Domain- and Language-Adaptable Natural Language Interface for Property Graphs

Despite the growing adoption of Property Graph Databases, like Neo4j, interacting with them remains difficult for non-technical users due to the reliance on formal query languages. Natural Language Interfaces (NLIs) address this by translating natural language (NL) into Cypher. However, existing sol...

Full description

Saved in:
Bibliographic Details
Main Authors: Ioannis Tsampos, Emmanouil Marakakis
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Computers
Subjects:
Online Access:https://www.mdpi.com/2073-431X/14/5/183
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327372591955968
author Ioannis Tsampos
Emmanouil Marakakis
author_facet Ioannis Tsampos
Emmanouil Marakakis
author_sort Ioannis Tsampos
collection DOAJ
description Despite the growing adoption of Property Graph Databases, like Neo4j, interacting with them remains difficult for non-technical users due to the reliance on formal query languages. Natural Language Interfaces (NLIs) address this by translating natural language (NL) into Cypher. However, existing solutions are typically limited to high-resource languages; are difficult to adapt to evolving domains with limited annotated data; and often depend on Machine Learning (ML) approaches, including Large Language Models (LLMs), that demand substantial computational resources and advanced expertise for training and maintenance. We address these limitations by introducing a novel dependency-based, training-free, schema-agnostic Natural Language Interface (NLI) that converts NL queries into Cypher for querying Property Graphs. Our system employs a modular pipeline-integrating entity and relationship extraction, Named Entity Recognition (NER), semantic mapping, triple creation via syntactic dependencies, and validation against an automatically extracted Schema Graph. The distinctive feature of this approach is the reduction in candidate entity pairs using syntactic analysis and schema validation, eliminating the need for candidate query generation and ranking. The schema-agnostic design enables adaptation across domains and languages. Our system supports single- and multi-hop queries, conjunctions, comparisons, aggregations, and complex questions through an explainable process. Evaluations on real-world queries demonstrate reliable translation results.
format Article
id doaj-art-4a6aebf845294a899926fa99af2b9b8f
institution Kabale University
issn 2073-431X
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Computers
spelling doaj-art-4a6aebf845294a899926fa99af2b9b8f2025-08-20T03:47:53ZengMDPI AGComputers2073-431X2025-05-0114518310.3390/computers14050183Domain- and Language-Adaptable Natural Language Interface for Property GraphsIoannis Tsampos0Emmanouil Marakakis1Department of Electrical & Computer Engineering, Hellenic Mediterranean University, 71410 Heraklion, GreeceDepartment of Electrical & Computer Engineering, Hellenic Mediterranean University, 71410 Heraklion, GreeceDespite the growing adoption of Property Graph Databases, like Neo4j, interacting with them remains difficult for non-technical users due to the reliance on formal query languages. Natural Language Interfaces (NLIs) address this by translating natural language (NL) into Cypher. However, existing solutions are typically limited to high-resource languages; are difficult to adapt to evolving domains with limited annotated data; and often depend on Machine Learning (ML) approaches, including Large Language Models (LLMs), that demand substantial computational resources and advanced expertise for training and maintenance. We address these limitations by introducing a novel dependency-based, training-free, schema-agnostic Natural Language Interface (NLI) that converts NL queries into Cypher for querying Property Graphs. Our system employs a modular pipeline-integrating entity and relationship extraction, Named Entity Recognition (NER), semantic mapping, triple creation via syntactic dependencies, and validation against an automatically extracted Schema Graph. The distinctive feature of this approach is the reduction in candidate entity pairs using syntactic analysis and schema validation, eliminating the need for candidate query generation and ranking. The schema-agnostic design enables adaptation across domains and languages. Our system supports single- and multi-hop queries, conjunctions, comparisons, aggregations, and complex questions through an explainable process. Evaluations on real-world queries demonstrate reliable translation results.https://www.mdpi.com/2073-431X/14/5/183property graph databasesknowledge graphsnatural language interfaces (NLIs)automated query generationtext-to-cypher translation
spellingShingle Ioannis Tsampos
Emmanouil Marakakis
Domain- and Language-Adaptable Natural Language Interface for Property Graphs
Computers
property graph databases
knowledge graphs
natural language interfaces (NLIs)
automated query generation
text-to-cypher translation
title Domain- and Language-Adaptable Natural Language Interface for Property Graphs
title_full Domain- and Language-Adaptable Natural Language Interface for Property Graphs
title_fullStr Domain- and Language-Adaptable Natural Language Interface for Property Graphs
title_full_unstemmed Domain- and Language-Adaptable Natural Language Interface for Property Graphs
title_short Domain- and Language-Adaptable Natural Language Interface for Property Graphs
title_sort domain and language adaptable natural language interface for property graphs
topic property graph databases
knowledge graphs
natural language interfaces (NLIs)
automated query generation
text-to-cypher translation
url https://www.mdpi.com/2073-431X/14/5/183
work_keys_str_mv AT ioannistsampos domainandlanguageadaptablenaturallanguageinterfaceforpropertygraphs
AT emmanouilmarakakis domainandlanguageadaptablenaturallanguageinterfaceforpropertygraphs