Ontology Graph Query Suggestions Using Semantic Partitions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for providing query suggestions rely on query logs or document corpora, which can lead to inaccurate or irrelevant suggestions due to dependence on user data and sensitivity to document corpora, limiting their effectiveness when query logs are unavailable or when user queries differ from the training data.
Innovation Solution
The system uses an ontology graph to map partially completed query requests to semantically relevant ontology elements, generating suggestions based on functional partitions within the graph, eliminating the need for query logs and reducing sensitivity to document corpora by focusing on concept relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If query logs or document corpora are used to generate query suggestions, then the system can provide suggestions based on user data, but the suggestions become inaccurate or irrelevant when query logs are unavailable or when user queries differ from training data
Solution Approach 1:
The patent introduces an ontology graph as an intermediary between the query input and suggestion generation process. The ontology graph contains structured knowledge about concepts, entities, and their relationships, serving as a mediator that transforms user queries into semantically meaningful suggestions. This intermediary enables the system to generate accurate suggestions based on semantic understanding rather than relying solely on statistical patterns from query logs or document corpora.
Solution Approach 2:
The patent changes the fundamental parameter of suggestion generation from statistical frequency-based (using query logs and document corpora) to semantics-based (using ontology graphs). By transforming the approach from counting co-occurrences to understanding conceptual relationships, the system achieves reliable suggestions even when query logs are unavailable or when queries differ from training data, as the ontology provides domain knowledge independent of specific user query patterns.
2Productivity
If the system relies on query logs for suggestions, then suggestions can be generated from user data, but the system becomes sensitive to the availability and quality of query logs
Solution Approach 1:
The ontology graph serves as an intermediary knowledge base that enables suggestion generation without direct dependency on query logs. The system can generate suggestions by querying the ontology for related concepts and entities, making the suggestion generation capability independent of the availability and quality of query logs while maintaining productivity.
Solution Approach 2:
The ontology graph provides a universal knowledge foundation that serves multiple functions: it enables suggestion generation, provides semantic understanding, and works across different domains and query types. This multi-functionality reduces data dependency by providing a consistent knowledge base that works regardless of the specific query log characteristics or domain.
3Loss of information
If document corpora are used for training, then suggestions can be generated from textual data, but the system becomes sensitive to the specific document corpora used
Solution Approach 1:
The patent changes the parameter of knowledge representation from unstructured or semi-structured text in document corpora to structured semantic relationships in an ontology graph. This transformation preserves semantic understanding while reducing sensitivity to specific corpora, as the ontology explicitly models concepts and relationships in a domain-independent manner that can serve multiple query scenarios.
Solution Approach 2:
The ontology graph acts as an intermediary layer between document corpora and suggestion generation. Instead of directly using patterns from document corpora, the system uses the ontology to interpret and structure the knowledge, thereby preserving semantic understanding while reducing sensitivity to the specific corpora used for training or reference.
Data Source
AI summary
One embodiment provides a method, including: utilizing at least one processor to execute computer code that performs the steps of: receiving a partially completed query request from a user, wherein the partially completed query request comprises at least one text string provided in a query field; mapping the at least one text string to at least one ontology element contained within at least one of a plurality of partitions in an ontology graph, the graph comprising (i) ontology elements represented as nodes and (ii) connections between ontology elements represented as edges, wherein each of the plurality of partitions corresponds to a subject area of a field and comprises ontology elements corresponding to concepts related to the subject area of the field; and generating, in view of the plurality of partitions, at least one suggestion for completing the query request to the user, wherein the at least one suggestion comprises a semantically relevant suggestion corresponding to the at least one text string provided in the query field. Other aspects are described and claimed.