IP Library › Granted Patent US 12,282,520
Granted Patent B2
US 12,282,520 · App. 18/454,132 · Granted Apr 22, 2025

Systems and methods for semantic keyword analysis

Inventors: Richard Israel Mallah (Boston, MA); Akos Lajos Balogh (Boston, MA)
Assignee: MarketMuse, Inc.
G06F16/9538G06F16/24575G06F16/24578G06F16/285G06F16/90324G06F16/90332G06F16/951G06F16/9535G06Q30/0256
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,282,520
App. No.
18/454,132
Granted
Apr 22, 2025
Kind
B2
Abstract

In various embodiments, a method for generating from one or more keywords a list of related topics for organic search includes receiving, by a topic tool, an input of one or more keywords for which to generate a list of related topics. The method may further include acquiring, by a crawler, content from a plurality of different web content sources via one or more networks. The method may also include applying, by the topic tool, to the acquired content an ensemble of one or more key phrase extraction algorithms, one or more graph analyses algorithms and one or more natural language processing algorithms to identify a set of semantically relevant topics scored by relevance. The method may also include generating, by the topic tool, from the set of semantically relevant topics, a knowledge graph of related topics for the input of the one or more keywords. The method may further include outputting, by the topic tool based at least partially on the knowledge graph, an enumerated list of topics ranked by at least a relevance score.

Claims (55)

1. A system for generating from one or more keywords a list of related topics for organic search, the system comprising:

a client computing device configured to:

send via a computer network connection, to a server, a request for keywords related to a topic or concept,

wherein, the server to which the request is sent utilizes:

a crawler configured to acquire content from a plurality of different web content sources via one or more computer network connections; and

a topic tool configured to execute on a processor on at least one server to:

receive an input including the request for keywords related to a topic or concept for which to generate a list of related topics;

apply to the acquired content an ensemble of algorithms, the ensemble comprising a predetermined sequence of:

one or more key phrase extraction algorithms to generate a set of keywords based on at least the acquired content,

one or more graph analyses algorithms to identify a set of topics semantically relevant to the set of keywords generated using the one or more key phrase extraction algorithms, and

one or more natural language processing algorithms to determine a relevance score for each topic of the set of semantically relevant topics;

generate from the set of semantically relevant topics, a knowledge graph of related topics for the input of the one or more keywords; and

receive via the computer network connection, from the server, output based at least partially on the knowledge graph, an enumerated list of topics ranked by at least the relevance score.

2. The system of claim 1 , wherein the server to which the request is sent further comprises a topic inventory tool configured to generate the input of one or more keywords from analyses of content from an identified web site.

3. The system of claim 1 , wherein the key phrase extraction algorithms comprise a Bayesian statistical ensemble.

4. The system of claim 1 , wherein the ensemble is further configured to perform a plurality of term ranking functions including one or more of the following: a core phrase term ranking function, a tail phrase term ranking function, a hyperdictionary graph traversal algorithm and a semantic knowledgebase path traversal score.

5. A method for generating from one or more keywords a list of related topics for organic search, the method comprising:

sending with a client device, via a computer network connection, to a server, a request for keywords related to a topic or concept,

wherein, the server to which the request is sent utilizes:

a crawler configured to acquire content from a plurality of different web content sources via one or more computer network connections; and

a topic tool configured to execute on a processor on at least one server to:

receive an input including the request for keywords related to a topic or concept for which to generate a list of related topics;

apply to the acquired content an ensemble of algorithms, the ensemble comprising a predetermined sequence of:

one or more key phrase extraction algorithms to generate a set of keywords based on at least the acquired content,

one or more graph analyses algorithms to identify a set of topics semantically relevant to the set of keywords generated using the one or more key phrase extraction algorithms, and

one or more natural language processing algorithms to determine a relevance score for each topic of the set of semantically relevant topics;

generate from the set of semantically relevant topics, a knowledge graph of related topics for the input of the one or more keywords; and

receiving with the client device, via the computer network connection, from the server, output based at least partially on the knowledge graph, an enumerated list of topics ranked by at least the relevance score.

