Querying Knowledge Graphs in Natural Language

Abstract

Knowledge graphs are a powerful concept for querying large amounts of data. These knowledge graphs are typically enormous and are often not easily accessible to end-users because they require specialized knowledge in query languages such as SPARQL. Moreover, end-users need a deep understanding of the structure of the underlying data models often based on the Resource Description Framework (RDF). This drawback has led to the development of Question-Answering (QA) systems that enable end-users to express their information needs in natural language. While existing systems simplify user access, there is still room for improvement in the accuracy of these systems. In this paper we propose a new QA system for translating natural language questions into SPARQL queries. The key idea is to break up the translation process into 5 smaller, more manageable sub-tasks and use ensemble machine learning methods as well as Tree-LSTM-based neural network models to automatically learn and translate a natural language question into a SPARQL query. The performance of our proposed QA system is empirically evaluated using the two renowned benchmarks - the 7th Question Answering over Linked Data Challenge (QALD-7) and the Large-Scale Complex Question Answering Dataset (LC-QuAD). Experimental results show that our QA system outperforms the state-of-art systems by 15% on the QALD-7 dataset and by 48% on the LC-QuAD dataset, respectively. In addition, we make our source code available.

A survey on semantic question answering systems

“…Examples of SQA systems using KB open domain are as follows: Swipe, Casia@V2, Deanna, SMIQ, YagoQA, YodaQA, Qanswer, gAnswer, ISOFT, SQUALL2SPARQL, Scalewelis, Intui2, Intui3, OAKIS, SemSek, Alexandria, RTV, Sorokin and Gurevych (2017), Athreya et al (2021) and Liang et al (2021). The most widely used and most popular open domain KB is DBpedia.…”

“…SWIP detects, factual, list, boolean, and count. Sorokin and Gurevych (2017), Athreya et al (2021) and Liang et al (2021) were evaluated in a test question set of QALD-7 that includes list, factual, boolean, count type questions. Mishra and Jain (2016) present the following categories for the analysis of questions: morphological analysis, syntactical analysis, semantic analysis, pragmatic and discourse analysis, expected answer type analysis, and focus recognition of questions.…”

“…The system of Liang et al (2021): the system applies tokenization, POS tagging, lemmatization, and dependency parsing. The question type classification is identified using machine learning (ML).…”

Ontology Driven Testing Strategies for IoT Applications

“…Pellet Reasoner is used to validate our ontology model. Previously researchers and ontology engineers use reasoner to validate their work [29][30][31]. It can be used in unification with Jena and OWL APIs.…”

Knowledge Graph-based Framework for Domain Expertise Elicitation and Reuse in e-Learning

“…LW represents the e-course that is traditionally designed by instructors. In our case LW is automatically generated from the www.ijacsa.thesai.org knowledge graph base which is queried to retrieve the knowledge subgraph containing the necessary knowledge concepts to fulfill the requirement of learning for a specific learner [17], [18]. These concepts are then ordered to form a tree of concepts.…”