Faceted Time-Series Search for Real-Time Ad-Hoc Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional business intelligence solutions face challenges in generating filtered reports on large datasets due to high computational costs and limitations on filterable fields, making real-time or near-real-time ad-hoc filtering difficult and time-consuming.
Innovation Solution
Implementing faceted search techniques based on document creation time, using index information to map documents to time ranges, allowing for ad-hoc filtering and real-time production of time-series data through search faceting and graphing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If traditional business intelligence solutions use aggregation and pre-computation techniques to improve performance, then query speed is improved, but the number of filterable fields is restricted to pre-defined fields only
Solution Approach 1:
The patent pre-computes and stores posting lists organized by time ranges (e.g., hourly, daily, weekly) during data indexing. This preliminary organization enables fast faceted search without restricting filterable fields, as the posting lists are structured to support multiple ad-hoc filtering scenarios while maintaining query performance.
2Adaptability or versatility
If traditional business intelligence solutions allocate a database column to each filterable field to enable filtering, then filtering capability is provided, but computing resources are excessively consumed on large datasets
Solution Approach 1:
The patent segments the data index into multiple posting lists, each corresponding to a specific time range (e.g., documents created in the last hour, day, week). This segmentation allows the system to process only relevant subsets of data during filtering operations, dramatically reducing computing resource consumption while maintaining versatile filtering capability across large datasets.
3Adaptability or versatility
If traditional business intelligence solutions perform ad-hoc filtering on large datasets, then flexible filtering is achieved, but the operation is too costly to perform in real-time or near-real-time
Solution Approach 1:
The patent pre-organizes data into posting lists segmented by time ranges during the indexing phase. When ad-hoc filtering is requested, the system leverages these pre-organized structures to rapidly retrieve and filter documents within specified time ranges, enabling flexible filtering operations to execute in real-time or near-real-time without excessive computational overhead.
Data Source
AI summary
Techniques and tools are described for producing time-series data using faceted search based on document creation time. For example, index information can be created from events (e.g., by adding meta-data and indexing the events as documents). In addition, index information can be created that maps document creation time to time ranges. Search queries can then be executed (e.g., comprising ad-hoc filters to filter on the meta-data), and search results can be faceted on the time ranges to produce time-series data. The time-series data can be graphed to display trends of activity (e.g., trends of events based on user activity).