Predictive Input Engine Candidate Ranking with Semantic Variety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing predictive input engines often fail to provide a satisfactory user experience by presenting lists of candidates that are either too focused on frequency of use or frequency and recency, leading to distracting and unproductive user interactions, as they may not accurately reflect user intent and can include undesired similar terms.
Innovation Solution
A predictive input engine that combines frequency and recency information to generate candidate lists, while incorporating a variety engine to ensure dissimilar options are presented at the top, thereby improving user experience by balancing relevance and novelty.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If predictive input engines present candidates sorted based on frequency of use, then the system can provide predictions, but the most frequent terms are not a good prediction of user intent and can be distracting
Solution Approach 1:
The patent changes the ranking parameters from frequency-based to a hybrid model incorporating recency, frequency, and semantic variety. This allows the system to balance statistical relevance with user intent prediction, reducing distractions while maintaining productivity.
Solution Approach 2:
The patent creates a composite ranking approach by combining multiple factors (recency, frequency, semantic similarity) into a unified candidate selection mechanism. This composite model better reflects user intent than any single factor alone.
2Adaptability or versatility
If predictive input engines present multiple phrases that begin with the input string, then more options are provided, but multiple similar yet undesired terms are presented causing distractions
Solution Approach 1:
The patent applies local quality by ensuring semantic diversity specifically at the top positions of the candidate list. The variety engine modifies local rankings to prevent similar terms from clustering at the top, while maintaining overall candidate diversity throughout the list.
Solution Approach 2:
Instead of ranking purely by frequency or recency, the patent inverts the approach by actively penalizing semantic similarity at top positions. This inversion prioritizes variety over raw frequency, reducing distractions while maintaining versatility.
3Reliability
If the candidate list is sorted by recency, then recent user preferences are captured, but the list lacks variety and presents similar terms
Solution Approach 1:
The patent merges recency-based ranking with semantic variety constraints. The variety engine combines these two objectives into a unified ranking system that maintains temporal relevance while ensuring diverse candidate presentation.
Solution Approach 2:
The patent introduces dynamic adjustment to the ranking process. The variety engine dynamically modifies candidate positions based on semantic similarity detection, making the ranking adaptive rather than static. This allows the system to maintain reliability while improving versatility.
Data Source
AI summary
A computing device maintains an input history in memory. This input history includes input strings that have been previously entered into the computing device. When the user begins entering characters of an input string, a predictive input engine is activated. The predictive input engine receives the input string and the input history to generate a candidate list of predictive inputs which are presented to the user. The user can select one of the inputs from the list, or otherwise continue entering characters. The computing device generates the candidate list by combining frequency and recency information of the matching strings from the input history. Additionally, the candidate list can be manipulated to present a variety of candidates. By using a combination of frequency, recency and variety, a favorable user experience is provided.