Multimodal knowledge graphs (MMKGs) integrate structural, visual, and textual modalities to enhance entity and relation representations. However, existing MMKGs completion methods often rely on static fusion strategies that overlook context-specific modality relevance, and they tend to underutilize structural information encoded in the graph topology. In this paper, we present SDMF-MKG, a structure-aware dynamic fusion framework designed to address modality bias and structural underrepresentation in MMKGs. The model incorporates three key components: a structure-guided semantic encoder that preserves topological signals, a dynamic weighting mechanism that adaptively calibrates modality contributions based on triple context, and a KL-regularized loss to encourage balanced modality utilization. We evaluate SDMF-MKG on four benchmark datasets spanning both multimodal-rich and structure-only settings. The model achieves state-of-the-art or competitive performance across most metrics, with notable gains on multimodal datasets such as VTKG-C. Ablation studies further confirm the complementary effects of structure awareness, adaptive fusion, and modality balancing.

Trade shows are key venues for presenting new products and research, yet physical exhibits can be costly and impractical. This study evaluates whether digital exhibitions utilizing VR and AR can effectively replace physical trade show presentations. Across three real-world experiments conducted at major international fairs, we deployed different digital exhibition formats and recorded visitor numbers, engagement ratios, and interaction durations. All exhibitions attracted substantially more visitors than the fairs’ average exhibitor-to-visitor ratios, exceeding them by 30 to 80%. In the first experiment, surveys captured user experience and success factors, while exhibitor interviews across all experiments provided insights into perceived impact. Results indicate that visitor interaction with digital content was the primary driver of engagement. By evaluating VR headsets, AR glasses, holographic projections, and mobile AR, we identified their respective strengths and limitations. Results indicate that well-designed digital exhibitions can serve as effective, cost-efficient alternatives to physical trade show presentations.

This study proposes a dual-view hyper-relational knowledge graph embedding model aimed at addressing the challenges of embedding complex relationships in knowledge graphs. Traditional methods primarily handle simple triplet relations and struggle with the complexity of hyper-relations. By integrating instance view and ontology view, our model, DVHE, captures hierarchical structural information between entities and is applied to link prediction tasks. Experimental results show that DVHE significantly outperforms existing single-view and dual-view models across multiple benchmark datasets, particularly in handling complex hyper-relations and hierarchical information. Ablation studies further validate the effectiveness of the model’s components, providing new insights for the development of knowledge graph embeddings.

Existing conversational recommender systems (CRS) use insufficient generality in incorporating external information using knowledge graphs. The recommendation module and generation module are loosely connected during model training and shallowly integrated during inference. A simple switching or copying mechanism is used to merge recommended items into generated responses. These problems significantly degrade the recommendation performance. To alleviate this problem, we propose a novel unified framework for collaboratively enhancing conversational recommendations using pre-trained language mod-els and knowledge graphs (CollRec). We use a fine-tuned pre-trained language model to efficiently extract knowledge graphs from conversational text descriptions, perform entity-based recommendations based on the generated graph nodes and edges, and fine-tune a large-scale pre-trained language model to generate fluent and diverse responses. Experimental results on the WebNLG 2020 Challenge dataset, ReDial dataset, and Reddit-Movie dataset show that our CollRec model significantly outperforms the state-of-the-art methods.

Knowledge graph embedding maps the semantics of entities and relations to a low-dimensional space by optimizing the vector distance between positive and negative triples. Traditional negative sampling techniques usually regard high-scoring triples as high-quality negative triples, but this not only easily introduces false negative triples, but also ignores important information in the graph structure. To address these issues, we propose an easily pluggable hierarchical knowledge graph embedding framework. High-quality corrupted entities are generated through semantic and structural information, and then margin estimation is used to generate high-quality negative triples, and the structural information of the entities is combined to perform link prediction on new facts. Experimental results show that our framework improves the performance of the original knowledge graph embedding model, in which the hierarchical subgraph negative sampling module outperforms other negative sampling techniques. The framework we proposed can be easily adapted to various knowledge graph embedding models and explain the prediction results.

