
          },
          "right": {
            "table": "district",
            "column": "district_id"
          }
        }
      ],
      "join": "client.district_id = district.district_id"
    },
    {
      "description": "<p>Disposition links a client to an account.</p>",
      "cardinality": "many-to-one",
      "declared_cardinality": "many-to-one",
      "pairs": [
        {
          "left": {
            "table": "disp",
            "column": "client_id"
          },
          "right": {
            "table": "client",
            "column": "client_id"
          }
        }
      ],
      "join": "disp.client_id = client.client_id"
    },
    {
      "description": "<p>Disposition links a client to an account.</p>",
      "cardinality": "many-to-one",
      "declared_cardinality": "many-to-one",
      "pairs": [
        {
          "left": {
            "table": "disp",
            "column": "account_id"
          },
          "right": {
            "table": "account",
            "column": "account_id"
          }
        }
      ],
      "join": "disp.account_id = account.account_id"
    },
    {
      "description": "<p>Permanent order belongs to an account.</p>",
      "cardinality": "many-to-one",
      "declared_cardinality": "many-to-one",
      "pairs": [
        {
          "left": {
            "table": "order",
            "column": "account_id"
          },
          "right": {
            "table": "account",
            "column": "account_id"
          }
        }
      ],
      "join": "order.account_id = account.account_id"
    },
    {
      "description": "<p>Transaction belongs to an account.</p>",
      "cardinality": "many-to-one",
      "declared_cardinality": "many-to-one",
      "pairs": [
        {
          "left": {
            "table": "trans",
            "column": "account_id"
          },
          "right": {
            "table": "account",
            "column": "account_id"
          }
        }
      ],
      "join": "trans.account_id = account.account_id"
    },
    {
      "description": "<p>Loan is granted for an account (at most one per account).</p>",
      "cardinality": "one-to-one",
      "declared_cardinality": "one-to-one",
      "pairs": [
        {
          "left": {
            "table": "loan",
            "column": "account_id"
          },
          "right": {
            "table": "account",
            "column": "account_id"
          }
        }
      ],
      "join": "loan.account_id = account.account_id"
    },
    {
      "description": "<p>Credit card is issued via a disposition.</p>",
      "cardinality": "many-to-one",
      "declared_cardinality": "many-to-one",
      "pairs": [
        {
          "left": {
            "table": "card",
            "column": "disp_id"
          },
          "right": {
            "table": "disp",
            "column": "disp_id"
          }
        }
      ],
      "join": "card.disp_id = disp.disp_id"
    }
  ],
  "glossary": [
    {
      "term": "POPLATEK MESICNE",
      "definition": "<p>Monthly (statement issuance frequency).</p>"
    },
    {
      "term": "POPLATEK TYDNE",
      "definition": "<p>Weekly (statement issuance frequency).</p>"
    },
    {
      "term": "POPLATEK PO OBRATU",
      "definition": "<p>After each transaction (statement issuance frequency).</p>"
    },
    {
      "term": "PRIJEM",
      "definition": "<p>Credit (incoming transaction).</p>"
    },
    {
      "term": "VYDAJ",
      "definition": "<p>Withdrawal/debit (outgoing transaction).</p>"
    },
    {
      "term": "VYBER",
      "definition": "<p>Cash withdrawal.</p>"
    },
    {
      "term": "VYBER KARTOU",
      "definition": "<p>Credit card withdrawal.</p>"
    },
    {
      "term": "VKLAD",
      "definition": "<p>Cash deposit.</p>"
    },
    {
      "term": "PREVOD Z UCTU",
      "definition": "<p>Collection from another bank.</p>"
    },
    {
      "term": "PREVOD NA UCET",
      "definition": "<p>Remittance to another bank.</p>"
    },
    {
      "term": "POJISTNE",
      "definition": "<p>Insurance payment.</p>"
    },
    {
      "term": "SIPO",
      "definition": "<p>Household payment.</p>"
    },
    {
      "term": "LEASING",
      "definition": "<p>Leasing payment.</p>"
    },
    {
      "term": "UVER",
      "definition": "<p>Loan payment.</p>"
    },
    {
      "term": "SLUZBY",
      "definition": "<p>Payment for statement.</p>"
    },
    {
      "term": "UROK",
      "definition": "<p>Interest credited.</p>"
    },
    {
      "term": "SANKC. UROK",
      "definition": "<p>Sanction interest charged when account has negative balance.</p>"
    },
    {
      "term": "DUCHOD",
      "definition": "<p>Old-age pension.</p>"
    },
    {
      "term": "DISPONENT",
      "definition": "<p>A person with limited access rights to an account.</p>"
    },
    {
      "term": "OWNER",
      "definition": "<p>Account owner with full rights (can issue orders and apply for loans).</p>"
    }
  ]
}