Obscene Word Classification via Misspelled Variant Generation

Solution Overview

Problem

Existing methods for classifying obscene words struggle to effectively identify misspelled variants, which can circumvent traditional filtering systems, requiring time-consuming database construction and energy-intensive approaches.

Innovation Solution

A machine learning algorithm is trained using artificially generated misspelled variations of obscene words, based on heuristics, to assign an indecency score, enabling the classification of misspelled words as obscene, thereby improving the filtering process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If a black list database approach is used to filter obscene words, then the filtering process becomes simple to implement, but it requires time-consuming database construction and cannot effectively identify misspelled variants

Engineering Contradiction:

Improveease of implementationVSAvoidtime-consuming database construction

Core Design Contradiction:

Ease of manufactureVSLoss of time

Solution Approach 1:

The system pre-generates misspelled variants of obscene words using heuristic rules (character substitution, transposition, deletion, insertion) and stores them in the black list database in advance. This preliminary action ensures that the database is ready for immediate use without requiring time-consuming construction during deployment, while effectively capturing misspelled variants that would otherwise circumvent filtering.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system automatically generates misspelled variants using algorithmic rules rather than requiring manual curation. The heuristic-based generation process (substituting characters with similar-looking ones, transposing adjacent characters, deleting characters, inserting random characters) enables the system to self-populate the database with relevant misspelled forms, reducing both time and resource requirements for database construction.

Inventive Principle:

Principle #25Self-service

2Ease of manufacture

If a black list database approach is used to filter obscene words, then the filtering process becomes simple to implement, but it allows ill-intended users to circumvent the filter by intentionally misspelling words

Engineering Contradiction:

Improveease of implementationVSAvoidfilter effectiveness

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The system pre-generates misspelled variants of obscene words using heuristic rules (character substitution, transposition, deletion, insertion) and stores them in the black list database in advance. This preliminary action ensures that the database is ready for immediate use without requiring time-consuming construction during deployment, while effectively capturing misspelled variants that would otherwise circumvent filtering.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system takes the harmful behavior of users intentionally misspelling words to circumvent filters and converts it into a benefit by using the same heuristic rules to proactively generate and include these misspelled variants in the black list. By anticipating and pre-including these circumvention attempts, the system turns potential weaknesses into strengthened filter coverage.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

3Ease of operation

If traditional filtering methods are used, then the system is easy to operate, but it requires energy-intensive approaches and manual labeling

Engineering Contradiction:

Improveease of operationVSAvoidenergy-intensive processing

Core Design Contradiction:

Ease of operationVSUse of energy by moving object

Solution Approach 1:

The system pre-generates misspelled variants of obscene words using heuristic rules (character substitution, transposition, deletion, insertion) and stores them in the black list database in advance. This preliminary action ensures that the database is ready for immediate use without requiring time-consuming construction during deployment, while effectively capturing misspelled variants that would otherwise circumvent filtering.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system automatically generates misspelled variants using algorithmic rules rather than requiring manual curation. The heuristic-based generation process (substituting characters with similar-looking ones, transposing adjacent characters, deleting characters, inserting random characters) enables the system to self-populate the database with relevant misspelled forms, reducing both time and resource requirements for database construction.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS12026465B2Method and system for classifying word as obscene word

Publication Date: 2024.07.02 Y E HUB ARMENIA LLC

AI summary

There is disclosed a method and system for classifying a word as an obscene word, the method comprising, at a training phrase: acquiring a first word, the first word corresponding to a given obscene word; generating a first set of misspelled words, the first set of misspelled words comprising a plurality of misspelled variations of the first word; generating a training pairs, the training pairs comprising: a set of positive training pairs comprising the first word paired with each misspelled variations of the first word; training a machine learning algorithm, the training comprising: determining, for each training pairs, a set of features representative of a property of the training pairs; generating an inferred function based on the set of features, the inferred function being configured to assign, in use, an indecency score, the decency score being indicative of a likelihood of the word being obscene.