6. The method of claim 5 , wherein the receiving an input further comprises receiving, by the topic tool, the input of one or more keywords from a topic inventory tool, the topic inventory tool generating the input keyword from analyses of content from an identified web site.

7. The method of claim 5 , wherein the acquiring further comprises acquiring content, by the crawler, from the plurality of different web content sources comprising web sites, news articles, blog posts and keyword data.

8. The method of claim 5 , wherein the acquiring further comprises cleansing and normalizing the acquired content.

9. The method of claim 5 , wherein the one or more key phrase extraction algorithms comprise a Bayesian statistical ensemble.

10. The method of claim 5 , wherein the applying further comprises performing a plurality of term ranking functions are performed including one or more of the following: a core phrase term ranking function, a tail phrase term ranking function, a hyperdictionary graph traversal algorithm and a semantic knowledgebase path traversal score.

11. The method of claim 5 , wherein the applying further comprises applying a weight to each of the one or more algorithms of the ensemble to generate the relevance score for the set of semantic relevance scored phrases.

12. The method of claim 5 , wherein the receiving output further comprises receiving the enumerated list of topics ranked by a measure of frequency comprising one or more of the following: frequency in page body, frequency in title, and number of pages where the topics occur.

13. The method of claim 5 , wherein the receiving output further comprises receiving the enumerated list of topics ranked by at least one of an attractiveness score, a volume score and a competition score.

14. The method of claim 5 , wherein the receiving output further comprises receiving the enumerated list of topics ranked by an estimated equivalent value associated with paid advertising.

15. The method of claim 5 , wherein the receiving output further comprises receiving the enumerated list of topics ranked by one or more of the following:

a measure of frequency, an attractiveness score, a volume score, and a competition score.

16. A non-transitory computer-readable storage medium comprising:

instructions that, when executed by a device comprising a processor, facilitate performance of generating topics for organic search operations comprising:

receiving, via a computer network connection, a request for keywords related to a topic or concept,

implementing a crawler configured to acquire content from a plurality of different web content sources via one or more computer network connections; and

implementing a topic tool to:

receive an input including the request for keywords related to a topic or concept for which to generate a list of related topics;

apply to the acquired content an ensemble of algorithms, the ensemble comprising a predetermined sequence of:

one or more key phrase extraction algorithms to generate a set of keywords based on at least the acquired content,

one or more graph analyses algorithms to identify a set of topics semantically relevant to the set of keywords generated using the one or more key phrase extraction algorithms, and

one or more natural language processing algorithms to determine a relevance score for each topic of the set of semantically relevant topics;

generate from the set of semantically relevant topics, a knowledge graph of related topics for the input of the one or more keywords; and

outputting, based at least partially on the knowledge graph, an enumerated list of topics ranked by at least the relevance score.

17. The medium of claim 16 , wherein the instructions for receiving an input further comprise instructions for receiving, by the topic tool, the input of one or more keywords from a topic inventory tool, the topic inventory tool generating the input keyword from analyses of content from an identified web site.

18. The medium of claim 16 , wherein the instructions for acquiring further comprise instructions for acquiring content, by the crawler, from the plurality of different web content sources comprising web sites, news articles, blog posts and keyword data.

19. The medium of claim 16 , wherein the one or more key phrase extraction algorithms comprise a Bayesian statistical ensemble.

