Page 358 - Special Topic Session (STS) - Volume 2

STS500 Neo S.K. et al.
In this pilot project, enterprises were broadly classified into five categories
according to their type of internet usage (Table 1). Enterprises classified under
categories B1, B2, C1 and C2 make up the internet economy and the scope of
the pilot project. Consumer-to-consumer economic activity was excluded
from the scope.

Table 1: Categorisation of enterprises according to internet usage

Category Definition Examples
A Enterprises without websites -

B1 Enterprises which do not generate Websites with information on
income directly from the internet products/services
and has passive internet presence
B2 Enterprises which do not generate Websites with subscription
income directly from the internet services or social media
and has active internet presence outreach
C1 Enterprises which generate income Online retail stores
directly online through sales of goods
C2 Enterprises which generate income Online web hosting services
directly online through sales of
services

2. Data Collection
The Uniform Resource Locators (URLs) or the website addresses of the
enterprises (if the enterprise has a website) were needed to classify the
enterprises. The URLs of enterprises were only available for a sample of
enterprises collected via traditional surveys. Hence, DOS explored using web-
based sources to obtain the URLs of enterprises.
The URLs were collected based on the four steps below:
1) Obtained a list of enterprises from DOS’s business register as the target
population of this pilot project.
2) Purchased URLs from the Singapore Network Information Centre (SGN For
the remaining enterprises without a URL, their names and addresses were
searched on Google Maps. If Google Maps recognised the enterprise and
displayed its URL, the URL is then extracted and merged back to the target
population.
3) Web scraped online business directories (e.g. Kompass and Orbis) and
licensing websites (e.g. the Monetary Authority of Singapore and Travel
Agents Directory) which contained URLs of enterprises. The URLs scraped
were then merged back to the target population.

347 | I S I W S C 2 0 1 9