Content Stream Population Using Quality Score Filtering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Social media platforms face challenges in ensuring that content streams provided to user devices include high-quality content, given the vast amounts of information shared across disparate sources.

Innovation Solution

A method that involves obtaining data related to specific topics, generating queries, evaluating search results based on quality characteristics such as precision, velocity, and feedback, and populating content streams with high-quality search results to optimize network bandwidth and reduce latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If all content items from disparate social media sources are included in content streams, then the quantity of content is increased, but the quality of content deteriorates

Engineering Contradiction:
Improvequantity of contentVSAvoidquality of content
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The system performs preliminary evaluation of content items before they are included in streams. Quality scores are calculated in advance based on multiple characteristics (engagement metrics, source reliability, recency), and only content meeting quality thresholds is selected for inclusion in streams, thus maintaining high quality while managing quantity effectively

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies different quality evaluation criteria and thresholds to different content sources and types. Each content item is assessed based on its specific characteristics and the particular stream context, allowing tailored quality control that maintains overall stream quality while incorporating diverse content

Inventive Principle:
Principle #3Local quality

2Quantity of substance

If content streams are populated with comprehensive search results, then the quantity of content is increased, but the network bandwidth consumption increases

Engineering Contradiction:
Improvequantity of contentVSAvoidnetwork bandwidth
Core Design Contradiction:
Quantity of substanceVSLoss of energy

Solution Approach 1:

The system extracts and transmits only the essential, high-quality content items that meet predetermined quality thresholds, rather than transmitting all search results. This selective extraction reduces network bandwidth consumption while ensuring that the delivered content maintains high quality standards

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Quality filtering and content selection are performed in advance before content transmission. By pre-evaluating and selecting only high-quality content items, the system minimizes the volume of data that needs to be transmitted over the network, thereby reducing bandwidth consumption

Inventive Principle:
Principle #10Preliminary action

3Quantity of substance

If content streams include all search results, then the quantity of content is increased, but the latency increases

Engineering Contradiction:
Improvequantity of contentVSAvoidlatency
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The system extracts and delivers only the most relevant and high-quality content items from search results, rather than processing and transmitting all results. This selective approach reduces the time required for content processing and delivery, thereby reducing latency while maintaining content quantity at acceptable levels

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Content quality evaluation and selection are performed in advance using pre-established quality metrics and thresholds. This preliminary filtering reduces the processing time required at content delivery time, thereby reducing latency while ensuring that the delivered content maintains high quality standards

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10409818B1Populating streams of content
Publication Date: 2019.09.10 GOOGLE LLC
  • US10409818B1 patent drawing
  • US10409818B1 patent drawing
  • US10409818B1 patent drawing

AI summary

Methods, systems, apparatus, including computer programs encoded on computer storage medium, for a bottom-up approach for generating high-quality content streams. In one aspect, the method includes actions of obtaining data identifying a plurality of content items, generating a plurality of queries for the particular topic, and for each query of the plurality of queries: obtaining a set of search results for the query that identify content items identified in the obtained data, and determining, from the search results for the query, a respective quality score for each of one or more quality characteristics. The method may also include actions such as identifying one or more first high-quality queries from the plurality of queries based on the respective quality scores for the one or more quality characteristics, and populating a stream of content for display on the user device using search results for the one or more first high-quality queries.