Event-Driven Data Aggregation for Near Real-Time Web Indexing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current search engines are inadequate for monitoring and indexing dynamic content on the Web in real time, as they typically take weeks to identify and catalog changes, and are not effective for finding content created within specific time ranges.

Innovation Solution

An event-driven method of data aggregation and indexing that utilizes a service-oriented architecture to receive event notifications from content sites, generate metadata with timestamps, and index changes within seven days, allowing for near real-time retrieval of content changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional search engines periodically crawl the Web to construct databases, then comprehensive content coverage is achieved, but the time required to identify and catalog changes increases to weeks or months

Engineering Contradiction:
Improvecontent freshnessVSAvoidindexing delay
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements preliminary action by using event notification mechanisms that trigger immediate indexing when content changes occur. Content sites send notifications before or at the moment of content publication, allowing the system to proactively retrieve and index changes without waiting for periodic crawls, thereby eliminating indexing delays

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system employs feedback mechanisms where content sites provide notifications about content changes back to the search engine. This feedback loop enables the system to respond immediately to content updates, continuously maintaining an up-to-date index without requiring time-consuming periodic rescans of the entire Web

Inventive Principle:
Principle #23Feedback

2Quantity of substance

If traditional search engines use comprehensive Web crawling, then all available content is indexed, but the system can only identify less than 10% of the content and misses time-sensitive information

Engineering Contradiction:
Improvecontent coverageVSAvoidindexing efficiency
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent extracts only the essential notification signals from content sites rather than processing entire web pages. By taking out just the critical change information through event notifications, the system achieves comprehensive content coverage while maintaining high indexing efficiency, as it processes minimal data packets instead of complete pages

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system introduces an intermediary notification mechanism between content sites and the search engine. This intermediary layer (event notification system) efficiently transmits content change information without requiring the search engine to directly crawl and process all content, thereby improving productivity while maintaining comprehensive coverage

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of operation

If search engines create reverse indexes using key word vocabulary, then search functionality is provided, but the system cannot effectively find content within specific time ranges

Engineering Contradiction:
Improvesearch capabilityVSAvoidtime-based querying
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent adds a temporal dimension to the traditional keyword-based reverse index. By incorporating timestamps and time-range metadata into the indexing structure, the system enables queries to filter results by specific time periods, transforming a two-dimensional keyword search into a multi-dimensional search space that includes time as a critical dimension

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS7596571B2Ecosystem method of aggregation and search and related techniques
Publication Date: 2009.09.29 SYNACOR INC
  • US7596571B2 patent drawing
  • US7596571B2 patent drawing
  • US7596571B2 patent drawing

AI summary

Techniques are described for effecting data aggregation in a network. Event notifications are received from a plurality of content sites on the network. Each event notification indicates occurrence of an event associated with a corresponding one of the content sites. Each event corresponds to a change to content associated with the corresponding content site. In response to each event notification, event metadata are generated. The event metadata include a time stamp for the event and change data corresponding to the change to the content. The event metadata are indexed in a plurality of databases with reference to a plurality of index metrics including the time stamp. The indexing of the event metadata for each event occurs within days, hours, or even minutes of receipt of the corresponding event notification thereby enabling retrieving of the changes to the content from the plurality of databases using any of the index metrics in near real time.