Customizable, real time intelligence channel

• US 10,223,465 B2
• Filed: 07/06/2017
• Issued: 03/05/2019
• Est. Priority Date: 03/11/2013
• Status: Active Grant

• Current Assignee: CREOpoint, Inc.
• Original Assignee: CREOpoint, Inc.
• Inventors: Goldenstein, Jean-Claude, Koemets, Kaimar, Fine, Ted, Jarv, Janno, Finn, Edward J.
• Primary Examiner(s): Hwa, Shyue Jiunn
• Application Number:
  US15/642,890
• Publication Number:
  US 20170308614A1
• Time in Patent Office: 607 Days
• Field of Search: 707722
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A customizable intelligence channel may be uniquely designed and implemented to contain only that information deemed to be relevant and material to a specific topic. The topic-related information provided via the intelligence channel may contain no extraneous results, irrelevant, and/or redundant information. The intelligence channel may be dynamically updated in real time, based on user input and/or crowdsourced data, to provide topic-related information.

Claims

1. A method of creating a customizable real-time intelligence channel, comprising:

determining a plurality of possible sources of information on a selected topic for the real-time intelligence channel;

receiving a plurality of topic-related keywords for a search query for each source of the plurality of sources, wherein each keyword of the plurality of topic-related keywords is related to the selected topic and is used to determine whether or not to include or exclude search results including the keyword;

receiving search results from at least one search utilizing the topic-related keywords to identify topic-related information within said plurality of sources of information;

selecting from the plurality of sources, based on the searching, at least one topic-related source comprising topic-related information;

a data processor implementing a filtering algorithm to pass items of information from the at least one selected topic related source based on a weighted influence and relevance ranking of each item of information from the at least one topic related source to the selected topic, wherein;

an influence score of each item of information from the selected topic-related source is calculated based on source online presence data obtained from social networks for each source and weighted based on at least one of academic research, publications, and information available online and offline and adjusted based on crowd-sourced information, and

a relevance score of each item of information from the selected topic-related source is calculated based on a similarity between the topic search query and the topic-related information and topic-related source and at least one of;

(1) a number of mentions of topic-related keywords in a headline of an item of information,

(2) a ratio of topic-related keywords to other words in the headline of the item of information,

(3) a ratio of topic related keywords to other words besides the headline in the item of information,

(4) a number of mentions of topic-related keywords of the item of information compared to irrelevant mentions of non-topic relevant information,

(5) a number of mentions of topic-related keywords in the topic-related source compared to irrelevant mentions of non-topic relevant information, and

(6) views and curation by users of a particular intelligence channel on at least one of particular items of information and items of information sourced from at least one of particular relevant sources and authors,

wherein implementing the filtering algorithm comprises filtering the topic-related information from the at least one selected topic-related source to remove at least one of an advertisement, information in a non-designated language, and duplicative information;

said data processor determining whether to pass each item of information from the at least one selected topic-related source through the filtering algorithm based on the weighted influence and relevance ranking of each item of information from the at least one selected topic-related source and reliability of the at least one topic related source, wherein relevance of search results from the topic-related source to the selected topic is based on actions of at least one user in at least one of (a) at least one of visiting, banning, and curating the source of the filtered topic-related information and (b) at least one of rating and deleting an item of information from the topic-related source; and

the data processor assembling the items of information that pass through said filtering algorithm into a data stream and presenting the assembled data stream of topic-related information via an interactive interface as the customizable real-time intelligence channel for the selected topic.

2. The method of claim 1, wherein the relevance of search results from the topic-related source to the selected topic is based on at least one of crowd-sourced data including data views and curation by users of the customizable real-time intelligence channel of the selected topic-related source of at least one of particular items of information and sources of items of information from at least one of particular topic-related sources and particular authors.

3. The method of claim 1, further comprising:

respectively formatting the topic-related keywords in accordance with a format of each source of the plurality of sources of information on the selected topic; and

searching each source of the plurality of sources of information on the selected topic utilizing the respectively formatted topic-related keywords.

4. The method of claim 1, further comprising at least one of updating and curating the plurality of topic-related keywords and information sources based on at least one of crowdsourcing data, research by subject matter experts, and input from the user.

5. The method of claim 4, wherein at least one of updating and curating the plurality of topic-related keywords and information sources comprises at least one of providing a source, removing a source, combining sources, and adjusting filtering by the filtering algorithm.

