Content Selection System Using Sentiment and Confidence Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content delivery systems fail to effectively account for user interactions with other Internet content when customizing content presentation, leading to missed opportunities in capturing user attention.
Innovation Solution
A data processing system identifies topics for a requested document, determines confidence and sentiment scores, and calculates a predicted acceptance score for candidate content items based on these scores and acceptance history data, selecting content for presentation based on this score.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If content is customized based on user characteristics and interests, then user attention is captured, but user interactions with other Internet content are not accounted for
Solution Approach 1:
The system performs preliminary analysis of user interactions with other Internet content before content selection. By examining click-through history, time spent on pages, and other engagement metrics in advance, the system builds a comprehensive user profile that informs subsequent content customization decisions, ensuring both personalization and contextual awareness are integrated
2Adaptability or versatility
If multiple topics are identified for a document, then topic coverage is improved, but confidence score calculation becomes complex
Solution Approach 1:
The system segments the confidence score calculation by treating each topic independently. For each identified topic, a separate confidence score is calculated based on that topic's specific characteristics and the user's interaction history with that topic. This modular approach allows comprehensive multi-topic coverage while keeping each individual confidence calculation manageable and interpretable
3Measurement precision
If sentiment analysis is performed on multiple topics, then sentiment accuracy is improved, but processing time increases
Solution Approach 1:
The system applies partial sentiment analysis by focusing computational resources on the most relevant topics for each user query. Rather than performing exhaustive sentiment analysis on all possible topics, the system identifies and analyzes only those topics with highest relevance scores and user interaction potential, achieving sufficient sentiment accuracy while significantly reducing processing time
Data Source
AI summary
Aspects and implementations of the present disclosure are directed to systems and methods of selecting content for presentation to a viewing user. In general, in some implementations, a data processing system identifies a topic for a document requested by a user, determines a confidence score for a correlation between the topic and the document, determines a sentiment score for a sentiment towards the topic, and determines, for a candidate content item, a predicted acceptance score based on the confidence score, the sentiment score, and acceptance history data for the candidate content item. The data processing system selects a candidate content item for presentation to the user based, at least in part, on the predicted acceptance score.