Real-Time Content Recommendation System Using Segmented User and Title Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content recommendation systems on computing devices, such as gaming consoles, rely solely on popularity trends, leading to users being shown the same content, neglecting individual user interests and being limited by interface space, resulting in decreased content quality and increased resource usage.
Innovation Solution
Generating two sets of affinity scores - user scores and title scores - based on user event records, which are combined in real-time to provide personalized content recommendations, taking into account individual user behavior and broader population trends, thus optimizing content relevance and reducing computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If content is suggested based on popularity trends, then content availability is improved, but content relevance to individual users deteriorates
Solution Approach 1:
The patent segments content recommendations into two distinct score sets: user scores based on individual user behavior patterns and title scores based on popularity trends. This segmentation allows the system to separately optimize for both content relevance to individual users and content availability, resolving the contradiction by processing and presenting these two dimensions of recommendation independently yet integrally.
Solution Approach 2:
The patent applies local quality by personalizing content recommendations to each user based on their specific behavior patterns, interaction history, and preferences. Instead of uniform popularity-based recommendations, the system tailors content suggestions to individual users' local characteristics while still incorporating broader popularity trends, thereby improving content relevance without sacrificing availability.
2Adaptability or versatility
If multiple content options are displayed, then content diversity is improved, but interface space requirements worsen
Solution Approach 1:
The patent applies partial action by selecting and displaying only the most relevant content options based on the combined user and title scores, rather than displaying all available content. This allows the system to provide diverse content recommendations that are personally tailored to each user, maximizing content diversity within the limited interface space available on gaming consoles.
3Measurement precision
If real-time content recommendations are generated, then content relevance is improved, but computational resources worsen
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing user scores and title scores in separate data structures before actual recommendation generation. During real-time recommendation requests, the system only needs to combine these pre-computed scores rather than performing complex analysis from scratch, significantly reducing computational resources required for real-time recommendations while maintaining high relevance.
Solution Approach 2:
The patent segments the computational workload into two independent score generation processes (user scores and title scores) that can be performed separately and cached, rather than computing a single comprehensive recommendation score each time. This segmentation allows the system to maintain real-time relevance by combining pre-computed results, thereby reducing overall computational resource consumption.
Data Source
AI summary
Methods, systems, apparatuses, and computer program products are described for providing content recommendations to a requestor. In implementations, user event records associated with media items and content types may be stored, for instance, in an event catalog. A set of user scores may be generated based on a user-content score model and the user event records, where each generated user score represents an affinity between a user and a corresponding content type of the plurality of content types. A set of title scores may also be generated for a media item based on a title-content score model, where each generated title score represents an affinity between the media item and a corresponding content type. In response to a request, the sets of scores may be combined to generate a set of content recommendations to provide to the requestor.