6. The method of claim 1, further comprising determining at least one of the selected topic and a customizable real-time intelligence channel for a user based on predictive data mining of information related to the user.

7. The method of claim 1, the data processor further filtering noise from the topic-related information to remove profanity and other items of information that at least one of do not add value, are of limited relevance to the topic, and that lack substantive content.

8. The method of claim 1, wherein the topic-related information is ranked based on at least one of the influence and relevance of topic-related information returned in search results for the selected topic and crowdsourced information developed through aggregate actions of users in at least one of banning and curating a topic-related source of the customizable real-time intelligence channel for the selected topic.

9. The method of claim 1, wherein filtering topic-related information from the at least one selected topic-related source for relevance to the topic comprises checking a source blacklist to remove information from banned sources.

10. The method of claim 9, further comprising obtaining at least one of source online presence metrics and context to said customizable real-time intelligence channel from at least one of social networks and a third party service for each source, said source online presence metrics including influence and relevance to the topic of the customizable real-time intelligence channel and at least one of a number of posts the source has shared, a number of social media followers, and at least one of a number of retweets, likes, and reposts.

11. The method of claim 1, further comprising performing semantic analysis of sources of information to select at least one of photographs, places, names, companies, addresses, and phone numbers to generate enhanced source content and using natural language processing to highlight elements of interest in said enhanced source content.

12. The method of claim 1, further comprising associating a time with storage of search results whereby users selecting the customizable real-time intelligence channel within a predetermined time interval may receive stored information and users selecting the customizable real-time intelligence channel outside of the predetermined time interval trigger may receive a new search inquiry.

13. The method of claim 1, further comprising combining customizable real-time intelligence channels.

14. The method of claim 1, wherein determining whether or not to include or exclude search results includes evaluating information indicative of a false positive search result.

15. The method of claim 1, further comprising filtering said plurality of sources by comparing a dictionary of continuously updated domain names and social media usernames with a source URL of each source and, when there is a match, at least one of approving the source and tagging the source for further review, but when there is not a match, at least one of removing at least one of the source and information from the stream and tagging the source for further review.

16. The method of claim 1, wherein the weighted influence and relevance ranking of each item of information from the selected topic-related source that may pass through the filtering algorithm is controllable by a user by moving an interactive interface element to select more or less filtering of each item of information from the selected topic-related source.

17. A system for creating a customizable real-time intelligence channel, comprising:

a data processor;

an interactive interface; and

a memory coupled to the data processor, the memory including executable instructions that when executed by the data processor cause the data processor to effectuate operations comprising;

determining a plurality of possible sources of information on a selected topic for the real-time intelligence channel;

receiving a plurality of topic-related keywords for a search query for each source of the plurality of sources, wherein each keyword of the plurality of topic-related keywords is related to the selected topic and is used to determine whether or not to include or exclude search results including the keyword;

receiving search results from at least one search utilizing the topic-related keywords to identify topic-related information within said plurality of sources of information;

selecting from the plurality of sources, based on the searching, at least one topic-related source comprising topic-related information;

implementing a filtering algorithm to pass items of information from the at least one selected topic related source based on a weighted influence and relevance ranking of each item of information from the at least one topic related source to the selected topic, wherein;

an influence score of each item of information from the selected topic-related source is calculated based on source online presence data obtained from social networks for each source and weighted based on at least one of academic research, publications, and information available online and offline and adjusted based on crowd-sourced information, and

a relevance score of each item of information from the selected topic-related source is calculated based on a similarity between the topic search query and the topic-related information and topic-related source and at least one of;

(1) a number of mentions of topic-related keywords in a headline of an item of information,

(2) a ratio of topic-related keywords to other words in the headline of the item of information,

(3) a ratio of topic-related keywords to other words besides the headline in the item of information,

(4) a number of mentions of topic-related keywords of the item of information compared to irrelevant mentions of non-topic relevant information,

(5) a number of mentions of topic-related keywords in the topic-related source compared to irrelevant mentions of non-topic relevant information, and

(6) views and curation by users of a particular intelligence channel on at least one of particular items of information and items of information sourced from at least one of particular relevant sources and authors,

wherein implementing the filtering algorithm comprises filtering the topic-related information from the at least one selected topic-related source to remove at least one of an advertisement, information in a non-designated language, and duplicative information;

determining whether to pass each item of information from the at least one selected topic-related source through the filtering algorithm based on the weighted influence and relevance ranking of each item of information from the at least one selected topic-related source and reliability of the at least one topic related source, wherein relevance of search results from the topic-related source to the selected topic is based on actions of at least one user in at least one of (a) at least one of visiting, banning, and curating the source of the filtered topic-related information and (b) at least one of rating and deleting an item of information from the topic-related source; and

assembling the items of information that pass through said filtering algorithm into a data stream and presenting the assembled data stream of topic-related information via the interactive interface as the customizable real-time intelligence channel for the selected topic.

