Commodity Documents — Information Retrieval Fields
Some documents collected by Bilby are relevant to one or more commodities that are tradeable on Chinese exchanges. Bilby labels such documents with the relevant commodity, quantifies the degree of this relevance with a floating point score, and assigns ordinal rankings to each day's documents, based on this score.
| Field Name | Type | Example/Possible Values | Description |
|---|---|---|---|
relevant_commodity | string | Aluminum , Coal , Coke , Coking Coal , Copper , Corn , Crude Oil , Ethylene Glycol , Gold , Hot Rolled Coil , Iron Ore , Lead , Methanol , Natural Gas , Natural Rubber , Nickel , PE , PP , PTA , PVC , Palladium , Palm Oil , Platinum , Polybutadiene , Rapeseed Meal , Rapeseed Oil , Silver , Soy Oil , Soybean , Soybean Meal , Steel Rebar , TSR 20 , Tin , Urea , Wheat , Wood Pulp , or Zinc | The commodity to which the underlying document is relevant. |
theme_relevance_score | number | 0 ≤ value ≤ 1 | The relevance score of the underlying document to the relevant_commodity. |
theme_relevance_rank | number | 1, 2, 3, ... | The rank of the underlying document, as measured by relevance_score, compared to other commodity-related documents published on the same day. |
relevant_commodity
Definition: The commodity to which the underlying document is relevant.
Possible values: One of the 37 labels: Aluminum , Coal , Coke ,
Coking Coal , Copper , Corn , Crude Oil , Ethylene Glycol , Gold ,
Hot Rolled Coil , Iron Ore , Lead , Methanol , Natural Gas ,
Natural Rubber , Nickel , PE , PP , PTA , PVC , Palladium ,
Palm Oil , Platinum , Polybutadiene , Rapeseed Meal , Rapeseed Oil ,
Silver , Soy Oil , Soybean , Soybean Meal , Steel Rebar , TSR 20 ,
Tin , Urea , Wheat , Wood Pulp , and Zinc.
theme_relevance_score
Definition: Bilby has created one classifier model for each of the 37
commodities currently found on Chinese exchanges (please consult the "possible
values" section in the relevant_commodity entry, below, for a complete list of
these commodities). Each model computes the relevance of the underlying document
to the given commodity, and assigns this value to the theme_relevance_score
variable.
Possible values: A float in the interval [0, 1] (i.e., between 0 and 1,
inclusive).
theme_relevance_rank
Definition: On any given day, the document with the Nth-highest
theme_relevance_score is assigned a theme_relevance_rank of N. For
example, the highest-scored document is given a rank of 1; the second-highest, a
rank of 2, and so on.
Possible values: Positive integers 1, 2, 3, ...
Note: The tabular data returned by the Bilby API is sorted first in
ascending order of publication date (i.e. published_at), and then in
descending order of theme_relevance_score. As a result, the
theme_relevance_rank is sorted in ascending order for each day.