Content Label Grouping via Co-occurrence and Topic Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches to utilizing hashtags in social networking systems often result in repetitive or redundant labels, leading to an undesirable user experience due to the inclusion of substantially similar or related labels, which can be uninteresting and inefficient for users.
Innovation Solution
A system and method to identify and group related content labels by using co-occurrence and topic similarity metrics, selecting one representative label based on social engagement, and providing suggestions to reduce redundancy, such as replacing multiple related hashtags with a single representative hashtag.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional approaches are used to utilize hashtags, then users can tag content items with various labels, but the hashtags become repetitive and redundant, reducing user experience
Solution Approach 1:
The patent merges similar or related hashtags into unified categories. The system identifies hashtags that are substantially similar or related and groups them together, presenting a consolidated view to users. This reduces redundancy by combining multiple repetitive labels into a single representative category while preserving the diversity of tagging options.
Solution Approach 2:
The patent changes the parameter of label representation from individual hashtags to grouped categories. By transforming the display format from multiple separate hashtags to consolidated groups with representative labels, the system reduces redundancy while maintaining adaptability. The grouping parameter organizes similar labels together, allowing users to select from diverse categories without encountering repetitive individual tags.
2Ease of operation
If multiple similar hashtags are provided to users, then labeling flexibility is maintained, but user experience deteriorates due to repetitiveness
Solution Approach 1:
The system merges multiple similar hashtags into unified groups, presenting users with consolidated categories rather than individual repetitive tags. This improves ease of operation by reducing the number of similar options users must evaluate, while maintaining labeling flexibility through representative labels that capture the essence of each group.
Solution Approach 2:
The patent uses representative hashtags that copy or represent the meaning of multiple similar tags within a group. Instead of displaying every individual hashtag, the system selects representative labels that capture the common theme, allowing users to efficiently select appropriate labels without being overwhelmed by repetitious options.
3Loss of information
If trending labels are displayed as individual items, then all labels are visible, but the list becomes uninteresting due to inclusion of similar labels
Solution Approach 1:
The patent merges similar trending labels into consolidated groups, ensuring that all label information remains visible through representative categories. This approach maintains complete label visibility while improving user interest by presenting organized, non-repetitive groups rather than a flat list of similar individual tags.
Solution Approach 2:
The system segments the trending labels list into distinct groups based on similarity relationships. By dividing the comprehensive label set into meaningful segments or categories, the patent maintains full label visibility while organizing information in an interesting, structured format that reduces repetitiveness and enhances user engagement.
Data Source
AI summary
Systems, methods, and non-transitory computer-readable media can acquire a set of labels associated with a set of content items. Each label in the set of labels can be associated with at least one content item in the set of content items. It can be determined that at least two labels, out of the set of labels, are related. The at least two labels can be determined to be related based on at least one of a co-occurrence metric associated with the at least two labels or a topic similarity metric associated with the at least two labels. One label can be selected, out of the at least two labels, as being representative of the at least two labels.