US 11,080,485 B2Granted

Systems and methods for generating and recognizing jokes

AssigneeTwenty Lane Media, LLC

InventorJoseph E. Toplyn

Life of this patent

Estimating the term…

Loading patent…

Abstract

Joke recognition methods include using server(s) coupled with data store(s) to communicatively couple with a first computing device through a telecommunications network. A first communication is provided to a user through a user interface of the first computing device or is received through the user interface. A second communication is provided to the user through the user interface or is received through the user interface. In response to providing or receiving the second communication, the server(s) determine whether the second communication, relative to the first communication, includes a joke and/or a punchline. Upon determining that the second communication includes a joke and/or a punchline, the server(s) initiate sending one or more responses to the first computing device. The response(s) initiate providing, through the user interface, an indication to the user that the second communication is recognized as a joke/punchline. Systems for joke recognition provide the disclosed joke recognition methods.

Claims20

Dependent claims

A system for recognizing jokes, comprising:

one or more data stores;
and
one or more servers communicatively coupled with the one or more data stores;
wherein the one or more servers are configured to:
communicatively couple with a first computing device through a telecommunications network;
receive a first communication through a user interface of the first computing device;
determine, in response to receiving the first communication, whether the first communication comprises one of a joke and a punchline;
and
initiate sending of one or more responses to the first computing device through the telecommunications network;
wherein the one or more responses are configured to initiate providing, through the user interface, an indication to the user that the first communication is recognized as one of a joke and a punchline;
and
wherein determining that the first communication comprises one of a joke and a punchline comprises one of:
determining that a wordplay score beyond a preset threshold exists between adjacent words or word chunks of the first communication, wherein the wordplay score is based on one or more of an edit distance score, a stop consonant score, an alliteration score, a score measuring a degree to which the adjacent words or word chunks comprise an identical number of syllables, a score measuring a degree to which the adjacent words or word chunks end with a common phoneme or letter, a score measuring a degree to which the adjacent words or word chunks share common vowel sounds, and a degree to which the adjacent words or word chunks rhyme;
determining that the first communication comprises a portmanteau;
determining that the first communication includes one or more words or word chunks that have a probability of occurrence below a predetermined threshold and that also have a calculated relatedness score, relative to one or more words or word chunks of an initial communication preceding the first communication, beyond a predetermined threshold;
and calculating a probability of one or more words, word chunks or concepts in the first communication being consistent with usage patterns of less than half of an intended audience.

13 more elements

A method for recognizing jokes, comprising:

providing one or more data stores;
providing one or more servers communicatively coupled with the one or more data stores;
and
using the one or more servers:
communicatively coupling with a first computing device through a telecommunications network;
receiving a first communication provided through a user interface of the first computing device;
determining, in response to receiving the first communication, whether the first communication comprises one of a joke and a punchline;
calculating a funniness score using the one or more servers;
and
initiating sending of one or more responses to the first computing device through the telecommunications network;
wherein the one or more responses are configured to initiate providing, through the user interface, an indication to the user that the first communication is recognized as one of a joke and a punchline;
and
wherein the funniness score is based at least in part on one or more of:
a length of one of the first communication and an initial communication preceding the first communication;
how close one or more words are to an end of the first communication;
a calculation of how closely related one or more words or word chunks of the first communication are semantically to one or more words or word chunks of the initial communication;
a calculated wordplay score between one or more words or word chunks of the first communication or the initial communication and one or more other words or word chunks of one of the first communication and the initial communication, wherein the calculated wordplay score is based on one or more of an edit distance score, a stop consonant score, an alliteration score, a score measuring a degree to which the adjacent words or word chunks comprise an identical number of syllables, a score measuring a degree to which the adjacent words or word chunks end with a common phoneme or letter, a score measuring a degree to which the adjacent words or word chunks share common vowel sounds, and a degree to which the adjacent words or word chunks rhyme;
a calculation of wordplay between one or more words or word chunks of the first communication or the initial communication and one or more words or word chunks not in the first communication and not in the initial communication but having a relatedness score, relative to one or more other words or word chunks of the first communication or initial communication, beyond a predetermined threshold;
a calculation related to a degree to which the first communication and the initial communication share words and grammatical structure;
and a calculation of how closely related one or more words or word chunks of the initial communication are semantically to one or more other words or word chunks of the initial communication.

17 more elements

A method for recognizing jokes, comprising:

using one or more servers, the one or more servers communicatively coupled with one or more data stores:
communicatively coupling with a first computing device through a telecommunications network;
one of providing a first communication to a user through a user interface of the first computing device and receiving the first communication through the user interface of the first computing device;
one of providing a second communication to the user through the user interface and receiving the second communication through the user interface;
determining, in response to providing or receiving the second communication, whether the second communication, in relation to the first communication, comprises one of a joke and a punchline;
and
initiating sending of one or more responses to the first computing device through the telecommunications network;
wherein the one or more responses are configured to initiate providing, through the user interface, an indication to the user that the second communication is recognized as one of a joke and a punchline;
and
wherein determining that the second communication comprises one of a joke and a punchline comprises one of:
determining that a wordplay score beyond a preset threshold exists between adjacent words or word chunks of the second communication, wherein the wordplay score is based on one or more of an edit distance score, a stop consonant score, an alliteration score, a score measuring a degree to which the adjacent words or word chunks comprise an identical number of syllables, a score measuring a degree to which the adjacent words or word chunks end with a common phoneme or letter, a score measuring a degree to which the adjacent words or word chunks share common vowel sounds, and a degree to which the adjacent words or word chunks rhyme;
determining that the second communication comprises a portmanteau;
determining that the second communication includes one or more words or word chunks that have a probability of occurrence below a predetermined threshold and that also have a calculated relatedness score, relative to one or more words or word chunks of the first communication, beyond a predetermined threshold;
and calculating a probability of one or more words, word chunks or concepts in the second communication being consistent with usage patterns of less than half of an intended audience.

11 more elements

Dependencies are inferred from the claim text.

Facts

Application: 17/247,482Filed 13 Dec 2020 · continuation in part
Priority: 24 Feb 2018
Published: US 2021/0103700 A18 Apr 2021
Granted: 3 Aug 2021
Term ends: Not estimatedThe source records don’t support an estimate.

Classification18 codes

G06F 40/289
Recognition of textual entities: phrasal analysis, e.g. finite state techniques or chunking
G06F 40/30
Handling natural language data: semantic analysis
G06F 40/10
Handling natural language data: text processing
G06F 40/211
Parsing: syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars

Family

Application62/634,845Us provisional application · filed 24 Feb 2018
Application16/250,972Parent of this continuation-in-part · filed 17 Jan 2019
Application16/831,526Parent of this continuation-in-part · filed 26 Mar 2020
US 10,642,939 B2Parent of this continuation-in-part · filed 5 May 2020
US 2021/0103700 A1Pre-grant publication · published 8 Apr 2021
US 10,878,817 B2Parent of this continuation-in-part
US 2019/0266250 A1Published application
US 2020/0227032 A1Published application
US 11,080,485 B2This patent

From the US filing history. Foreign family members aren’t listed.