Computer-based systems and methods for determining interest levels of consumers in research work product produced by a research department

• US 8,694,413 B1
• Filed: 02/23/2012
• Issued: 04/08/2014
• Est. Priority Date: 09/29/2011
• Status: Active Grant

• Current Assignee: Morgan Stanley Services Group Inc. (Morgan Stanley)
• Original Assignee: Morgan Stanley & Company LLC (Morgan Stanley)
• Inventors: Hurewitz, Barry S., Harvey, Vaughn S., Treder, Robert, Solomon, Scott, Moustafa, Omar, Rachidi, Thami
• Primary Examiner(s): Khattar, Rajesh
• Assistant Examiner(s): See, Carol
• Application Number:
  US13/402,998
• Time in Patent Office: 775 Days
• Field of Search: None
• US Class Current: 705/37
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

Computer-based systems and methods that determine interest profiles of contacts of work product. The contact interest profiles may indicate the contacts'"'"' interests in particular topics of research work product and/or analyst teams within the research department that produced the research work product. The contacts'"'"' interest profiles may be determined based on research work product accessed by the contacts according to topic and/or analyst team that produced the research work product, the interaction of the contacts with the research department, and any other reliable information indicative of the contacts'"'"' topic and/or team interests. Also, research work product items produced by the entity may be rated for each contact based on, for example, the topic(s) of the research work product items and the analyst teams that produced the research work product items, as well as the contacts'"'"' topic and/or analyst team interests.

Claims

1. A computer system for computing an interest score for one or more contacts of a research entity that produces research work product items that individually cover one or more of a plurality of topics, wherein the topics comprise at least one of publicly-traded stocks, industries, and geographic regions, wherein the one or more contacts are associated with accounts for holding and trading publicly-traded securities, and wherein the research entity comprises a plurality of teams that produce the research work product items, the computer system comprising:

a computer-based data storage system that stores data relating to the contacts and the research work product items produced by the entity;

at least one processor in communication with the computer-based data storage system; and

at least one computer memory device in communication with the processor, wherein the at least one computer memory device stores instructions that when executed by the at least one processor causes the at least one processor to compute a work product item interest scores for the one or more contacts for one or more work product items produced by the research entity, wherein the work product item interest score for a particular contact and a particular work product item is indicative of the particular contact'"'"'s likely interest in the particular work product item, wherein the at least one processor is programmed to compute the work product item interest scores by performing steps comprising;

determining for the one or more contacts at least one of the following sub-interest scores;

a stock sub-interest score for one or more of the stocks, wherein the stock sub-interest score for a stock for a contact is a score indicative of the contact'"'"'s interest in accessing research pertaining to the stock; and

a team sub-interest score for one or more analyst teams of the research entity, wherein the team sub-interest score for an analyst team for a contact is a score indicative of the contact'"'"'s interest in accessing research from the analyst team;

wherein the sub-interest scores are determined using one or more mathematical models, wherein the one or more mathematical models uses a plurality of derived variables that are indicative of the one or more contacts'"'"' interest in the stocks and/or analyst teams, wherein the plurality of derived variables comprise at least one of the following derived variable;

a first derived variable that is based on data regarding the research work product items produced by the entity that were accessed by the one or more contacts;

a second derived variable that is based on data regarding communication interactions between the one or more contacts and the teams of the research entity;

a third derived variable that is based on broker votes by the one or more contacts;

a fourth derived variable that is based on stock trades of the accounts associated with the one or more contacts; and

a fifth derived variable that is based on stock holdings of the accounts associated with the one or more contacts;

computing the work product interest scores for the one or more contacts for work product items produced by the entity, wherein the work product interest score for one contact for one work product item is computed by summing a plurality of weighted summands, wherein the plurality of weighted summands comprise the at least one of the;

the stock sub-interest score for the contact for the one or more stocks to which the work product item pertains; and

the team sub-interest score for the contact for the team that generated the work product item, wherein;

data indicating the one or more stocks covered by each research work product item produced by the research entity are stored in the computer-based data storage system; and

data indicating which team produced each research work product item produced by the research entity are stored in the computer-based data storage system, and

wherein the work product interest scores are for use by the research entity in notifying the one or more contacts about the work product items.

2. The computer system of claim 1, wherein:

the computer-based data storage system further stores;

trade data regarding trades of stocks by the accounts associated with the contacts; and

holdings data regarding holdings of stocks by the accounts associated with the contacts.

3. The computer system of claim 1, wherein the one or more mathematical models comprise one or more propensity models.

4. The computer system of claim 3, wherein the one or more propensity models are trained by blending two or more continuous regression-type models.

5. The computer system of claim 3, wherein the one or more propensity models comprise a blend of two or more propensity models.

