A question-answering (QA) system is essential for helping users retrieve relevant and accurate answers based on their queries. The precision of SPARQL query syntax generation is directly linked to the accuracy of the answers provided. Recently, many studies on knowledge graph-based natural language question-answering (KGQA) systems have leveraged the Neural Machine Translation (NMT) framework to translate input questions into SPARQL query syntax, a process known as Text-to-SPARQL. In NMT, cross-attention-based Transformers, ConvS2S, and BiLSTM models are commonly used for training. However, comparing the translation performance of these models is challenging due to their significant architectural differences. To address this issue, this paper integrates various encoder and cross-attention methods with a fixed LSTM decoder to form hybrid models, which are then trained and evaluated on QA systems.Beyond the hybrid models discussed, this study introduces an improved ConvS2S architecture featuring a Multi-Head Convolutional (MHC) encoder, designated as QAWizer_MHC. The MHC encoder incorporates the Transformer’s multi-head attention mechanism to compute dependencies within the input sequence. Additionally, the enhanced ConvS2S model captures local hidden features across different receptive fields within the input sequence. Experimental results demonstrate that QAWizer_MHC outperforms other models, achieving BLEU-1 scores of 76.52% and 83.37% on the QALD-9 and LC-QuAD-1.0 datasets, respectively. Furthermore, in end-to-end system evaluations on the same datasets, the model attained Macro F1 scores of 52% and 66%, respectively, surpassing other KGQA systems. The experimental findings indicate that even with limited computational resources and general embeddings, a well-designed encoder–decoder architecture that integrates cross-attention can achieve performance comparable to large pre-trained models.

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.

To mitigate the semantic gap between a visual feature and linguistic representation in social image retrieval, researchers have proposed an automatic image annotation approach, which employs multiple visual features and text matching to label images. This paper used a linked open data approach and ontology to construct a model, an automatic semantic image annotation model for social image retrieval. These models enable users to label images through automatic semantic annotation and to identify the underlying intents of semantics, thereby fulfilling user needs and enhancing the retrieval accuracy.

Most clustering algorithms build disjoint clusters. However, clusters might be overlapped because documents may belong to two or more categories in the real world. For example, a paper discussing the Apple Watch may be categorized into either 3C, Fashion, or even Clothing and Shoes. Therefore, overlapping clustering algorithms have been studied such that a resource can be assigned to one or more clusters. Formal Concept Analysis (FCA), which has many practical applications in information science, has been used in disjoin clustering, but has not been studied in overlapping clustering. To make overlapping clustering possible by using FCA, we propose an approach, including two types of transformation. From the experimental results, it shows that the proposed fuzzy overlapping clustering performed more efficiently than existing overlapping clustering methods. The positive results confirm the feasibility of the proposed scheme used in overlapping clustering. Also, it can be used in applications such as recommendation systems.

Today, networks can be seen as a large-scale knowledge base which contains a huge amount of resources. To be able to use them effectively, metadata should be associated with each sharable resource. As a result, authoring systems supporting metadata have been widely studied. In this paper, we attempt to develop an authoring system supporting metadata. From our previous research, it is found that "RELATION" is an important metadata, and a relation ontology was proposed. However, existing authoring systems do not support the relation ontology. Therefore, we continued our previous work and developed a Web-based authoring system supporting the relation ontology. With the authoring system, we believe the reusability of network resources can be significantly increased.

Abstract Since electronic data interchange (EDI) is one of the most important components in electronic commerce and since Extensible Markup Language (XML) provides Internet developers with a powerful vehicle for exchanging messages, XML/EDI has received much attention from almost all well‐known international enterprises and is believed to be the next generation EDI. However, since XML allows developers to design their own elements and attributes, it is almost certain businesses will receive XML documents with unknown elements. Generally, human intervention is required to solve the problem. Therefore, it is indispensable to design an efficient scheme to resolve the unknown elements. In this paper, we propose a ReScUE XML/EDI model such that transformation templates will be generated automatically for documents with unknown elements, and the documents will be converted into expected format. This model increases the flexibility of XML/EDI systems and reduces human intervention. Copyright © 2003 John Wiley & Sons, Ltd.

In a (t, n) threshold proxy signature scheme, the original signer delegates the power of signing messages to a designated proxy group of n members. Any t or more proxy signers of the group can cooperatively issue a proxy signature on behalf of the original signer, but (t - 1) or less proxy signers cannot. Previously, all of the proposed threshold proxy signature schemes have been based on the discrete logarithm problem and do not satisfy all proxy requirements. In this paper, we propose a practical, efficient, and secure (t, n) threshold proxy signature scheme based on the RSA cryptosystem. Our scheme satisfies all proxy requirements and uses only a simple Lagrange formula to share the proxy signature key. Furthermore, our scheme requires only 5 percent of the computational overhead and 8 percent of the communicational overhead required in Kim's scheme.

The simulation of many-body, many-particle systems has a wide range of applications in areas such as biophysics, chemistry, astrophysics, etc. It is known that the force calculation contributes 90% of the simulation time. This is mainly due to the fact that the total number of interactions in the force is O(N/sup 2/), where N is the number of particles in the system. The fast multipole algorithm proposed by Greengard and Rokhlin (1987) reduces the time complexity to O(N). In this paper, we design an efficient, parallel fast multipole algorithm in 3D. For portability, our parallel program is implemented using the Message Passing Interface. Is it possible to obtain high performance for a computationally-intensive application using a LAN of workstations? In this paper, we attempt to answer this question, which is commonly asked by those researchers who have no access to parallel computers or supercomputers.