Cross-domain Topic Space for Video Recommendation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Social media platforms struggle to effectively connect disparate sources of multimedia data, limiting the ability to recommend relevant videos based on topical similarity between old and newly uploaded content, and failing to accurately measure video popularity using traditional metrics alone.

Innovation Solution

A common topic space is created using a bidirectional bridge between social stream and multimedia domains, employing an Online Streaming LDA model to update topics in real-time, allowing for video recommendations and popularity ranking based on social trends.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional video recommendation systems are used, then video recommendations can be provided, but the ability to connect old and new videos based on topical similarity is limited

Engineering Contradiction:

Improveability to connect old and new videos based on topical similarityVSAvoidtopical relevance between social streams and media items

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent introduces a topic space as an intermediary layer between social streams and media items. This topic space contains topics that serve as mediators to connect microblog entries from social streams with relevant media items across different domains, enabling the system to bridge the gap between disparate data sources and improve topical relevance matching

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The topic space is designed as a universal structure that can handle multiple types of data including microblog entries, video content, and other media items. This multi-functional topic space enables the system to process and connect diverse content types from different domains through common topical representations

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Measurement precision

If traditional popularity metrics are used, then video popularity can be measured, but the accuracy of popularity measurement is limited

Engineering Contradiction:

Improveaccuracy of video popularity measurementVSAvoidsocial trend information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The system incorporates feedback from social streams into the popularity measurement process. By monitoring social media discussions and trends related to video topics, the system receives continuous feedback that supplements traditional metrics like view counts, providing a more comprehensive and accurate picture of video popularity that reflects current social sentiment and trending topics

Inventive Principle:

Principle #23Feedback

3Speed

If real-time social data is utilized, then relevant information can be obtained quickly, but the complexity of processing and integrating data from multiple domains increases

Engineering Contradiction:

Improvespeed of obtaining relevant informationVSAvoidcomplexity of processing and integrating data from multiple domains

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The patent segments the complex multi-domain data processing task into distinct components: social stream processing, topic extraction, media domain processing, and recommendation generation. Each component operates independently on its specific data type, extracting topics or features separately before integrating results through the topic space, thereby reducing overall system complexity while maintaining real-time processing capability

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS9152709B2Cross-domain topic space

Publication Date: 2015.10.06 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Some examples include receiving a microblog entry from a social stream domain. Further, some implementations include determining, based on a topic space associated with the social stream domain and a media domain, a topic that is associated with the microblog entry. Some implementations include determining, based on the topic space, one or more media items that are associated with the topic.