6. The computer system of claim 1, wherein:

determining at least one of the stock sub-interest score and the team sub-interest score comprises determining both the stock sub-interest score and the team sub-interest score; and

the plurality of weighted summands used to compute the work product interest scores comprises both the stock sub-interest score and the team sub-interest score.

7. The computer system of claim 1, wherein the plurality of derived variables used by the one or more mathematical models comprises at least two of the first derived variable, the second derived variable, the third derived variable, the fourth derived variable, and the fifth derived variable.

8. The computer system of claim 7, wherein the plurality of derived variables used by the one or more mathematical models comprises at least the first and second derived variables.

9. The computer system of claim 8, wherein the plurality of derived variables used by the one or more mathematical models comprises at least a sixth derived variable that is based on data regarding a type of the research work product items, wherein the types of research work product items include at least one of:

research work product items that change a rating for a stock;

research work product items that identify a target price movement for a stock; and

research work product items that change an earnings estimate for a stock.

10. A computer-implemented method for computing an interest score for one or more contacts of a research entity that produces research work product items that individually cover one or more of a plurality of topics, wherein the topics comprise at least one of publicly-traded stocks, industries, and geographic regions, wherein the one or more contacts are associated with accounts for holding and trading publicly-traded securities, and wherein the research entity comprises a plurality of teams that produce the research work product items, the method comprising:

storing data in a computer-based data storage system, wherein the stored data comprises data relating to the one or more contacts and the research work product items produced by the entity;

computing, by a computer system in communication with the computer-based data storage system, a work product item interest scores for the one or more contacts for one or more work product items produced by the research entity, wherein the work product item interest score for a particular contact and a particular work product item is indicative of the particular contact'"'"'s likely interest in the particular work product item, wherein the work product item interest scores are computed by performing steps comprising;

determining for the one or more contacts at least one of the following sub-interest scores;

a stock sub-interest score for one or more of the stocks, wherein the stock sub-interest score for a stock for a contact is a score indicative of the contact'"'"'s interest in accessing research pertaining to the stock; and

a team sub-interest score for one or more analyst teams of the research entity, wherein the team sub-interest score for an analyst team for a contact is a score indicative of the contact'"'"'s interest in accessing research from the analyst team;

wherein the sub-interest scores are determined using one or more mathematical models, wherein the one or more mathematical models uses a plurality of derived variables that are indicative of the one or more contacts'"'"' interest in the stocks and/or analyst teams, wherein the plurality of derived variables comprise at least one of the following derived variable;

a first derived variable that is based on data regarding the research work product items produced by the entity that were accessed by the one or more contacts;

a second derived variable that is based on data regarding communication interactions between the one or more contacts and the teams of the research entity;

a third derived variable that is based on broker votes by the one or more contacts;

a fourth derived variable that is based on stock trades of the accounts associated with the one or more contacts; and

a fifth derived variable that is based on stock holdings of the accounts associated with the one or more contacts;

computing the work product interest scores for the one or more contacts for work product items produced by the entity, wherein the work product interest score for one contact for one work product item is computed by summing a plurality of weighted summands, wherein the plurality of weighted summands comprise the at least one of;

the stock sub-interest score for the contact for the one or more stocks to which the work product item pertains; and

the team sub-interest score for the contact for the team that generated the work product item, wherein;

data indicating the one or more stocks topics covered by each research work product item produced by the research entity are stored in the computer-based data storage system; and

data indicating which team produced each research work product item produced by the research entity are stored in the computer-based data storage system, and

wherein the work product interest scores are for use by the research entity in notifying the one or more contacts about the work product items.

11. The method of claim 10, wherein:

trade data regarding trades of stocks by the accounts associated with the contacts are stored in the computer-based data storage system; and

holdings data regarding holdings of stocks by the accounts associated with the contacts are stored in the computer-based data storage system.

12. The method of claim 10, wherein the one or more mathematical models comprise one or more propensity models.

13. The method of claim 12, wherein the one or more propensity models comprise a blend of two or more propensity models.

14. The method of claim 10, wherein:

determining at least one of the stock sub-interest score and the team sub-interest score comprises determining both the stock sub-interest score and the team sub-interest score; and

the plurality of weighted summands used to compute the work product interest scores comprises both the stock sub-interest score and the team sub-interest score.

15. The method of claim 10, wherein the plurality of derived variables used by the one or more mathematical models comprises at least two of the first derived variable, the second derived variable, the third derived variable, the fourth derived variable, and the fifth derived variable.

16. The method of claim 15, wherein the plurality of derived variables used by the one or more mathematical models comprises at least the first and second derived variables.

17. The method of claim 16, wherein the plurality of derived variables used by the one or more mathematical models comprises at least a sixth derived variable that is based on data regarding a type of the research work product items, wherein the types of research work product items include at least one of:

research work product items that change a rating for a stock;

research work product items that identify a target price movement for a stock; and

research work product items that change an earnings estimate for a stock.