Multi-Phrase Search Segmentation via Knowledge Graph Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current devices fail to accurately distinguish between separate search commands in multi-phrased search strings, leading to misinterpretation of user inputs.
Innovation Solution
A media guidance application that translates phrases into strings of word types and determines their associations using a knowledge graph to identify whether phrases are conversationally connected, allowing for accurate separation of individual commands within a multi-phrased search.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If devices process multi-phrased search strings as single commands, then processing simplicity is maintained, but search accuracy deteriorates due to inability to distinguish separate commands
Solution Approach 1:
The system segments a multi-phrased search string into individual search commands by detecting phrase boundaries. A trained model analyzes the input string to identify where one search command ends and another begins, splitting the string into separate searchable phrases. This segmentation enables accurate processing of multiple distinct search intents within a single user input.
Solution Approach 2:
The patent introduces a trained model as an intermediary component between the search string input and the search execution. This model serves as a mediator that analyzes the input, determines phrase boundaries, and guides the subsequent search processing. The intermediary model resolves the ambiguity in multi-phrased inputs without requiring complex rule-based parsing logic throughout the entire system.
2Reliability
If devices treat all input strings as single commands, then operational simplicity is maintained, but command identification accuracy deteriorates
Solution Approach 1:
The system performs preliminary analysis of the search string before executing searches. The trained model pre-processes the input to identify and separate distinct commands, preparing the data in advance for accurate search execution. This preliminary action ensures that each search command is correctly identified and processed independently, improving reliability without requiring complex operations during the actual search execution.
3Measurement precision
If phrase separation is not implemented, then processing speed is maintained, but search result relevance deteriorates due to misinterpretation of user intent
Solution Approach 1:
The system segments the search string into individual commands using a trained model that quickly identifies phrase boundaries. This segmentation occurs in advance of the actual search execution, allowing each separated command to be processed independently and efficiently. The segmentation improves result relevance by ensuring each search targets the correct user intent while maintaining speed through optimized phrase separation algorithms.
Data Source
AI summary
Systems and methods are provided herein for using a trained knowledge graph to accurately determine whether two phrases are conversationally connected. To this end, three phrases are received. The phrases are translated to strings of word types by determining what type of word each word of each phrase represents, and replacing each word of the first phrase with its respective type. A knowledge graph is then accessed to determine strengths of association between a combinations of each string of word types and any conversational category of a plurality of conversational categories. If any of the strengths of association exceed a threshold, the corresponding strings may be determined to be part of a conversation.