Skip to content

Commodities Knowledge Model

Overview

Commodities (display name: Commodities EN v#) is a categorization and extraction Knowledge Model that aims at classifying texts dealing with the commodity market and extracts specific commodities items plus the derivative and exchange traded funds (ETF) contracts. It is focused on the main sectors involved in such market, namely agriculture, energy, livestock, metallurgy, and chemistry, including any related financial instrument, like derivatives and ETFs.

This Knowledge Model addresses the collection of information about specific commodity sectors or instruments for financial analysis.

Categorization

The Commodities Knowledge Model taxonomy covers the financial instruments of six main sectors, as follows:

1000 Commodity Market
    1100 Agriculture
        1110 Derivatives
        1120 ETFs
    1200 Livestock
        1210 Derivatives
        1220 ETFs
    1300 Energy, Oil and Gas
        1310 Derivatives
        1320 ETFs
    1400 Precious Metals
        1410 Derivatives
        1420 ETFs
    1500 Industrial Metals
        1510 Derivatives
        1520 ETFs
    1600 Chemicals and Petrochemicals
        1610 Derivatives
        1620 ETFs

where:

  • Agriculture: for products like crops, cocoa, any kind of vegetable oil, milk, cotton, and several plant-based raw materials.
  • Livestock: live cattle, wool, and meat commodities.
  • Energy, Oil, and Gas: electric energy, renewable energy, gas, crude oil and fuel.
  • Precious Metals: gold, silver, platinum, and palladium.
  • Industrial Metals: ferrous and non-ferrous metals like copper, steel or lithium.
  • Chemicals and Petrochemicals: petrochemicals other than fuel, like polymers and other chemicals.

For each sector, Derivatives category refers to commodity-based futures, or options contracts, while the ETFs category is triggered by mentions to ETF contracts based on commodities.

Extraction groups and classes

COMMODITIES

The COMMODITIES group is used to extract references to commodities.
Its classes are:

  • agriculture
  • livestock
  • energy
  • industrial_metals
  • precious_metals
  • chemicals

DERIVATIVES

The DERIVATIVES group extracts commodity-based derivative contracts. Its only class is security.

ETFS

The ETFS group extracts ETFs contracts based on commodities. Its only class is etfs.

Output structure

The model output has the same structure as any other model and is affected by the functional properties of the workflow block.
The peculiar parts of the output are the result of categorization, i.e. the categories array, and the result of information extraction, i.e. the extractions array.

Example

In the extraction output, the template key corresponds to the concept of group and template fields correspond to classes.
Considering the text:

SOYB
Fund Summary
The Teucrium Soybean Fund (SOYB) provides investors an easy way to gain exposure to the price of soybeans futures in a brokerage account.
SOYB: Did You Know?
Soybeans are one of the most important agricultural commodities, used throughout the global economy as feed, oils, wood substitutes, foam, ink, and crayons.
Demand for soybeans is rising exponentially due to world population growth and the expansion of the global middle class.
Soybean prices have a historically low correlation with U.S. equities making SOYB a potentially attractive option for portfolio diversification.

the JSON output excerpt for categorization is:

"categories": [
    {
        "frequency": 10.71,
        "hierarchy": [
            "Agriculture"
        ],
        "id": "1100",
        "label": "Agriculture",
        "namespace": "commodities_en",
        "positions": [
            {
                "end": 1151,
                "start": 1143
            }
        ],
        "score": 90,
        "winner": true
    },
    {
        "frequency": 2.38,
        "hierarchy": [
            "Agriculture",
            "Derivatives"
        ],
        "id": "1110",
        "label": "Derivatives",
        "namespace": "commodities_en",
        "positions": [
            {
                "end": 1090,
                "start": 1082
            }
        ],
        "score": 20,
        "winner": true
    }
]

For the extraction, it is:

"extractions": [
    {
        "fields": [
            {
                "name": "agriculture",
                "positions": [
                    {
                        "end": 1151,
                        "start": 1143
                    },
                    {
                        "end": 1319,
                        "start": 1311
                    }
                ],
                "value": "soybean"
            }
        ],
        "namespace": "commodities_en",
        "template": "COMMODITIES"
    },
    {
        "fields": [
            {
                "name": "precious_metal",
                "positions": [
                    {
                        "end": 214,
                        "start": 199
                    }
                ],
                "value": "precious metal"
            }
        ],
        "namespace": "commodities_en",
        "template": "COMMODITIES"
    }
]