Page 50 - Reclaim YOUR DIGITAL GOLD (without audio)

RECLAIM YOUR DIGITAL GOLD

machine learning (ML) have greatly simplified data
organization and categorization. To accomplish this, the
AI and ML must first be linked to the appropriate data
source or sources before being trained/programmed to
process it.

Let’s start with a look at the various data collection
methods and data sources that result from data
harvesting.

HOW IS DATA COLLECTED?

The methods and data sources most commonly used
for building machine learning models are summarized
below:

1. Traditional Data Collection Methods
Many businesses have had technology systems that
process and/or collect data over time, making such
data invaluable to their operations today. Some of
these systems are made up of databases that contain
information about products, inventory, sales, and
customers, as well as transactional level data.

The retail industry has some of the most comprehensive
and historical data due to the various POS (Point Of Sale)
systems and manufacturer databanks. When combined
with the data continuum from product manufacturer to
retailer to consumer level data, this can be a powerful
data lake. Any industry could claim the same; however,
the manufacturing, banking, insurance, CPG (Consumer
Package Goods), and automotive industries have led the
way in amassing such traditional data aggregates over
decades.

30