Storytelling is a common creative activity for children. During storytelling, children need support and are subjected to cognitive challenges. This paper explores co-creative agent - StoryDrawer, which supports in creating oral stories through collaborative drawing. StoryDrawer works with two strategies: Child says AI draws; scribbles completes. These strategies allow draw their as an externalization provoke unexpected ideas. presents the interaction design, implementation of conducts user...

Semantically-based watermarking scheme embeds the information by transforming semantic structure of sentence. In this paper we propose an optimized algorithm which introduces TMR string reranking mechanism before watermark embedding. The has feature resisting sentence reordering attack to a certain extent. To evaluate effectiveness our algorithm, conduct few experiments with different conditions and make comparison among results. show that is effective better robustness.

Observation is common for children to connect with nature, increasing their knowledge and awareness of biodiversity. However, it challenging them make document observations due a lack observation drawing skills. Therefore, we designed an interactive tool, Bio Sketchbook, which scaffolds in systematic observation, observational drawing, acquisition. It can recognize plant species generate contour drawings from children's photographs, guiding observe draw multi-dimensional features digital...

Multi-relation embedding is a popular approach to knowledge base completion that learns representations of entities and relations compute the plausibility missing triplet. The effectiveness depends on sparsity KB falls for infrequent only appeared few times. This paper addresses this issue by proposing new model exploiting entity-independent transitive relation patterns, namely Transitive Relation Embedding (TRE). TRE alleviates problem predicting while enjoys generalisation power embedding....

Observational sketching is a common method to facilitate repeated observations that increase individual knowledge and awareness of biodiversity. However, this presents high barrier entry for some novices, particularly young children. In paper, we introduce Bio Sketchbook, novice AI-assisted partner help children draw from observation guide them repeatedly observe different biological features. We describe our design process goals, the interaction results preliminary user study with six Our...

We designed and built Bubble Beats, a breathing exercise game based on music rhythm, which maps into musical gameplay mechanics using sensors-based technology. Beats aims to explore the effectiveness of combining rhythm exercises. In game, children blow bubbles with visual guidance. Initial user tests show that can effectively motivate improve their adherence will continue iterate upgrade system attract match more diverse population, for example, patients recovering from pneumonia other...

This paper describes N-XKT (Neural encoding based on eXplanatory Knowledge Transfer), a novel method for the automatic transfer of explanatory knowledge through neural mechanisms. We demonstrate that is able to improve accuracy and generalization science Question Answering (QA). Specifically, by leveraging facts from background corpora, model shows clear improvement zero-shot QA. Furthermore, we show can be fine-tuned target QA dataset, enabling faster convergence more accurate results. A...

To explore the latent information of Human Knowledge, analysis for Knowledge Bases (KBs) (e.g. WordNet, Freebase) is essential. Some previous KB element embedding frameworks are used KBs structure and completion. These use low-dimensional vector space representation large scale entities relations in KB. Based on that, which not contained by can be measured. The idea reasonable, while current methods have some issues to get proper embeddings elements. entity-relation-entity triplet, most KB,...

Life insurance application requires in-person meetings with underwriters, tedious paperwork, and an average waiting period of six weeks before offer can be made. This outdated process has become a barrier for broader consumer adoption, resulting large coverage gap. In this work, we aim to closing gap by leveraging data mining techniques optimize the questionnaire form. Our experiment on 10 years identified that only ~2% all questions have shown high relevancy determining risks applicants,...

Denial of service (DoS) attack can be easily launched for the simple discovery Protocol (SSDP) on Universal Plug and Play (UPnP). This paper proposes a novel DoS defense scheme UPnP. The connection between UPnP devices nodes will limited, useless descriptions received by control points filtered through setting packet-filter with traffic anomaly detection active rule-based XML firewall points. Finally, analyzes performance this scheme, which shows its particular advantages.

This paper presents an automatic term extraction method based on Markov process. The aims to extract multi-word domain terms from English corpora. proves that the extracting process is a chain firstly, and then gives steps of Markov-based method. In order evaluate our method, we use corpus related computer science got by Web crawlers, methods introduced in paper. experiment data shows out performs other methods.

This paper proposes a new strategy of extracting boundary points from scanning point cloud (SPC) data sheet metal parts (SMPs). is suitable for bending SMPs with slowly changing surfaces. To cope the problem that SMP too thin to have enough its lateral surface be calculated outline, are obtained by moving ridge which maximum curvature on marginal along theoretical position direction. In this article, explained firstly and then carried out two different experimental SMPs. The contains several...

Augmented reality (AR) is a technology that integrates 3D virtual objects into the physical world in real-time, while (VR) immerses users an interactive environment. The fast development of augmented and technologies has reshaped how people interact with world. This presentation will outline results from two unique AR one Web-based VR coastal engineering projects, motivating next stage package for students, engineers, planners.

User preference prediction aims to predict a users future preferences on large number of items according his/her history. To achieve this goal, many models have been proposed, but mainly for explicit data, such as 5-star ratings. Nevertheless, real-world data are often in implicit format, purchase action, and the is not always large. In paper, we demonstrate use latent factor solving task predicting user low-dimensional dataset.

In recent years, problem solving, automatic proof and human-like test-tasking have become a hot spot of research.This paper focus on the study solving physical in Chinese.Based analysis corpus, it is found that are made up ntuples which contain concepts relations between concepts, n-tuples can be expressed form UP-graph (The graph understanding problem), semantic expression problem.UP-graph base generated by using knowledge (PKG).However, current hard to used because cannot store methods for...

The abstract is based on the project of "extended reality" for effective communication hazards from extreme coastal events, such as tsunamis and hurricanes. attends to use augmented reality (AR) mixed (MR) allow, example, a resident see digital tsunami crashing onshore bulldozing through community, all while standing their beach or in driveway. This type experience provides an emotional impact long-lasting memory that will guide future planning decisions proactivity. In this abstract, we...

Multi-hop machine reading comprehension is a challenging task in natural language processing as it requires more reasoning ability across multiple documents. Spectral models based on graph convolutional networks have shown good inferring abilities and lead to competitive results. However, the analysis of some are inconsistent with those humans. Inspired by concept grandmother cells cognitive neuroscience, we propose heterogeneous attention network model named ClueReader imitate cell concept....