Continual relation extraction (CRE) aims to extract relations towards the continuous and iterative arrival of new data. To address the problem of catastrophic forgetting, some existing research endeavors have focused on exploring memory replay methods by storing typical historical learned instances or embedding all observed relations as prototypes by averaging the hidden representation of samples and replaying them in the subsequent training process. However, this prototype generation method overlooks the rich semantic information within the label namespace and are also constrained by memory size, resulting in inadequate descriptions of relation semantics by relation prototypes. To this end, we introduce an approach termed Label-Guided Relation Prototype Generation. Initially, we enhance the representations of label embeddings through a technique named label knowledge infusion. Following that, we utilize the multi-head attention mechanism to form relation prototypes, allowing them to capture diverse aspects of typical instances. The embeddings of relation labels are utilized at this stage, leveraging their contained semantics. Additionally, we propose a feature-based distillation loss function called multi-similarity distillation, to ensure the model retains prior knowledge after learning new tasks. The experimental results indicate that our method has achieved competitive performance compared to the state-of-the-art baseline models in CRE.

The hyper-relation knowledge graph constructed by hyper-relation facts can effectively express most of the information in the real world. Similar to traditional knowledge graphs, it also has more or less missing information, which makes link prediction on the hyper-relation knowledge graph equally important.However, due to the extension from binary relations to n-ary relations, the hyper-relation knowledge graph has both graph representations based on hypergraphs and sequential representations based on semantics.Most existing models mainly focus on handling only one type of representation within the internal structure of hyper-relations, which limits the comprehensive learning of hyper-relation knowledge graphs.Moreover, there is a large amount of neighbor information in hyper-relation knowledge graphs, but most existing models simply feed the sequential representations into Transformer layers, which greatly reduces the model’s ability to handle neighbors.Therefore, based on the two aforementioned issues, this paper presents a Hyper-Relation Knowledge Graph Completion model that integrates neighbor information(HKNI). HKNI not only effectively handles the two types of representations within the hyper-relation, but also efficiently processes neighbor information.In the end, we compared our model with existing baseline models and experimental results demonstrated that our model achieves the best performance in link prediction tasks.

Procedure for early forecasting of apple yields in Slovenian orchards is presented with the help of a modern application that works in the Android environment and uses a fast 5G transmission network. To recognize fruits in images of ‘Golden delicious’ trees, we used the YOLO pre-learning network, which is based on convolutional neural networks and regression techniques to determine the position of apples in the image. To model the fruit yield, specific cultivar-adjusted growth curve is used, which makes it possible to predict fruit mass from the current and expected fruit diameter and the ratio between diameter and mass. The procedure was tested on a series of 20 images captured from six-years-old orchard revealing on average 71.84% accuracy in fruits counting, 108.27% accuracy in diameter calculating and 97.90% accuracy in yield forecasting. With the first estimation, we have shown that the automated method for yield forecasting is an excellent tool for accurately estimating the yield of an individual plot so we will continue to upgrade it in the future.

This paper proposes a new string transformation technique called Move with Interleaving (MwI). Four possible ways of rearranging 2D raster images into 1D sequences of values are applied, including scan-line, left-right, strip-based, and Hilbert arrangements. Experiments on 32 benchmark greyscale raster images of various resolutions demonstrated that the proposed transformation reduces information entropy to a similar extent as the combination of the Burrows-Wheeler transform followed by the Move-To-Front or the Inversion Frequencies. The proposed transformation MwI yields the best result among all the considered transformations when the Hilbert arrangement is applied.

