Determining importance of investment identifier to content of content item
In some implementations, a computing system can determine how important an investment identifier is to a content item that mentions the investment identifier. For example, a content item may describe a story, event, etc., related to an investment identifier. The content item may mention the investment identifier by mentioning the investment identifier, proxies for the investment identifier, or other equivalents associated with the investment identifier. The computing system can determine locations in the content item where the investment identifier is mentioned and/or how frequently the investment identifier is mentioned. Based on the locations and/or frequency of mentions, the computing system can determine an importance score that represents how important the investment identifier is to the story described by the content item. The importance score can be stored in metadata for the content item and used when determining which content items to present to a user.
1. A method comprising:
receiving, by a server device, a content item including content;
obtaining, by the server device, a dictionary of terms, each term in the dictionary of terms associated with a respective investment identifier;
determining, by the server device, locations of occurrences of the terms within the content of the content item;
generating, by the server device, location characteristics for each location of the locations of occurrences of the terms within the content of the content item;
generating, by the server device, importance scores for each investment identifier associated with the terms occurring within the content item based on the location characteristics.
2. The method of claim 1 , further comprising:
determining that a particular investment identifier has been occurred within the content of the content item when the particular investment identifier or an equivalent to the investment identifier is located within the content item.
3. The method of claim 2 , wherein an equivalent to the investment identifier includes a name of a person, a name of a product, nickname, or an alias associated with the investment identifier.
4. The method of claim 1 , wherein the location characteristics include a character count from a beginning of the content of the content item.
5. The method of claim 1 , wherein the location characteristics include a first location of an occurrence of a first investment identifier relative to a second location of an occurrence of a second investment identifier within the content item.
6. The method of claim 1 , wherein the location characteristics include a structural element of the content item in which the term occurs, where the structural element corresponds to a title, a paragraph, a first line in a paragraph, a last line in a paragraph, an abstract section, a summary section, or a combination thereof.
7. The method of claim 1 , wherein the importance scores for each of the investment identifiers are generated based on a machine learning model having features corresponding to the location characteristics of each dictionary term occurring in the content item.
8. A non-transitory computer readable medium including one or more sequences of instructions that, when executed by one or more processors, cause the processors to perform operations comprising:
receiving, by a server device, a content item including content;
obtaining, by the server device, a dictionary of terms, each term in the dictionary of terms associated with a respective investment identifier;
determining, by the server device, locations of occurrences of the terms within the content of the content item;
generating, by the server device, location characteristics for each location of the locations of occurrences of the terms within the content of the content item;
generating, by the server device, importance scores for each investment identifier associated with the terms occurring within the content item based on the location characteristics.
9. The non-transitory computer readable medium of claim 8 , wherein the instructions cause the processors to perform operations comprising:
determining that a particular investment identifier has been occurred within the content of the content item when the particular investment identifier or an equivalent to the investment identifier is located within the content item.
10. The non-transitory computer readable medium of claim 9 , wherein an equivalent to the investment identifier includes a name of a person, a name of a product, nickname, or an alias associated with the investment identifier.
11. The non-transitory computer readable medium of claim 8 , wherein the location characteristics include a character count from a beginning of the content of the content item.
12. The non-transitory computer readable medium of claim 8 , wherein the location characteristics include a first location of an occurrence of a first investment identifier relative to a second location of an occurrence of a second investment identifier within the content item.
13. The non-transitory computer readable medium of claim 8 , wherein the location characteristics include a structural element of the content item in which the term occurs, where the structural element corresponds to a title, a paragraph, a first line in a paragraph, a last line in a paragraph, an abstract section, a summary section, or a combination thereof.
14. The non-transitory computer readable medium of claim 8 , wherein the importance scores for each of the investment identifiers are generated based on a machine learning model having features corresponding to the location characteristics of each dictionary term occurring in the content item.
15. A system comprising:
one or more processors; and
a non-transitory computer readable medium including one or more sequences of instructions that, when executed by the one or more processors, cause the processors to perform operations comprising:
receiving, by a server device, a content item including content;
obtaining, by the server device, a dictionary of terms, each term in the dictionary of terms associated with a respective investment identifier;
determining, by the server device, locations of occurrences of the terms within the content of the content item;
generating, by the server device, location characteristics for each location of the locations of occurrences of the terms within the content of the content item;
generating, by the server device, importance scores for each investment identifier associated with the terms occurring within the content item based on the location characteristics.
16. The system of claim 15 , wherein the instructions cause the processors to perform operations comprising:
determining that a particular investment identifier has been occurred within the content of the content item when the particular investment identifier or an equivalent to the investment identifier is located within the content item.
17. The system of claim 16 , wherein an equivalent to the investment identifier includes a name of a person, a name of a product, nickname, or an alias associated with the investment identifier.
18. The system of claim 15 , wherein the location characteristics include a character count from a beginning of the content of the content item.
19. The system of claim 15 , wherein the location characteristics include a first location of an occurrence of a first investment identifier relative to a second location of an occurrence of a second investment identifier within the content item.
20. The system of claim 15 , wherein the location characteristics include a structural element of the content item in which the term occurs, where the structural element corresponds to a title, a paragraph, a first line in a paragraph, a last line in a paragraph, an abstract section, a summary section, or a combination thereof.
21. The system of claim 15 , wherein the importance scores for each of the investment identifiers are generated based on a machine learning model having features corresponding to the location characteristics of each dictionary term occurring in the content item.