In-Memory Index for Real-Time Search Ranking

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Search engines often prioritize results based on historical user traffic, leading to outdated information being prominently displayed, making it difficult for users to find recent trends or topics of interest.

Innovation Solution

Implementing an in-memory index that rapidly and periodically updates metadata by recording recent significant user behavior, allowing for real-time ranking of search query results to reflect emerging trends and periodic events.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If search results are ranked based on historical user traffic, then search results receive stable and consistent ranking, but outdated information is prominently displayed and recent trends are hidden

Engineering Contradiction:

Improveranking consistencyVSAvoidrecent trend visibility

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent segments the search index into two distinct components: a main inverted index for stable, historical ranking and an in-memory index for recent, significant user behavior. This segmentation allows each index to serve its specific purpose without interfering with the other, resolving the contradiction between ranking consistency and recent trend visibility.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a dynamic in-memory index that is continuously updated with recent user behavior data, contrasting with the static main inverted index. This dynamic component enables the system to adapt to emerging trends while maintaining the stability of the historical ranking system.

Inventive Principle:

Principle #15Dynamics

2Quantity of substance

If search results are ranked based on historical user traffic, then comprehensive search coverage is maintained, but users must scroll through outdated results to find recent information

Engineering Contradiction:

Improvesearch result coverageVSAvoidtime to find recent information

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-computing and storing significant user behavior data in the in-memory index before it is needed for ranking. This allows the ranking system to quickly access recent trend information without having to process entire search results during query time, reducing user search time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The in-memory index acts as an intermediary layer between the main inverted index and the final search results. It mediates by providing recent behavior signals that adjust the ranking of results from the main index, allowing users to find recent information quickly while maintaining comprehensive coverage.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If an in-memory index is continuously updated with recent user behavior, then recent trends are prominently displayed, but system complexity increases

Engineering Contradiction:

Improveresponsiveness to recent trendsVSAvoidindex system complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies local quality by making the in-memory index selective rather than comprehensive. It only stores and processes 'significant' user behavior data that meets specific criteria, rather than tracking all user interactions. This reduces the complexity of the in-memory index while maintaining its ability to capture important recent trends.

Inventive Principle:

Principle #3Local quality

4Speed

If search ranking prioritizes recent user behavior, then up-to-date information is provided, but historical popularity signals are diminished

Engineering Contradiction:

Improveinformation freshnessVSAvoidhistorical popularity accuracy

Core Design Contradiction:

SpeedVSReliability

Solution Approach 1:

The patent merges signals from both the main inverted index (historical popularity) and the in-memory index (recent behavior) to produce the final search results. This combination allows the system to maintain the reliability of historical popularity signals while incorporating the freshness of recent user behavior, achieving a balanced ranking that values both timeframes.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS8244701B2Using behavior data to quickly improve search ranking

Publication Date: 2012.08.14 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Systems and methods for applying user behavior data to improve search query result ranking are provided. Upon receiving an update file indicating that recent, significant user behavior data is available for a document associated with an inverted index, the update file is published periodically and frequently to an index server. After filtering out the relevant update information from the update file, the index server extracts identifiers of the documents having the associated user behavior data. The update file and the identifier of the documents are utilized to update an in-memory index containing representations of metadata indicative of the user behavior. The in-memory index is continuously updated and utilized to serve search query results in response to user search queries. Search query results from the in-memory index are ranked using the user behavior data prior to serving. Thus, results associated with recent, significant user-behavior metadata receive prominent placement on the search results page.