tlevine / match-companies Goto Github PK

The present package produces a data table about the entities who have bid on contracts for projects funded by the World Bank Group.

Within any project funded by the World Bank, many different contracts may be involved. Several different bidders will bid on each contract. Bidders are usually single companies, but they are sometimes joint ventures of multiple companies. Each row in the "companies.csv" table corresponds to a company inside a bidder inside a contract inside a project.

We have the following fields for each company.

project.name

Name of the project that the company bid on

contract.uri

Page on the World Bank website for the contract that the company bid on

contract.number

Identification number for the contract

bidder.name

This is the name of the bidder. This is not necessarily a company because a bidder can be a joint venture of multiple companies.

bidder.country

Country for the bidder

bid.status

Bids are "awarded" (evaluated and chosen) "evaluated" (evaluated and not chosen) or "rejected" (rejected before evaluation for not following formal bid procedures).

original.company.name

Name of the company as listed on the World Bank website

original.company.country

Country of the company as listed on the World Bank website

opencorporates.company.name

Original company names were reconciled with the Open Corporates reconciliation API. If a match was found, this field contains the Open Corporates name; otherwise, this field is empty.

opencorporates.company.uri

Original company names were reconciled with the Open Corporates reconciliation API. If a match was found, this field contains the Open Corporates URI; otherwise, this field is empty.

address

Dunno

country

Separate field for the country of the bidder

duration

Dunno

date.signature

Dunno

score.financial

Dunno

score.technical

Dunno

score.final

Dunno

original.price.opening

Raw text for the opening price

original.price.evaluated

Raw text for the evaluated price

original.price.contract

Raw text for the contract price

price.opening.currency

Currency of the opening price

price.opening.amount

Numeric opening price

price.evaluated.currency

Currency of the evaluated price

price.evaluated.amount

Numeric evaluated price

price.contract.currency

Currency of the contract price

price.contract.amount

Numeric contract price

method.procurement

?

method.selection

?

scope

?

name

?

reason.rejection

?

small.contract.notice

?

ranking.final

?

Maxime Rusaev produced some of these fields, (in the ./scraper directory.) and then Thomas Levine split those up further (in the reconcile.py file).

The resulting table is called "companies.csv".

You can access the resulting dataset at http://small.dada.pink/match-companies-output/companies.csv or by running

./reconcile.py