An NMT-Based Approach to Translate Natural Language Questions to SPARQL Queries

Lin, Jia-Huei; Lu, Eric Jui-Lin

doi:10.1007/978-981-16-4118-3_1

Cited by 3 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As can be seen, the previous studies [2,3] which can perfectly deal with closed-domain question answering datasets all fail to handle the LC-QuAD [9] dataset, where the highest accuracy they reached is only about 60%. On the contrary, our preliminary study [18] shows an outstanding accuracy (78.07%) using QALD-7 dataset, where the QALD datasets are considered as significantly more complex datasets LC-QuAD [19]. The preliminary experimental results not only show that our approach is feasible to solve open-domain question answering datasets, but also encourage us for further investigation.…”

Section: Sparql Generation Using Nmt Modelsmentioning

confidence: 71%

“…Experimental results of this approach on the QALD-7 dataset have been published in our preliminary study [18]. The outstanding results encouraged us to pursue further.…”

Section: Introductionmentioning

confidence: 94%

“…In [18], we proposed an NMT-based approach with the Transformer model to automatically translate natural language questions to SPARQL queries and presented our preliminary results. A Transformer model is chosen to train our SPARQL generator due to its relatively high efficiency than RNN-based models as well as the effectiveness in various natural language processing tasks [17].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SPARQL Generation with an NMT-based Approach

Lin

2022

JWE

Self Cite

View full text Add to dashboard Cite

SPARQL is a powerful query language which has been widely used in various natural language question answering (QA) systems. As the advances of deep neural networks, Neural Machine Translation (NMT) models are employed to directly translate natural language questions to SPARQL queries in recent years. In this paper, we propose an NMT-based approach with Transformer model to generate SPARQL queries. Transformer model is chosen due to its relatively high efficiency and effectiveness. We design a format to encode a SPARQL query into a simple sequence with only RDF triples reserved. The main purpose of this step is to shorten the sequences and reduce the complexity of the target language. Moreover, we employ entity type tags to further resolve mistranslated problems. The proposed approach is evaluated against three open-domain question answering datasets (QALD-7, QALD-8, and LC-QuAD) on BLEU score and accuracy, and obtains outstanding results (83.49%, 90.13%, and 76.32% on BLEU score, respectively) which considerably outperform all known studies.

show abstract

Section: Sparql Generation Using Nmt Modelsmentioning

confidence: 71%

“…Experimental results of this approach on the QALD-7 dataset have been published in our preliminary study [18]. The outstanding results encouraged us to pursue further.…”

Section: Introductionmentioning

confidence: 94%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SPARQL Generation with an NMT-based Approach

Lin

2022

JWE

Self Cite

View full text Add to dashboard Cite

show abstract

“…The evaluation stage uses the generated query syntax to retrieve answers from the knowledge base and provides them to the user. Currently, research on query generation syntax can be classified broadly into two categories [8], i.e., (1) template-free approaches [9][10][11][12], which involve constructing SPARQL query syntax by utilizing the syntactic structure of question sentences, and (2) template-based approaches [13,14], which involve mapping question sentences to manually or automatically create SPARQL query templates.…”

Section: Introductionmentioning

confidence: 99%

“…There are three main categories of template-free approaches, i.e., using knowledge graph structures [10], dependency structures [9], and machine translation [11,12]. The knowledge graph structure approach [10] utilizes a knowledge graph structure and a subgraph search algorithm to find all possible RDF triples and their matching subgraphs in the entity mapping step.…”

Section: Introductionmentioning

confidence: 99%

Boosting Question Answering Systems with Multi-Label Classification Techniques

Chen

Jin-De

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Enhancing SPARQL Query Performance With Recurrent Neural Networks

Chen,

Lu,

Lin

2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

DBpedia is one of the most resourceful link databases today, and to access information in DBpedia databases, we need to use query syntax (e.g., SPARQL). However, not all users know SPARQL, so we must use a natural language query system to translate the user's query into the corresponding query syntax. Generating query syntax through the query system is both time-consuming and expensive. To improve the efficiency of query syntax generation from user questions, the multi-label template approach, specifically Light-QAwizard, is utilized. Light-QAwizard transforms the problem into one or more singlelabel classifications using multi-label learning template approaches. By implementing Light-QAwizard, query costs can be reduced by 50%, but it introduces a new label during the transformation process leading to sample imbalance, compromised accuracy, and limited scalability. To overcome these limitations, this paper employs employ two multi-label learning methods, namely Binary Relevance (BR) and Classifier Chains (CC), for question transformation. By employing Recurrent Neural Networks (RNNs) as a multi-label classifier for generating RDF triples, we predict all the labels that align with the query intentions. To better account for the relationship between RDF triples, we integrate BR into an ensemble learning approach, resulting in the Ensemble BR. Experimental results demonstrate that our proposed method outperforms previous research in terms of improving query accuracy.

show abstract

An NMT-Based Approach to Translate Natural Language Questions to SPARQL Queries

Cited by 3 publications

References 5 publications

SPARQL Generation with an NMT-based Approach

SPARQL Generation with an NMT-based Approach

Boosting Question Answering Systems with Multi-Label Classification Techniques

Enhancing SPARQL Query Performance With Recurrent Neural Networks

Contact Info

Product

Resources

About