Search

The uncertainty is an inherent feature of Knowledge Graph (KG), which is often modelled as confidence scores of relation facts. Although Knowledge Graph Embedding (KGE) has been a great success recently, it is still a big challenge to predict confidence of unseen facts in KG in the continuous vector space. There are several reasons for this situation. First, the current KGE is often concerned with the deterministic knowledge, in which unseen facts’ confidence are treated as zero, otherwise as one. Second, in the embedding space, uncertainty features are not well preserved. Third, approximate reasoning in embedding spaces is often unexplainable and not intuitive. Furthermore, the time and space cost of obtaining embedding spaces with uncertainty preserved are always very high. To address these issues, considering Uncertain Knowledge Graph (UKG), we propose a fast and effective embedding method, UKGsE, in which approximate reasoning and calculation can be quickly performed after generating an Uncertain Knowledge Graph Embedding (UKGE) space in a high speed and reasonable accuracy. The idea is that treating relation facts as short sentences and pre-handling are benefit to the learning and training confidence scores of them. The experiment shows that the method is suitable for the downstream task, confidence prediction of relation facts, whether they are seen in UKG or not. It achieves the best tradeoff between efficiency and accuracy of predicting uncertain confidence of knowledge. Further, we found that the model outperforms state-of-the-art uncertain link prediction baselines on CN15k dataset.

It is a challenge to transform legacy XML-based data into ontology for applications of the semantic web, such as semantic-based integrations, intelligent web searching, and internet based knowledge reasoning. We propose a new technique to transform XML data into ontology data automatically by modeling XML documents semantically. Firstly, we provide the XML a semantically interpretation by developing a graph-based formal language, w-graph. Then, the result of the interpretation can be automatically mapped into OWL web ontology language with semantics preserved. The proof of semantics preserved also been considered, and automatic mapping tool has been developed.

The extensible markup language (XML), a standard format of web information, has a clear syntax but unfortunately an ambiguous formal semantics, which results in being not used directly in semantic web applications. So it is tough job to reuse XML-based data intelligently in the semantic web. To address this problem, a new formal technique of obtaining ontology data automatically from XML documents is proposed. We provide the XML a semantical interpretation by developing a graph-based formal language, which then can be automatically mapped into web ontology language OWL with semantics preserved. The semantic validity and entailment problem are also concerned. The automatical mapping tool has also been developed.

Emotional agents are useful to variety of computer application. This paper focuses on the emotion surprise. Surprise is the automatic reaction to a mismatch, which plays an important role in the behaviors of intelligent agents. We represent psychology theories of surprise through a fuzzy inference system, as fuzzy logic helps to capture the fuzzy and complex nature of emotions. The system takes four factors related to surprise as inputs and the degree of surprise as output. We put forward fuzzy inference rules and reasoning parameters for the system. The results of the inference can be used in single or multi-agent system simulation.

A big challenge for automatically analyzing patent claims written in Chinese language is how to obtain semantically information from claims which are usually a non-structured, but free text. In this paper, a set of techniques is been provided to extract some valuable semantically information from claims in Chinese language. The method could automatically discovery some usable semantically information from the patent claim texts by means of regular expression pattern and predefined ontology model. It can extract not only surface semantic information but also deeper semantic information. Furthermore, the extracted semantically information is automatically translated into web ontology language (OWL), a machine readable semantic structure specification. The work proposes a potential semantic solution in Chinese language patent analyzing based on patent claims, such as semantic searching, legal status and invalidation checking, and discovering new technique trends. A case study on electronic engineering domain patent claims is also provided.

One of key issues of the Semantic Web applications is the lack of semantic data (ontologies). Although the vast majority of data are stored in the popular relational databases, they are still not easily available for many next generation Web applications. Therefore, one of core challenges of Semantic Web is whether these applications can automatically retrieve semantic information from the existed relational databases. This paper proposes a middle graph-based formal model language, W-graph, a bridge between relational databases and ontologies, which abstracts semantic information from relational database instances semi-automatically and then generates an OWL ontology automatically. This method not only maps relational database schemata to ontologies, but also populates ontologies with data stored in databases. Moreover, a proof of semantic preserving on the mapping is provided, and a case study and an implemented prototype tool are also reported.

The analysis of safety properties of a system is an important way to study correctness of design and implementation,and that is an urgent requirement in some systems, such as hybrid embedded systems. But it is always difficult to make good and accurate safety analyze. We proposed a formalism, TCSP for hybrid embedded systems, by introducing a new operator (emitting signal operator) into TCSP and giving details of its structural operational semantics to analyze formally safe specification of hybrid embedded systems. Not only has safety for actions of hybrid embedded systems been analyzed, but also safety for predicates.

We propose a process algebra TCSPhs obtained by extending process algebra timed CSP (timed communicating sequential processes [10, 11]) with a new propositional signal operator, inspired by [2]. The proposed process algebra makes it possible to deal well with behaviors and security features of hybrid system. Motivation is twofold. On the one hand we offer structural operational semantics of the proposed process algebra language. On the other hand we concentrate on investigating safety specification aspects of hybrid system within that language.