Granted Patent B2

US 11,182,440 · App. 15/094,775 · Granted Nov 23, 2021

Inventors: Peter Joseph Sweeney (Kitchener, CA); Robert Charles Good (Waterloo, CA)

Assignee: PRIMAL FUSION INC.

G06F16/951G06F16/24578G06F16/3334G06F16/367G06F40/30G06Q30/0277

Abstract

Claims (31)

1. A method of supplementing search queries based on identifying trendy terms within a stream of digital content received via a network interface, the method comprising:

receiving time-stamped digital content containing a plurality of terms and a time stamp associated with the plurality of terms;

calculating, with a processor, a trend score for each term in the plurality of terms contained in the time-stamped digital content, the trend score being based at least in part on a frequency of appearance of the respective term from the plurality of terms, a decay function and the time stamp;

outputting one or more trendy terms, said trendy terms being identified based on the calculated trend scores;

receiving a first search query;

extracting an active concept from said first search query;

generating a semantic network comprising a plurality of concepts related to said active concept, said plurality of concepts related to said active concept including concepts derived from said one or more trendy terms and at least one reference corpora;

categorizing each of said plurality of concepts related to said active concept into one of a plurality of categories based on whether each respective concept was derived from said one or more trendy terms or said at least one reference corpora, wherein said plurality of categories includes a) concepts identified as being very trendy terms in real-time data, b) concepts identified as being less trendy terms in said real-time data and absent from said at least one reference corpora, c) concepts identified from both said real-time data and said at least one reference corpora, and d) concepts identified from said at least one reference corpora and absent from said real-time data;

presenting said plurality of concepts related to said active concept to a user, said presenting including identifying said category given to each of said plurality of concepts;

selecting at least one of said concepts related to said active concept;

generating a second search query based at least in part on the first search query and the selected concept; and

outputting search results obtained from a search conducted based, at least in part, on the second search query.

2. The method of claim 1 , wherein:

the time-stamped content comprises content from at least one of blogs or Internet streams.

3. The method of claim 1 , wherein calculating the trend score comprises:

calculating a frequency of occurrence of a term in the time-stamped content relative to the frequency of occurrence of other terms in the time-stamped content; and

modifying the frequency of occurrence of the term based on the decay function.

4. The method of claim 1 , wherein the decay function is defined based at least in part on a rate-of-decay parameter.

5. The method of claim 1 , wherein identifying comprises:

ranking terms in the plurality of terms based on the calculated trend scores; and

identifying a number of top-ranked terms as the trendy terms.

6. The method of claim 1 , further comprising:

determining a language of the time-stamped content based on an average Unicode point value.

7. The method of claim 1 , further comprising:

adding to a semantic lexicon a term in the plurality of terms with a trend score in a predetermined range.

8. The method of claim 1 , further comprising:

selecting a term in the plurality of terms with a trend score in a predetermined range;

calculating a second trend score for the term based at least in part on the frequency of occurrence of the term in one or more user-issued queries; and

categorizing the selected term based at least in part on the determination.

9. The method of claim 8 , further comprising:

purchasing the selected term as an advertising keyword based on the categorization.

Assignments (3)

Continuity (8)