White Papers Making more of data using AI | Page 4

Data Management

A ComTech Advisory Whitepaper

Although the DataGenic survey found that the initial interest in the use of social media to support
commodity-trading activities was very low (Figure 1), when respondents were challenged to think
about social media as a source of trade intelligence, they became significantly more interested.
In particular, the respondents struggled to understand how any valuable trend data could be
extracted from the vast volume of noise. Artificial Intelligence is already used in some industries
to help identify potentially useful signals from very noisy and voluminous data and surely this
could be applied also to social media data.
Figure 1: Relative Importance of Different Types of Data (5 is very important, 3 is average and 1 is
unimportant)
Social Media
Algorithmic Data
Technical Indicators
News
Historical Data
Weather Data
Fundamental Data
Realtime Market Prices
0

1

2

3

4

5

At the same time as the quantity of data sources and volume o bFFw&