18. The system of claim 17, wherein the relevance ranking includes at least one of crowd-sourced data including data views and curation by users of the customizable real-time intelligence channel of the selected topic-related source of at least one of particular items of information and sources of items of information from at least one of particular topic-related sources and particular authors.

19. The system of claim 17, further comprising instructions for at least one of updating and curating the plurality of topic-related keywords and information sources based on at least one of crowdsourcing data, research by subject matter experts, and input from the user.

20. The system of claim 17, further comprising instructions for determining at least one of the topic and a customizable real-time intelligence channel for a user based on predictive data mining of information related to the user.

21. The system of claim 17, wherein the interactive interface comprises:

a selectable source region providing individually selectable access to a plurality of sources of topic-related information, wherein each source of the plurality of topic-related sources comprises information related to a topic and is individually selectable;

a content region rendering topic-related information and providing access to a source of the rendered topic-related information; and

an adjustable filter region providing adjustable filtering of the content region in response to adjustment of the interactive interface element to adjust a weighting of the source of the rendered topic-related information based on at least one of the influence on and relevance of the source of the rendered topic-related information to the topic, wherein content of the content region is dynamically modified, in real time, in response to adjustment of the interactive interface element.

22. The system of claim 17, wherein the data processor further processes instructions to filter noise from the topic-related information to remove profanity and items of information that at least one of do not add value, are of limited relevance to the topic, and that lack substantive content.

23. A non-transitory computer readable storage medium comprising executable instructions that when executed by a processor cause the processor to effectuate operations to perform a method of creating a customizable real-time intelligence channel, comprising:

determining a plurality of possible sources of information on a selected topic for the real-time intelligence channel;

receiving a plurality of topic-related keywords for a search query for each source of the plurality of sources, wherein each keyword of the plurality of topic-related keywords is related to the selected topic and is used to determine whether or not to include or exclude search results including the keyword;

receiving search results from at least one search utilizing the topic-related keywords to identify topic-related information within said plurality of sources of information;

selecting from the plurality of sources, based on the searching, at least one topic-related source comprising topic-related information;

implementing a filtering algorithm to pass items of information from the at least one selected topic related source based on a weighted influence and relevance ranking of each item of information from the at least one topic related source to the selected topic, wherein;

an influence score of each item of information from the selected topic-related source is calculated based on source online presence data obtained from social networks for each source and weighted based on at least one of academic research, publications, and information available online and offline and adjusted based on crowd-sourced information, and

a relevance score of each item of information from the selected topic-related source is calculated based on a similarity between the topic search query and the topic-related information and topic-related source and at least one of;

(1) a number of mentions of topic-related keywords in a headline of an item of information,

(2) a ratio of topic-related keywords to other words in the headline of the item of information,

(3) a ratio of topic-related keywords to other words besides the headline in the item of information,

(4) a number of mentions of topic-related keywords of the item of information compared to irrelevant mentions of non-topic relevant information,

(5) a number of mentions of topic-related keywords in the topic-related source compared to irrelevant mentions of non-topic relevant information, and

(6) views and curation by users of a particular intelligence channel on at least one of particular items of information and items of information sourced from at least one of particular relevant sources and authors,

wherein implementing the filtering algorithm comprises filtering the topic-related information from the at least one selected topic-related source to remove at least one of an advertisement, information in a non-designated language, and duplicative information;

determining whether to pass each item of information from the at least one selected topic-related source through the filtering algorithm based on the weighted influence and relevance ranking of each item of information from the at least one selected topic-related source and reliability of the at least one topic related source, wherein relevance of search results from the topic-related source to the selected topic is based on actions of at least one user in at least one of (a) at least one of visiting, banning, and curating the source of the filtered topic-related information and (b) at least one of rating and deleting an item of information from the topic-related source; and

assembling the items of information that pass through said filtering algorithm into a data stream and presenting the assembled data stream of topic-related information via an interactive interface as the customizable real-time intelligence channel for the selected topic.