Event Classification Using N-gram Analysis and Random Forest

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems for automatically classifying web pages as related to specific events, such as sporting events or concerts, rely heavily on manual keyword selection and maintenance, which is tedious, time-consuming, and impractical for a large number of events, especially those with rapidly changing information.

Innovation Solution

The development of classification models that use n-gram analysis to automatically identify pages discussing specific events by collecting and ranking event-related n-grams from reference and publisher pages, training binary classifiers like Random Forest classifiers, and applying TF-IEF scoring to determine the relevance of n-grams to events.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual keyword selection and maintenance is used for event classification, then classification accuracy can be maintained through human expertise, but the process becomes tedious and time-consuming especially for large numbers of events

Engineering Contradiction:

Improveclassification accuracyVSAvoidtime for keyword maintenance

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system enables automatic generation and maintenance of event-related keywords through machine learning models that self-update based on learned patterns from training data, eliminating the need for manual keyword curation while maintaining classification accuracy

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system transforms the static manual keyword approach into a dynamic automated system by changing the parameter of keyword generation from human-driven to algorithm-driven, using trained classifiers that automatically adapt keywords based on event data patterns

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If manual keyword selection is used for event classification, then control over classification criteria is maintained, but the system becomes impractical for a large number of events with rapidly changing information

Engineering Contradiction:

Improvescalability to large number of eventsVSAvoidease of system implementation

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The system replaces the mechanical manual process of keyword selection with an automated computational system using machine learning classifiers that process event data and generate classification keywords automatically, enabling scalability to large numbers of events

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary training of classification models using historical event data before deployment, so that when new events are introduced, the system can quickly adapt using pre-established classification patterns rather than requiring manual keyword creation from scratch

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10504145B2Automated classification of network-accessible content based on events

Publication Date: 2019.12.10 GUMGUM INC

AI summary

Systems and methods are provided for generating and using classification models to automatically classify pages or other source content as including text about one or more real-world events. Generating the classification models may include analyzing text content of a large number of different pages from both a reference source and from more dynamic sources, such as from publisher sources via a network. Features for training classifiers may be determined based in part on the top n-grams identified among pages that have been determined to be associated with a given event.