Streaming Event Analysis for Search Recall
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing e-commerce user interfaces face inaccuracies and inefficiencies in search result generation due to outdated or incomplete rules, leading to irrelevant information and increased resource usage, as they rely on manual analysis and human-intervention methods that fail to capture consumer intent effectively.
Innovation Solution
The implementation of data processing techniques that analyze historical and current streaming interaction events to dynamically adjust search logic and rules, improving search recall by identifying scenarios where relevant items were not initially found, and updating search criteria based on consumer engagement, exposure, and relevancy scores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis and human-intervention methods are used to generate search rules, then search result generation can be performed, but the accuracy and completeness of search results deteriorate due to outdated or incomplete rules
Solution Approach 1:
The system enables search rules to automatically update themselves by analyzing streaming interaction events and consumer behavior patterns. The search rule generation process serves itself by using actual user interaction data to continuously refine and improve search rules without requiring manual intervention, thereby resolving the contradiction between accuracy and automation.
Solution Approach 2:
The system implements a feedback mechanism where search results and consumer interactions are continuously monitored and fed back into the search rule generation process. This closed-loop feedback system allows the automated search rule generation to improve accuracy over time by learning from actual consumer behavior, resolving the contradiction between automation and result quality.
2Reliability
If manual analysis methods are used to update search rules, then search rules can be maintained, but resource consumption increases due to additional computer and network resources required
Solution Approach 1:
The system uses self-service automation where search rules automatically update themselves by processing streaming interaction events. This eliminates the need for manual analysis resources while maintaining rule completeness, as the automated system efficiently processes and learns from user interaction data to continuously refine search rules.
Solution Approach 2:
The system performs preliminary analysis of consumer interaction patterns and pre-computes search rule improvements before they are needed. By proactively analyzing streaming events and preparing search rule updates in advance, the system reduces the computational resources required during actual search operations while maintaining high reliability.
3Productivity
If static search rules are used, then search operations can be performed efficiently, but search recall deteriorates when rules become outdated or incomplete
Solution Approach 1:
The system transforms static search rules into dynamic, adaptive rules that automatically evolve based on streaming interaction events. The search rules continuously adapt to changing consumer behavior patterns while maintaining operational efficiency, resolving the contradiction between efficiency and adaptability through real-time automated updates.
Solution Approach 2:
The system performs preliminary analysis of interaction patterns and pre-adapts search rules before changes in consumer behavior become problematic. By proactively detecting trends in streaming events and updating rules in advance, the system maintains both high productivity and adaptability without requiring frequent reactive adjustments.
Data Source
AI summary
Techniques and system configurations for identifying search recall activities and performing changes to search criteria based on streaming event data are disclosed. In an example, electronic operations used for identifying search recall scenarios based on streaming events in a user interface include: obtaining a plurality of streaming events that represent user interaction from a user interface sessions, from inputs that are used to locate and select items; identifying search recall scenarios from the events based on vicinity and exposure, where the items are not retrieved by initial searches; determining changes to selection criteria, to locate and select the items; and updating the selection criteria such that subsequent searches are configured to locate the items using the updated search criteria. With this technique, search data and rules such as exclusion and inclusion lists or category/product information rules may be automatically updated to successfully locate product items in subsequent searches.