Pre-web Classification for Low Latency Query Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines face high latency in returning results due to lengthy pre-web result ranking processes, especially when processing large volumes of queries, which can lead to a negative user experience.
Innovation Solution
Implementing a low-latency pre-web classification system that uses a vertical intent classification index to identify and rank candidate entity identifiers for query data before it is processed by a web ranking service, allowing for efficient processing and reduced latency in returning results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pre-web result ranking is performed to bias results towards a particular topic or entity, then result relevance is improved, but processing time increases causing high latency
Solution Approach 1:
The patent applies preliminary action by performing vertical intent classification and entity identification on query data before it reaches the web ranking service. The pre-web classification component analyzes the query intent and identifies candidate entities in advance, so that when the web ranking service receives the pre-processed query, it can immediately bias results towards the identified topic or entity without performing lengthy ranking processes, thus reducing latency while maintaining result relevance
Solution Approach 2:
The patent segments the query processing into distinct components: a pre-web classification component that performs vertical intent classification and entity identification, and a web ranking service that performs result ranking. This segmentation allows the classification and entity identification to be performed separately before web indexing, enabling the web ranking service to focus only on ranking results biased towards the pre-identified topic or entity, thereby reducing overall processing time while maintaining relevance
2Productivity
If main web indexes are searched to return results, then query processing is completed, but a large part of allowable latency is consumed
Solution Approach 1:
The patent performs preliminary action by conducting vertical intent classification and candidate entity identification before the web ranking service processes the query. This pre-processing establishes the query intent and candidate entities in advance, allowing the main web index search to be more targeted and efficient, thereby completing query processing while consuming less of the allowable latency
Solution Approach 2:
The patent introduces an intermediary pre-web classification component that sits between the query reception and the web ranking service. This intermediary performs vertical intent classification and entity identification, creating a bridge that prepares the query data in a format that enables more efficient web index searching, thus improving productivity while reducing the time consumed in the main processing pipeline
Data Source
AI summary
Non-limiting examples of the present disclosure describe low latency pre-web classification of query data. In examples, processing is performed where query data may be analyzed in a low latency manner that includes providing a vertical intent classification and entity identification for query data before a web ranking service processes the query data. Query data may be received. A vertical intent classification index may be searched using the query data. In examples, the vertical intent classification index may comprise a set of files that can be used to determine one or more candidate entity identifiers for the query data. The one or more entity identifiers may be ranked. The query data, a vertical intent classification for the vertical intent classification index and the one or more ranked candidate entity identifiers may be transmitted for processing associated with a web ranking service. Other examples are also described.