Markets, Tickers and Indices Knowledge Model
Overview
The Markets, Tickers and Indices Knowledge Model (display name: Markets, Tickers and Indices EN v#) for English texts like news articles, technical analyses and financial reports, has two goals:
- Categorize texts according to the geographical positioning of the stock market(s) mentioned.
- Extract entities that are usually related to stock markets, namely stock exchanges, publicly listed companies, their corresponding stock symbols (tickers), and stock indices.
Categorization
The category tree is:.
01 North America
01.01 New York
01.02 Toronto
02 South America
02.01 São Paulo
03 Africa
03.03 Johannesburg
04 Australia-Oceania
04.01 Sydney
05 Asia
05.01 Shanghai
05.02 Tokyo
05.03 Hong Kong
05.04 Shenzhen
05.05 Mumbai
05.06 Riyadh
05.07 Seoul
05.08 Taipei
05.09 Tehran
06 Europe
06.01 Frankfurt
06.02 Stockholm
06.03 Zurich
06.04 Amsterdam
06.05 Milan
06.06 Brussels
06.07 Oslo
06.08 Paris
06.09 Lisbon
06.10 London
06.11 Madrid
The 25 cities with a specific category cover the most important markets. For other cities, the model predicts the category corresponding to their continent.
Mexico and Nicaragua markets are assimilated to North America.
Continent categories are also triggered when international entities are mentioned and do not refer to a specific city of the taxonomy: for example, Euronext triggers 06 Europe, while Euronext Amsterdam triggers 06.04 Amsterdam.
Extraction groups and classes
The model extracts stock exchanges, publicly listed companies, tickers and stock market indices.
STOCK_EXCHANGE
The STOCK_EXCHANGE group extracts, in its only class Stock_exchange, marketplaces where stockbrokers and traders can buy and sell securities, such as shares of stock, bonds, and other financial instruments.
PUBLIC_COMPANY_AND_TICKER
The PUBLIC_COMPANY_AND_TICKER extracts information about publicly listed companies (PLCs).
Its classes are:
| Class | Description |
|---|---|
| Company | Name of the PLC |
| Ticker | Stock symbol |
| Listed_on | A market where the company is listed |
STOCK_INDEX
The STOCK_INDEX group extracts, in its only class Index, mentions to stock market indices like Standard & Poor's 500 Index or Dow Jones Industrial Average (DJIA), but also specialized indices that track a particular industry or segment.
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
Considering the text:
ATLANTA, NEW YORK & HONG KONG--(BUSINESS WIRE)-- Intercontinental Exchange, Inc. (NYSE: ICE), a leading global provider of data, technology and market infrastructure, today announced that KB Securities has selected the NYSE FANG+Daily 2x Leveraged Index (Interest Rate Adjusted Total Return) for its KB Leverage FANG Plus ETN(H).
The NYSE FANG+Daily 2x Leveraged Index (Interest Rate Adjusted TR) delivers a daily return that is approximately twice that of the NYSE FANG+ Index (Total Return) with the application of an overnight interest rate to account for the leverage. The NYSE FANG+ Index (TR) is an equal-dollar weighted index with the objective of tracking the performance of highly-traded growth stocks of technology and tech-enabled companies in the technology, media & communications and consumer discretionary sectors. The KB Leverage FANG Plus ETN(H) began trading on April 19, 2022.
KB Securities, a wholly owned subsidiary of KB Financial Group, is a leading investment bank in Korea that provides asset management services to retail customers as well as financing advisory and investment services to corporate customers. KB Securities is committed to providing investment products that are tailored to their clients' demands across various asset classes.
"This new ETN will allow Korean market participants to access key "FAANG" stocks, such as Facebook (Meta Platforms), Apple, Amazon, Netflix and Google (Alphabet), through a single instrument", said Magnus Cattan, Head of ICE Fixed Income & Data Services, Asia Pacific. "We are pleased to have worked with one of Korea's largest investment banks to launch the first ETN that tracks an ICE index in the country".
