In the field of question answering-based knowledge graphs, due to the complexity of the construction of knowledge graphs, a domain-specific knowledge graph often cannot contain some common-sense knowledge, which makes it impossible to answer questions that involve common-sense and domain knowledge at the same time. Therefore, this study proposes a knowledge graph-based question answering method in the computer science domain, which facilitates obtaining complete answers in this domain. In order to solve the problem of natural language problems being difficult to match with structured knowledge, a series of logic rules are first designed to convert natural language into triples of the question. Then, a semantic query expansion strategy based on WordNet is proposed and a priority marking algorithm is proposed to mark the order of triples of the question. Finally, when a question triple corresponds to multiple triples in the knowledge graph, it can be solved by the proposed SimCSE-based similarity method. The designed logic rules can deal with each type of question in a targeted manner according to the different question words and can effectively transform the question text into question triples. In addition, the proposed priority marking algorithm can effectively mark the order in the triple of the question. MKBQA can answer not only computer science-related questions but also extended open domain questions. In practical applications, answering a domain question often cannot rely solely on one knowledge graph. It is necessary to combine domain knowledge and common-sense knowledge. The MKBQA method provides a new idea and can be easily migrated from the field of computer science to other fields. Experiment results on real-world data sets show that, as compared to baselines, our method achieves significant improvements to question answering and can combine common-sense and domain-specific knowledge graphs to give a more complete answer.
Entity alignment is an important task in knowledge fusion, which aims to link entities that have the same real-world identity in two knowledge graphs. However, in the process of constructing a knowledge graph, some noise may inevitably be introduced, which must affect the results of the entity alignment tasks. The triple confidence calculation can quantify the correctness of the triples to reduce the impact of the noise on entity alignment. Therefore, we designed a method to calculate the confidence of the triples and applied it to the knowledge representation learning phase of entity alignment. The method calculates the triple confidence based on the pairing rates of the three angles between the entities and relations. Specifically, the method uses the pairing rates of the three angles as features, which are then fed into a feedforward neural network for training to obtain the triple confidence. Moreover, we introduced the triple confidence into the knowledge representation learning methods to improve their performance in entity alignment. For the graph neural network-based method GCN, we considered entity confidence when calculating the adjacency matrix, and for the translation-based method TransE, we proposed a strategy to dynamically adjust the margin value in the loss function based on confidence. These two methods were then applied to the entity alignment, and the experimental results demonstrate that compared with the knowledge representation learning methods without integrating confidence, the confidence-based knowledge representation learning methods achieved excellent performance in the entity alignment task.
Graph convolutional network-based methods have become mainstream for cross-language entity alignment. The graph convolutional network has multi-order characteristics that not only process data more conveniently but also reduce the interference of noise effectively. Although the existing methods have achieved good results for the task of cross-language entity alignment, they have often overlooked the same entity names in the real corpus, resulting in an entity-matching result that was not ideal. Therefore, this study proposed a neighboring-entity-screening rule by combining the entity name and the attribute (NENA) to reduce the influence of these issues. We used the NENA-screening rule to filter and delete redundant equivalent entities and to construct a dual-relation graph as auxiliary evidence for scenarios when the attribute information may be insufficient.This study adopted a graph convolutional network in order to embed knowledge graphs and entity names into a unified vector space, and then a down-sampling method was used to extract the neighboring entities of each entity, thus forming sub-graphs of the two knowledge graphs. We embedded the sub-graphs into the GCN, as the new input, and then we used a cross-graph-matching module to finally achieve alignment. Our results on the DBP15K dataset showed that our approach significantly improved the overall entity alignment.On the sub-dataset ZH-EN of DBP15K, the value of Hits@1 improved by 1.38%, as compared to the best approach mentioned in this paper, and it was useful for the construction and completion of the open knowledge graph.
10 February 2023 in Data Technologies and Applications
Purpose: Intent detection and slot filling are two important tasks in question comprehension of a question answering system. This study aims to build a joint task model with some generalization ability and benchmark its performance over other neural network models mentioned in this paper.Design/methodology/approach: This study used a deep-learning-based approach for the joint modeling of question intent detection and slot filling. Meanwhile, the internal cell structure of the long short-term memory (LSTM) network was improved. Furthermore, the dataset Computer Science Literature Question (CSLQ) was constructed based on the Science and Technology Knowledge Graph. The datasets Airline Travel Information Systems, Snips (a natural language processing dataset of the consumer intent engine collected by Snips) and CSLQ were used for the empirical analysis. The accuracy of intent detection and F1 score of slot filling, as well as the semantic accuracy of sentences, were compared for several models.Findings: The results showed that the proposed model outperformed all other benchmark methods, especially for the CSLQ dataset. This proves that the design of this study improved the comprehensive performance and generalization ability of the model to some extent.Originality/value: This study contributes to the understanding of question sentences in a specific domain. LSTM was improved, and a computer literature domain dataset was constructed herein. This will lay the data and model foundation for the future construction of a computer literature question answering system.
About Me
Her areas of interest include knowledge graph and explainable AI.
Her areas of interest include knowledge graph and explainable AI.