A number of studies on network analysis have focused on language networks based on free word association, which reflects human lexical knowledge, and have demonstrated the small-world and scale-free properties in the word association network. Nevertheless, there have been very few attempts at applying network analysis to distributional semantic models, despite the fact that these models have be...

In this article we present a neural network model of sentence generation. The network has both technical and conceptual innovations. Its main technical novelty is in its semantic representations: the messages which form the input to the network are structured as sequences, so that message elements are delivered to the network one at a time. Rather than learning to linearise a static semantic re...

In 1993, we started to develop a tool to help people manage structured personal and business information. Initially named ORCCA (Online Resources for Corporate Citizens Action), it was fully implemented as a Semantic Networks Editor and Manager. Initially designed as a personal desktop tool, it evolved towards a collaborative server allowing small groups to share Semantic Networks. It has been ...

This paper presents a lightweight Multi-task Semantic Attention Network (MTSAN) to collectively deal with object detection as well semantic segmentation aiding real-time applications of the Advanced Driver Assistance Systems (ADAS). proposes Module (SAM) that introduces contextual clues from subnet guide subnet. The SAM significantly boosts up performance and computational cost by considerably ...

Automatic hypermedia presentation generation uses an information source semantic network first to select the content and then to compose it in the presentation so that the semantic relations between the information items are conveyed to the user. A hypermedia presentation can be considered as an example of strongly-guided navigation through a hypermedia structure. This can provide a high level ...

OBJECTIVES This paper proposes a novel semantic method for auditing associative relations in biomedical terminologies. We tested our methodology on two Unified Medical Language System (UMLS) knowledge sources. METHODS We use the UMLS semantic groups as high-level representations of the domain and range of relationships in the Metathesaurus and in the Semantic Network. A mapping created betwee...