Use Python faker lib, create tables: 
1. Industry: 3 rows, 1 column which is industry_name: medical supplies, logistics, or construction
2. Owner: 50 rows, with following columns:
  - owner_id: a citizen id having 12 digits. Must be unique
  - full name: Last Name + MiddleName + First Name. Last Name can be Tran/Nguyen/Le. First Name can be Tung/Linh/Hung/Huy/Khoa/Khang/Diem. Middle name can be Van/Huy/Thi/Xuan/Viet.
  - phone: personal phone number starts with +84 (Vietnam region). Must also be unique.
  You should generate unique phone and owner_id first before generating other fields

3. Employee: 
  - empl_persnbr: Employee code. Must be unique. Something like acb-x where x is random between 0~100
  - empl_name: Employee First name. Use faker to generate this. This should also be unique. 
  - area: Random between (Hà Nội/Hải Phòng/Hồ Chí Minh/Quy Nhơn/Đà Nẵng/Khánh Hòa/Kiên Giang/Cần Thơ/Bình Thuận/Nghệ An/Hà Tĩnh)

  You should generate all the unique empl_names and empl_persnbr before generating the rest of this table
4. Company: 70 rows, with the following fields:
  - tax_id: 10 digits. Must be unique
  - owner_id: must be one of the owner_id in Owner
  - org_name: name of the company. Must be unique. Composed of three random letters + JSC/LTD/CORP/GROUP + International/National/Global
  - org_type: KHHH or KHMOI
  - website: org_name + '.com' or '.com.vn'
  - industry_name: Must be one of the three known industries in the Industry table
  - segment: implies the size of the company. Acceptable values are SME (small companies) or MMLC (big companies)
  - phone: company phone number starts with +8402 (Vietnam region)
  - empl_persnbr: code of an employee in Employee table.
  - yoe: year of establishment. A random year between 2002 and 2020

  You should generate the unique org_name and tax_id first before generating other fields.

4. Ledger: number of rows = Number of rows in Company multiplied by a random number between 1.4 and 1.8
  - provider_tax_id: tax_id of the selling company. Must be one of the known companies in Company
  - customer_tax_id: tax_id of the buying company. Must be one of the known companies in Company, but different from provider_tax_id
  - vat_count: number of transactions (a random number between 1 and 10)
  - total_vat_amount: total transaction values. A random number between 5 and 5000, multiplied by 1000000. Unit: VND

  Ledger is uniquely identified by the string provider_tax_id + customer_tax_id. To generate this table, you must generate all the permutations of provider_tax_id + customer_tax_id first, then generate the numeric fields as random numpy arrays.

5. Bank. Only one field which is bank_name, must be one of AGR/ACB/VCB/TCB/VPB/MSB/BID/TTH/MMM/UZU

6. CompBankBalance:  balance of company at bank
  - tax_id: tax_id of the company. Must be one of the known companies in Company 
  - bank_name: one of the bank_name values in Bank
  - balance: Random amount between -1 and 15 multiplied by 1000000000
  Each company must have at least one bank account, no more than 3 accounts