Digital Content Alerting via Statistical Distribution Thresholds

Solution Overview

Problem

Existing digital content object tracking systems struggle to accurately identify and alert users to relevant news stories, often leading to over-frequent or irrelevant notifications due to the difficulty in determining what constitutes a 'big' story, which can result in user rejection of alerting sources.

Innovation Solution

A computer system and method that utilizes a user interface and alerting module to configure digital content object panel specifications, applying them to a database to determine alert criteria, fit values to a distribution function, set a threshold for alerting, and adjust based on user preferences, allowing for targeted and context-aware alerts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If alerting is applied to all digital content objects matching user interests, then user coverage is improved, but alert frequency becomes too high causing alarm fatigue

Engineering Contradiction:

Improveuser coverageVSAvoidalarm fatigue

Core Design Contradiction:

Adaptability or versatilityVSObject-generated harmful factors

Solution Approach 1:

The patent applies parameter changes by using statistical distribution parameters (mean, standard deviation) to dynamically adjust alert thresholds. Instead of fixed thresholds, the system calculates thresholds based on the statistical properties of engagement metrics for similar content, allowing adaptive filtering that maintains user coverage while reducing excessive alerts.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces mechanical filtering systems with statistical modeling. Instead of using rigid rule-based filters, the system employs distribution functions and statistical parameters to model user engagement patterns, enabling more nuanced and accurate alert filtering that adapts to varying content types and user behaviors.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Quantity of substance

If alerting is applied to all digital content objects matching user interests, then comprehensiveness is improved, but relevance deteriorates due to inclusion of low-interest stories

Engineering Contradiction:

ImprovecomprehensivenessVSAvoidrelevance

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent replaces mechanical filtering systems with statistical modeling. Instead of using rigid rule-based filters, the system employs distribution functions and statistical parameters to model user engagement patterns, enabling more nuanced and accurate alert filtering that adapts to varying content types and user behaviors.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent applies parameter changes by using statistical distribution parameters (mean, standard deviation) to dynamically adjust alert thresholds. Instead of fixed thresholds, the system calculates thresholds based on the statistical properties of engagement metrics for similar content, allowing adaptive filtering that maintains user coverage while reducing excessive alerts.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If simple alerting rules are used, then system complexity is reduced, but accuracy of identifying big stories deteriorates

Engineering Contradiction:

Improvesystem complexityVSAvoidaccuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces mechanical filtering systems with statistical modeling. Instead of using rigid rule-based filters, the system employs distribution functions and statistical parameters to model user engagement patterns, enabling more nuanced and accurate alert filtering that adapts to varying content types and user behaviors.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10621680B2System and method for alerting users to digital content objects of potential interest

Publication Date: 2020.04.14 NEWSWHIP MEDIA

AI summary

A method and system for processing digital content objects, such as news stories, are provided. A user specifies digital content objects of interest. The user specification is then applied to a source of digital content objects in order to obtain a batch of digital content objects matching the specification. A value of a variable, such as a social media impact metric, is then determined for each of the digital content objects of the batch and these values are fitted to a distribution function in order to determine parameter values for the distribution function. A threshold value for alerting is then determined based on the parameterized distribution function. The specification can then continue to be applied to the source of digital content objects and when new digital content objects are found that match the specification, their values are compared against the threshold value for alerting and the user is alerted only in respect of new digital content objects that have values which exceed the threshold value.