20. The medium of claim 16 , wherein the instructions for applying further comprise instructions for performing a plurality of term ranking functions are performed including one or more of the following: a core phrase term ranking function, a tail phrase term ranking function, a hyperdictionary graph traversal algorithm and a semantic knowledgebase path traversal score.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 23, 2023
From: MALLAH, RICHARD ISRAEL; BALOGH, AKOS LAJOS
To: MARKETMUSE, INC.
Reel/Frame 064675/0442 →
Continuity (5)
Continuation 17188600 · Mar 1, 2021
Continuation 16523267 · Jul 26, 2019
Continuation 14928210 · Oct 30, 2015
Provisional Application 62073560 · Oct 31, 2014
Related Publication 20230394101A1 · Dec 7, 2023
References Cited (39)
US 8676798B1 · Datta et al. · 2014 [cited by applicant]
US 10162882B2 · Franceschini et al. · 2018 [cited by applicant]
US 11061913B2 · Eifert · 2021 [cited by examiner]
US 11074262B2 · Eifert · 2021 [cited by examiner]
US 11132988B1 · Steedman Henderson · 2021 [cited by examiner]
US 11270061B2 · Herzig · 2022 [cited by examiner]
US 11379540B2 · James · 2022 [cited by examiner]
US 11574522B1 · Mendell · 2023 [cited by examiner]
US 11615251B1 · Schilling · 2023 [cited by examiner]
US 20020129014A1 · Kim et al. · 2002 [cited by applicant]
US 20070174340A1 · Gross · 2007 [cited by applicant]
US 20090063468A1 · Berg · 2009 [cited by examiner]
US 20100030648A1 · Manolescu et al. · 2010 [cited by applicant]
US 20110213655A1 · Henkin · 2011 [cited by examiner]
US 20110288931A1 · Kuhn et al. · 2011 [cited by applicant]
US 20120226713A1 · Park et al. · 2012 [cited by applicant]
US 20160125087A1 · Mallah · 2016 [cited by examiner]
US 20160232162A1 · Dzumla · 2016 [cited by examiner]
US 20170235848A1 · Van Dusen et al. · 2017 [cited by applicant]
US 20180366013A1 · Arvindam · 2018 [cited by applicant]
US 20200380043A1 · Miyazawa · 2020 [cited by examiner]
US 20210065569A1 · Arvindam · 2021 [cited by examiner]
US 20210182352A1 · Mallah · 2021 [cited by examiner]
US 20210294974A1 · Gao · 2021 [cited by examiner]
US 20220415290A1 · Jeffrey · 2022 [cited by examiner]
Non-Final Office Action received for U.S. Appl. No. 17/188,600 dated Jan. 5, 2023, 30 pages. [cited by applicant]
Final Office Action received for U.S. Appl. No. 14/928,210 dated Oct. 25, 2018, 45 pages. [cited by applicant]
International Preliminary Report on Patentability received for PCT Application Serial No. PCT/US2015/058289 dated May 11, 2017, 7 pages. [cited by applicant]
Notice of Allowance received for U.S. Appl. No. 14/928,210 dated May 1, 2019, 39 pages. [cited by applicant]
Non-Final Office Action received for U.S. Appl. No. 14/928,210 dated May 3, 2018, 38 pages. [cited by applicant]
International Search Report and Written Opinion received for PCT Application Serial No. PCT/US2015/058289 dated Feb. 2, 2016, 8 pages. [cited by applicant]
International Search Report and Written Opinion received for PCT Application Serial No. PCT/US2015/058285 dated Feb. 1, 2016, 9 pages. [cited by applicant]
Kim et al. “Retrieving and Ranking Methods for Finding Match Candidates”, International Conference on Complex, Intelligent, and Software Intensive Systems, 2011, 5 pages. [cited by applicant]
Non-Final Office Action received for U.S. Appl. No. 14/928,220 dated Feb. 25, 2019, 85 pages. [cited by applicant]
Notice of Allowance received for U.S. Appl. No. 14/928,220 dated Jul. 25, 2019, 74 pages. [cited by applicant]
Geadas et al. “Ensemble Learning for Keyword Extraction from Event Descriptions”, International Joint Conference on Neural Networks (IJCNN), Jul. 6-11, 2014, 8 pages. [cited by applicant]
Zhu et al., “Combination of Unsupervised Keyphrase Extraction Algorithms”, International Conference on Asian Language Processing, 2013, 4 pages. [cited by applicant]
Non final office action received for U.S. Appl. No. 16/523,267 dated Dec. 7, 2020, 73 Pages. [cited by applicant]
Notice of Allowance received for U.S. Appl. No. 17/188,600 dated May 22, 2023, 50 pages. [cited by applicant]