Search Engine Numeric Relevance Decay Computation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines face challenges in efficiently updating numeric relevance scoring modifiers over time, leading to increased memory requirements and computational overhead, as they rely on controlling applications to manage and update these values, which can result in performance issues and resource bottlenecks.
Innovation Solution
The search engine is configured to manage and update numeric relevance scoring modifiers independently, allowing controlling applications to set limits and decay rates, while the search engine applies adjustments regularly or on demand, ensuring valid values within acceptable ranges without the need for frequent transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If controlling applications manage and update numeric relevance scoring modifiers, then relevance computation can be performed, but computational overhead and resource usage increase
Solution Approach 1:
The patent extracts the modifier management functionality from controlling applications and places it directly into the search engine. The search engine now independently manages numeric relevance scoring modifiers, applying decay computations and validity checks without requiring controlling applications to track or update these values. This extraction eliminates the computational overhead associated with modifier management in controlling applications while maintaining relevance computation accuracy.
2Measurement precision
If modifier values are tracked and updated frequently by controlling applications, then relevance scores remain current, but transaction volume and performance impact increase
Solution Approach 1:
The search engine performs self-service by independently computing and applying modifier decay values. Instead of relying on controlling applications to manage modifiers, the search engine automatically tracks time elapsed since last modification, computes appropriate decay amounts, and applies these adjustments directly to modifier values. This self-service approach maintains current relevance scores while eliminating the performance impact of frequent transactions from controlling applications.
3Adaptability or versatility
If decay computation is performed by controlling applications, then modifier values can be adjusted, but the burden of computing falls on external applications
Solution Approach 1:
The patent inverts the traditional approach by having the search engine perform decay computations rather than controlling applications. The search engine now calculates time-based decay values and applies them to modifiers independently. Controlling applications simply set initial modifier values and decay rates, while the search engine handles all subsequent decay computations and adjustments, significantly reducing the operational burden on external applications.
Data Source
AI summary
Search engines today are capable of incorporating numeric scoring modifiers from controlling applications into their relevance computations. Challenges arise in keeping these modifiers current, given that they may change over time. Embodiments provide a new way to compute numeric value decay for efficient relevance computation without having to rely on a controlling application. The controlling application can set a value for a modifier of an object managed by the controlling application and the controlling application can perform operations on the modifier. However, the controlling application does not need to keep track of the modifier and compute the modifier value independently. Rather, a search engine is configured to perform decay computation(s) and adjust the modifier value on a regular basis or on demand. The search engine ensures that modifier values for all the objects indexed by the search engine are always valid—within acceptable ranges and with acceptable adjustments.