Method and system for improving search results by recency boosting customer support content for a customer self-help system associated with one or more financial management systems

• US 10,552,843 B1
• Filed: 12/05/2016
• Issued: 02/04/2020
• Est. Priority Date: 12/05/2016
• Status: Active Grant

• Current Assignee: Intuit, Inc.
• Original Assignee: Intuit, Inc.
• Inventors: Podgorny, Igor A., Indyk, Benjamin, Goodyear, Todd Frey, Ma, Irene F.
• Primary Examiner(s): Peng, Huawen A
• Application Number:
  US15/368,806
• Time in Patent Office: 1,156 Days
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06Q 30/01 Customer relationship services
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/306 User profiles

Abstract

Disclosed methods and systems improve search results by recency boosting customer support content for a customer self-help system associated with one or more financial management systems. The customer self-help system retrieves content relevance from a variety of sources, such as media outlets, taxation agencies and news feeds for the financial management system. The customer self-help system generates content relevance weights from the content relevance data, and applies the content relevance weights to customer support content maintained by the customer self-help system. In response to receiving a search query from a user, the customer self-help system provides relevant portions of customer support content that has been recency boosted (e.g., adjusted by the content relevance weights), to increase the likelihood that the customer support content provided to the user is relevant to the user'"'"'s search query.

Claims

1. A computing system implemented method for improving search results by recency boosting customer support content for a customer self-help system associated with one or more financial management systems, comprising:

providing, with one or more computing systems, a customer self-help system associated with a financial management system;

storing, with the customer self-help system, customer support content data representing customer support content, in memory allocated for use by the customer self-help system;

determining content relevance weight data for the customer support content data by applying content relevance data to a content relevance analytics model, the content relevance data being acquired by the customer self-help system from one or more sources of content relevance data, the content relevance weight data representing content relevance weights;

determining topic term probabilities, wherein the topic term probabilities are for topic terms that are assigned to individual portions of the customer support content represented by the customer support content data;

determining context characteristics probabilities, wherein the context characteristics probabilities represent a probabilistic relationship between topics and user context characteristics data;

receiving search query terms data, by the customer self-help system, from a user of the customer self-help system, the search query terms data representing one or more search query terms that formulate at least part of a search query;

providing search algorithm data representing a search algorithm that matches the search query with portions of the customer support content that are topically related to the search query;

applying search query data to the search algorithm data to identify the portions of the customer support content data that are topically related to the search query, the search query data including the search query terms data;

determining context characteristics associated with the user of the customer self-help system, wherein the context characteristics are present during the submission of the search query terms data, and further wherein the context characteristics include user financial data characteristics and user profile characteristics;

determining topic relevance scores for customer support content associated with the search query by summing the topic term probabilities associated with topic terms in the identified portions of the customer support content data and the context characteristics probabilities for the context characteristics associated with the user;

determining weighted relevance scores for customer support content associated with the search query by multiplying the topic relevance scores by the content relevance weights associated with the customer support content;

applying the weighted relevance scores to the customer support content data to transform the customer support content into recency boosted customer support content data, the recency boosted customer support content data representing recency boosted customer support content;

updating user experience display data to include portions of the recency boosted customer support content data that are topically related to the search query, the user experience display data representing a user experience display; and

transmitting the user experience display data to a user computing system, to display the portions of the recency boosted customer support content data that are topically related to the search query, in response to the search query terms data received from the user, to assist the user in finding an answer to the search query.

2. The computing system implemented method of claim 1, further comprising:

maintaining user financial data representing financial information for users of the financial management system; and

maintaining user profile data representing user profiles for users of the financial management system,

wherein the search query data includes the user profile data and the user financial data.

3. The computing system implemented method of claim 1, wherein the customer self-help system is part of the financial management system.

4. The computing system implemented method of claim 1, wherein the financial management system is selected from a group of financial management systems including:

a tax return preparation system, a business financial management system, and a personal financial management system.

5. The computing system implemented method of claim 1, further comprising:

determining topic relevance scores for topics associated with the customer support content data;

sorting the topic relevance scores; and

selecting the portions of the customer support content data that is topically relevant to the search query, at least partially based on the topic relevance scores.

