Exponential Decay Scoring for Dynamic Media Ranking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Video hosting servers face challenges in efficiently ranking and maintaining a dynamic set of popular media items, as existing methods often result in static popularity lists dominated by extremely popular content, failing to incorporate new items and user engagement effectively.

Innovation Solution

A system computes scores for media items by determining positive user actions, combining score contributions, and applying exponential decay, while filtering out items not featured externally or not meeting certain criteria, to create a diverse and current popular media items data set.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If existing popularity ranking methods are used, then the popular media items data set is easy to generate, but the data set becomes static and dominated by extremely popular content, failing to incorporate new items effectively

Engineering Contradiction:
Improveease of generating popular media items data setVSAvoidability to incorporate new items and reflect current popularity
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent applies dynamics by implementing an exponential decay function that continuously reduces the score of media items over time. This dynamic mechanism ensures that the popular media items data set automatically updates to reflect current popularity trends, allowing new items to rise to prominence while preventing the data set from becoming static and dominated by historically popular content.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of score calculation by incorporating multiple factors including view counts, user engagement metrics, and recency weights through exponential decay. This multi-parameter approach enables the system to effectively evaluate and rank both new and established media items, improving adaptability while maintaining generation efficiency.

Inventive Principle:
Principle #35Parameter changes

2Device complexity

If simple view count ranking is used, then the ranking process is simple and fast, but it fails to capture user engagement and diversity of content

Engineering Contradiction:
Improvecomplexity of ranking processVSAvoidaccuracy of popularity measurement
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent segments the popularity measurement into multiple distinct components: view counts, user engagement metrics (likes, comments, shares), and recency factors. Each segment is calculated and weighted separately, then combined to produce a comprehensive popularity score. This segmentation enables precise measurement of popularity while maintaining a structured and manageable ranking process.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a universal scoring system that handles multiple types of media items (videos, images, articles) and multiple engagement metrics through a single exponential decay framework. This multi-functional approach accurately captures diverse content popularity without requiring separate complex ranking processes for each content type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Quantity of substance

If no filtering is applied, then all media items are included in ranking, but the data set quality decreases with low engagement items

Engineering Contradiction:
Improvenumber of media items in data setVSAvoidquality and engagement level of media items
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent implements feedback mechanisms by setting minimum threshold values for view counts and user engagement metrics. Media items that fail to meet these thresholds are filtered out of the popular media items data set. This feedback-based filtering ensures that only quality content with sufficient user engagement is included, maintaining data set reliability while preserving a meaningful quantity of items.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS9092489B1Popular media items data set with exponential decay
Publication Date: 2015.07.28 GOOGLE LLC
  • US9092489B1 patent drawing
  • US9092489B1 patent drawing
  • US9092489B1 patent drawing

AI summary

A processing device computes scores for a plurality of media items. A score for a media item is computed by determining a plurality of positive user actions associated with the media item, combining a plurality of score contributions from the plurality of positive user actions to determine a value for the score, and applying an exponential decay to the value for the score. The media items are ranked based on the scores. One or more of the media items that have not been featured on any external website or that fail to satisfy another filtering criterion are filtered. A popular media item data set is then populated with highest ranked media items from a remainder of the plurality of media items.