"The KB Leverage FANG Plus ETN(H) will provide our investors access to some of the most recognizable companies in the world", said Ho Young Kim, Head of Equity Trading at KB Securities. "We believe that this product offers Korean market participants a unique tool to capture the performance of U.S. listed technology and tech-enabled companies".
ICE's global family of indices serve as the performance benchmark for $1.5 trillion in fund assets managed by investors around the globe. For more information about ICE's Indices, please visit: https://www.theice.com/market-data/indices.
About Intercontinental Exchange
Intercontinental Exchange, Inc. (NYSE: ICE) is a Fortune 500 company that designs, builds and operates digital networks to connect people to opportunity. We provide financial technology and data services across major asset classes that offer our customers access to mission-critical workflow tools that increase transparency and operational efficiencies. We operate exchanges, including the New York Stock Exchange, and clearing houses that help people invest, raise capital and manage risk across multiple asset classes. Our comprehensive fixed income data services and execution capabilities provide information, analytics and platforms that help our customers capitalize on opportunities and operate more efficiently. At ICE Mortgage Technology, we are transforming and digitizing the U.S. residential mortgage process, from consumer engagement through loan registration. Together, we transform, streamline and automate industries to connect our customers to opportunity.
the categorization output is:
"categories": [
{
"frequency": 88.01,
"hierarchy": [
"North America",
"New York"
],
"id": "01.01",
"label": "New York",
"namespace": "markets_tickers_indices_en",
"positions": [
{
"end": 17,
"start": 9
},
{
"end": 80,
"start": 49
},
{
"end": 82,
"start": 81
},
{
"end": 86,
"start": 82
},
{
"end": 87,
"start": 86
},
{
"end": 91,
"start": 88
},
{
"end": 92,
"start": 91
},
{
"end": 150,
"start": 144
},
{
"end": 2328,
"start": 2297
},
{
"end": 2330,
"start": 2329
},
{
"end": 2334,
"start": 2330
},
{
"end": 2335,
"start": 2334
},
{
"end": 2339,
"start": 2336
},
{
"end": 2340,
"start": 2339
},
{
"end": 2711,
"start": 2688
}
],
"score": 779,
"winner": true
}
]
In the extraction output, the template key corresponds to the concept of group and template fields correspond to classes:
"extractions": [
{
"fields": [
{
"name": "Company",
"positions": [
{
"end": 80,
"start": 49
},
{
"end": 2328,
"start": 2297
}
],
"value": "Intercontinental Exchange, Inc."
},
{
"name": "Listed_on",
"positions": [
{
"end": 86,
"start": 82
},
{
"end": 2334,
"start": 2330
}
],
"value": "NYSE"
},
{
"name": "Ticker",
"positions": [
{
"end": 91,
"start": 88
},
{
"end": 2339,
"start": 2336
}
],
"value": "ICE"
}
],
"namespace": "markets_tickers_indices_en",
"template": "PUBLIC_COMPANY_AND_TICKER"
},
{
"fields": [
{
"name": "Stock_Exchange",
"positions": [
{
"end": 2711,
"start": 2688
}
],
"value": "New York Stock Exchange"
}
],
"namespace": "markets_tickers_indices_en",
"template": "STOCK_EXCHANGE"
},
{
"fields": [
{
"name": "Index",
"positions": [
{
"end": 253,
"start": 219
},
{
"end": 368,
"start": 334
}
],
"value": "NYSE FANG+Daily 2x Leveraged Index"
}
],
"namespace": "markets_tickers_indices_en",
"template": "STOCK_INDEX"
},
{
"fields": [
{
"name": "Index",
"positions": [
{
"end": 477,
"start": 461
},
{
"end": 593,
"start": 577
}
],
"value": "NYSE FANG+ Index"
}
],
"namespace": "markets_tickers_indices_en",
"template": "STOCK_INDEX"
},
{
"fields": [
{
"name": "Index",
"positions": [
{
"end": 1663,
"start": 1654
}
],
"value": "ICE index"
}
],
"namespace": "markets_tickers_indices_en",
"template": "STOCK_INDEX"
}
]