6. The computing system implemented method of claim 1, wherein the search algorithm includes an analytics model that is a probabilistic topic model.

7. The computing system implemented method of claim 6, wherein the probabilistic topic model determines a topic relevance between the customer support content and the one or more search query terms, by identifying topics that are associated with each content entry of the customer support content and by identifying topics that are associated with the one or more search query terms.

8. The computing system implemented method of claim 6, wherein the probabilistic topic model includes one or more of a Latent Dirichlet Allocation algorithm, a Latent Semantic Indexing (“

LSI”

) algorithm, a query clustering algorithm, and a query de-duplication algorithm.

9. The computing system implemented method of claim 1, further comprising:

acquiring, with the customer self-help system, the content relevance data from one or more sources of content relevance data by;

retrieving electronic media feeds from the one or more sources of content relevance data; and

applying contents of the electronic media feeds to a probabilistic topic model to determine topics of the electronic media feeds;

ranking the topics of the electronic media feeds by prevalence within the electronic media feeds; and

defining content relevance weight data for the topics of the electronic media feeds at least partially based on said ranking the topics of the electronic media feeds.

10. The computing system implemented method of claim 1, wherein sources of the content relevance data includes one or more of the financial management system, the customer self-help system, a taxation agency, and media outlets.

11. The computing system implemented method of claim 1, wherein the content relevance data is selected from a group of content relevance data including:

product news data, search trends data, tax law updates data, and financial news data.

12. The computing system implemented method of claim 1, wherein the search algorithm includes an analytics model that is a predictive model.

13. The computing system implemented method of claim 12, wherein the predictive model is trained using a training operation that is selected from a group of predictive model training operations, consisting of:

regression;

logistic regression;

decision trees;

artificial neural networks;

support vector machines;

linear regression;

nearest neighbor methods;

distance based methods;

naive Bayes;

linear discriminant analysis; and

k-nearest neighbor algorithm.

14. The computing system implemented method of claim 1, wherein the customer support content is selected from a group of customer support content, consisting of:

crowd-sourced customer support content;

service provider content;

definitions content, wherein the definitions content includes explanations of acronyms of tax-specific terms, definitions of tax-specific terms, or the acronyms of tax-specific terms and the definitions of tax-specific terms;

potential search queries; and

user financial data for the user.

15. A computing system implemented method for improving search results by recency boosting customer support content for a customer self-help system associated with one or more financial management systems, comprising:

providing, with one or more computing systems, a customer self-help system associated with a financial management system;

storing, with the customer self-help system, customer support content data representing customer support content, in memory allocated for use by the customer self-help system;

acquiring, with the customer self-help system, content relevance from one or more sources of content relevance data;

determining content relevance weight data for the customer support content data by applying the content relevance data to a content relevance analytics model, the content relevance weight data representing content relevance weights, the content relevance analytics model being at least partially based on a probabilistic topic model;

determining topic term probabilities, wherein the topic term probabilities are for topic terms that are assigned to individual portions of the customer support content represented by the customer support content data;

determining context characteristics probabilities, wherein the context characteristics probabilities represent a probabilistic relationship between topics and user context characteristics data;

receiving search query terms data, by the customer self-help system, from a user of the customer self-help system, the search query terms data representing one or more search query terms that formulate at least part of a search query;

providing search algorithm data representing a search algorithm that matches the search query with portions of the customer support content, at least partially based on topics that are common to both the search query and the portions of the customer support content;

applying search query data to the search algorithm data to identify the portions of the customer support content data that are topically related to the search query, the search query data including the search query terms data;

determining context characteristics associated with the user of the customer self-help system, wherein the context characteristics are present during the submission of the search query terms data, and further wherein the context characteristics include user financial data characteristics and user profile characteristics;

determining topic relevance scores for customer support content associated with the search query by summing the topic term probabilities associated with topic terms in the identified portions of the customer support content data and the context characteristics probabilities for the context characteristics associated with the user;

determining weighted relevance scores for customer support content associated with the search query by multiplying the topic relevance scores by the content relevance weights associated with the customer support content;

applying the weighted relevance scores to the customer support content data to transform the customer support content into recency boosted customer support content data, the recency boosted customer support content data representing recency boosted customer support content;

updating user experience display data to include portions of the recency boosted customer support content data that are topically related to the search query, the user experience display data representing a user experience display; and

transmitting the user experience display data to a user computing system, to display the portions of the recency boosted customer support content data that are topically related to the search query, in response to the search query terms data received from the user, to assist the user in finding an answer to the search query.

