US 12,591,737 B2Granted

Word offensiveness detection and processing

AssigneeIpar, LLC

InventorJoseph L. Spears

Life of this patent

Estimating the term…

Loading patent…

Abstract

Computer-implemented systems and methods are provided for identifying language that would be considered obscene or otherwise offensive to a user or proprietor of a system. A plurality of offensive words are received, where each offensive word is associated with a severity score identifying the offensiveness of that word. A string of words is received. A distance between a candidate word and each offensive word in the plurality of offensive words is calculated, and a plurality of offensiveness scores for the candidate word are calculated, each offensiveness score based on the calculated distance between the candidate word and the offensive word and the severity score of the offensive word. A determination is made as to whether the candidate word is an offender word, where the candidate word is deemed to be an offender word when the highest offensiveness score in the plurality of offensiveness scores exceeds an offensiveness threshold value.

Claims42

Dependent claims

A method, using one or more processors, comprising:

accessing a first group of one or more offensive words, the first group of one or more offensive words being associated with a first type of offensive content;
accessing a second group of one or more offensive words, the second group of one or more offensive words being associated with a second type of offensive content;
receiving, from a selection made through a user interface, types of offensive content that should be filtered from a string of words within a submitted review text that is generated by the user, wherein the types of offensive content comprise the first type of offensive content and the second type of offensive content and wherein the first type of offensive content and the second type of offensive content are displayed via the user interface;
generating a consolidated list of offensive words that includes the first group of one or more of offensive words and the second group of one or more offensive words based on the selection of the types of offensive content that should be filtered;
receiving the string of words;
determining a distance of each word in the string of words to the consolidated list of offensive words, wherein the distance is determined based on any individual, character-level differences between individual characters within each word and individual characters within each offensive word in the consolidated list of offensive words;
detecting, based on the determined distances, offensive words within the string of words;
and
processing the string of words based upon the detection of offensive words in the string of words.

6 more elements

A system, comprising:

one or more data processors;
a non-transitory computer-readable medium encoded with instructions for commanding the one or more data processors to execute steps of a method that comprises:
accessing, using one or more processors, a first group of one or more offensive words, the first group of one or more offensive words being associated with a first type of offensive content;
accessing, using one or more processors, a second group of one or more offensive words, the second group of one or more offensive words being associated with a second type of offensive content;
receiving, using one or more processors, from a selection made through a user interface, types of offensive content that should be filtered from a string of words within a submitted review text that is generated by the user, wherein the types of offensive content comprise the first type of offensive content and the second type of offensive content and wherein the first type of offensive content and the second type of offensive content are displayed via the user interface;
generating a consolidated list of offensive words that includes the first group of one or more of offensive words and the second group of one or more offensive words based on the identification of the types of offensive content that should be filtered;
receiving, using one or more processors, a string of words;
determining, using one or more processors, a distance of each word in the received string of words to the consolidated list of offensive words, wherein the distance is determined based on any individual, character-level differences between individual characters within each word and individual characters within each offensive word in the consolidated list of offensive words;
and
detecting, using one or more processors and based on the determined distances, offensive words within the string of words;
and
processing the string of words, using one or more processors, based upon the detection of offensive words in the string of words.

9 more elements

A method, using one or more processors, comprising:

providing a plurality of lists, each list being associated with a type of undesirable communication, each list comprising a plurality of alphanumeric strings;
receiving, from a selection made through a user interface, assigned user preferences regarding handling of different types of undesirable communication within a string of words in a submitted review text that is generated by the user, wherein at least one user preference is selected from the group consisting of:
blocking a communication, limiting access to the communication, providing a warning, and allowing the communication and wherein at least one user preference comprises the type of undesirable communication associated with at least one list of the plurality of lists and wherein the plurality of lists are displayed via the user interface;
receiving an inbound communication that is associated with an inbound alphanumeric string;
and
comparing the inbound alphanumeric string to alphanumeric strings on the plurality of lists by:
determining, using one or more processors, a distance of the inbound alphanumeric string to the alphanumeric strings in the plurality of lists, wherein the distance is determined based on any individual, character-level differences between individual characters within each word and individual characters within each offensive word in the consolidated list of offensive words;
determining, using one or more processors and based on the determined distances, a type of undesirable communication associated with the inbound alphanumeric string;
and
handling the inbound communication according to a user preference assigned to the type of undesirable communication, wherein the inbound communication is blocked, access to the inbound communication is limited, a warning is provided, or the communication is allowed.

7 more elements

A method, comprising:

accessing, using one or more processors, a first group of one or more undesirable words, the first group of one or more undesirable words being associated with a first type of undesirable content;
accessing, using one or more processors, a second group of one or more undesirable words, the second group of one or more undesirable words being associated with a second type of undesirable content;
receiving, from a selection made through a user interface, types of undesirable content that should be filtered from a string of words within a submitted review text that is generated by the user, wherein the types of offensive content comprise the first type of offensive content and the second type of offensive content and wherein the first type of offensive content and the second type of offensive content are displayed via the user interface;
generating a consolidated list of undesirable words based on the identification of the types of undesirable content that should be filtered;
receiving, using one or more processors, content, wherein words are associated with the received content, wherein one or more detected undesirable words is selected from the words associated with the received content that matches words from the consolidated list;
determining, using one or more processors, a distance of each word associated with the received content to the consolidated list, wherein the distance is determined based on any individual, character-level differences between individual characters within each word and individual characters within each offensive word in the consolidated list of offensive words;
detecting, using one or more processors and using the determined distances, undesirable words associated with the received content;
and
processing the content, using one or more processors, based upon the detection of undesirable words.

6 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/427,883Filed 31 May 2019 · continuation
Priority: 29 Jan 2010
Published: US 2019/0286677 A119 Sep 2019
Granted: 31 Mar 2026
Term ends: Not estimatedThe source records don’t support an estimate.

Classification4 codes

G06F 40/242
Lexical tools: dictionaries
G06F 16/9535
Querying, e.g. by the use of web search engines: search customisation based on user profiles and personalisation
G06F 40/10
Handling natural language data: text processing
G06F
Computing and data processing

Family

Application12/696,991Parent of this continuation · filed 29 Jan 2010
Application13/654,722Parent of this continuation · filed 18 Oct 2012
Application15/645,063Parent of this continuation · filed 10 Jul 2017
US 2019/0286677 A1Pre-grant publication · published 19 Sep 2019
US 10,534,827 B2Parent of this continuation
US 9,703,872 B2Parent of this continuation
US 8,296,130 B2Parent of this continuation
US 2011/0191105 A1Published application
US 2013/0041655 A1Published application
US 2017/0308615 A1Published application
US 12,591,737 B2This patent

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