16. The computing system implemented method of claim 15, wherein the financial management system is selected from a group of financial management systems including:

a tax return preparation system, a business financial management system, and a personal financial management system.

17. The computing system implemented method of claim 15, further comprising:

determining topic relevance scores for topics associated with the customer support content data;

sorting the topic relevance scores; and

selecting the portions of the customer support content data that are topically relevant to the search query, at least partially based on the topic relevance scores.

18. The computing system implemented method of claim 15, wherein the search algorithm includes an analytics model that is a second probabilistic topic model.

19. The computing system implemented method of claim 18, wherein the second probabilistic topic model includes one or more of a Latent Dirichlet Allocation algorithm, a Latent Semantic Indexing (“

LSI”

) algorithm, a query clustering algorithm, and a query de-duplication algorithm.

20. The computing system implemented method of claim 15, wherein sources of the content relevance data includes one or more of the financial management system, the customer self-help system, a taxation agency, and media outlets.

21. The computing system implemented method of claim 15, wherein the content relevance data is selected from a group of content relevance data including:

product news data, search trends data, tax law updates data, and financial news data.

22. The computing system implemented method of claim 15, wherein the customer support content is selected from a group of customer support content, consisting of:

crowd-sourced customer support content;

service provider content;

definitions content, wherein the definitions content includes explanations of acronyms of tax-specific terms, definitions of tax-specific terms, or the acronyms of tax-specific terms and the definitions of tax-specific terms;

potential search queries; and

user financial data for the user.

23. A system that increases a likelihood of providing relevant search results by recency boosting customer support content for a customer self-help system associated with one or more financial management systems, comprising:

one or more processors; and

memory having instructions which, if executed by the one or more processors, cause the one or more processors to perform a process for improving search results by recency boosting customer support content for a customer self-help system associated with one or more financial management systems, the process comprising;

providing, a customer self-help system associated with a financial management system;

storing, with the customer self-help system, customer support content data representing customer support content, in memory allocated for use by the customer self-help system;

determining content relevance weight data for the customer support content data by applying content relevance data to a content relevance analytics model, the content relevance data being acquired by the self-help system from one or more sources of content relevance data, the content relevance weight data representing content relevance weights;

determining topic term probabilities, wherein the topic term probabilities are for topic terms that are assigned to individual portions of the customer support content represented by the customer support content data;

determining context characteristics probabilities, wherein the context characteristics probabilities represent a probabilistic relationship between topics and user context characteristics data;

receiving search query terms data, by the customer self-help system, from a user of the customer self-help system, the search query terms data representing one or more search query terms that formulate at least part of a search query;

providing search algorithm data representing a search algorithm that matches the search query with portions of the customer support content that are topically related to the search query;

applying search query data to the search algorithm data to identify the portions of the customer support content data that are topically related to the search query, the search query data including the search query terms data;

determining context characteristics associated with the user of the customer self-help system, wherein the context characteristics are present during the submission of the search query terms data, and further wherein the context characteristics include user financial data characteristics and user profile characteristics;

determining topic relevance scores for customer support content associated with the search query by summing the topic term probabilities associated with topic terms in the identified portions of the customer support content data and the context characteristics probabilities for the context characteristics associated with the user;

determining weighted relevance scores for customer support content associated with the search query by multiplying the topic relevance scores by the content relevance weights associated with the customer support content;

applying the weighted relevance scores to the customer support content data to transform the customer support content into recency boosted customer support content data, the recency boosted customer support content data representing recency boosted customer support content;

updating user experience display data to include portions of the recency boosted customer support content data that are topically related to the search query, the user experience display data representing a user experience display; and

transmitting the user experience display data to a user computing system, to display the portions of the recency boosted customer support content data that are topically related to the search query, in response to the search query terms data received from the user, to assist the user in finding an answer to the search query.