Knowledge Graphs (KGs) describe objective facts in the form of RDF triples, each triple contains sufficient semantic information and triple structure information. Knowledge Graph Completion (KGC) is to acquire new knowledge by predicting hidden relationships between entities and adding the new knowledge to the KG. At present, the mainstream KGC approaches only applied the triple structure information or only utilized the semantic information of the text. This paper proposes an approach (TSTR) using BERT and deep neural networks to fully extract the semantic information of knowledge, and designs an aggregated re-ranking scheme that incorporates existing graph embedding approach to learn the structural information of triples. In experiments, the approach achieves state-of-the-art performance on three benchmark datasets, and outperforms recent KGC approaches on sparsely connected datasets.

With the rise of deep learning technology, the field of medical image segmentation has undergone rapid development. In recent years, convolutional neural networks (CNNs) have brought many achievements and become the consensus in medical image segmentation tasks. Although many neural networks based on U-shaped structures and methods, such as skip connections have achieved excellent results in medical image segmentation tasks, the properties of convolutional operations limit their ability to effectively learn local and global features. To address this problem, the Transformer from the field of natural language processing (NLP) was introduced to the image segmentation field. Various Transformer-based networks have shown significant performance advantages over mainstream neural networks in different visual tasks, demonstrating the huge potential of Transformers in the field of image segmentation. However, Transformers were originally designed for NLP and ignore the multidimensional nature of images. In the process of operation, they may destroy the 2D structure of the image and cannot effectively capture low-level features. Therefore, we propose a new multi-scale cross-attention method called M-VAN Unet, which is designed based on the Visual Attention Network (VAN) and can effectively learn local and global features. We propose two attention mechanisms, namely MSC-Attention and LKA-Cross-Attention, for capturing low-level features and promoting global information interaction. MSC-Attention is designed for multi-scale channel attention, while LKA-Cross-Attention is a cross-attention mechanism based on the large kernel attention (LKA). Extensive experiments show that our method outperforms current mainstream methods in evaluation metrics such as Dice coefficient and Hausdorff 95 coefficient.

The specific task of vehicle re-identification is how to quickly and correctly match the same vehicle in different scenarios. In order to solve the problem of inter-class similarity and environmental interference in vehicle images in complex scenes, one fusion attention method is put forward based on the idea of obtaining the distinguishing features of details—the mechanism for the vehicle re-identification method. First, the vehicle image is preprocessed to restore the image’s attributes better. Then, the processed image is sent to ResNet50 to extract the features of the second and third layers, respectively. Then, the feature fusion is carried out through the two-layer attention mechanism for a network model. This model can better focus on local detail features, and global features are constructed and named SDLAU-Reid. In the training process, a data augmentation strategy of random erasure is adopted to improve the robustness. The experimental results show that the mAP and rank-k indicators of the model on VeRi-776 and the VehicleID are better than the results of the existing vehicle re-identification algorithms, which verifies the algorithm’s effectiveness.

Reliable forest growth forecasting requires detailed tree data for forest simulation, while manual on-site collection of relevant data is work-intensive and unfeasible in larger forests. This paper proposes a complete methodology for fully automated forest growth simulation that relies primarily on airborne topographic Light Detection And Ranging (LiDAR) point clouds of individual trees. The proposed method estimates tree parameters and performs growth of individual trees based on an individual-based forest growth simulator, named BWINPro. In addition, competition and detailed asymmetric tree crown growth are modeled regarding the shading of tree crowns, which is estimated from the surrounding environment and neighbor trees. The result of the proposed approach is a new point cloud for subsequent analyses. The proposed method was validated by comparing canopy height models derived from the point clouds of the simulated trees with canopy height models derived from more recent ground truth point clouds. The results demonstrate the efficacy of the proposed method which achieves a 9.4% higher accuracy than the averaged linear regression model and, in the case of datasets with more distinct self-standing trees, where a tree crown boundary plays major role, a 4.1% higher accuracy than the directly fitted linear regression model.

With rapid development of the Internet, people have undergone tremendous changes in the way they obtain information. In recent years, knowledge graph is becoming a popular tool for the public to acquire knowledge. For knowledge graph of Chinese history and culture, most researchers adopted traditional named entity recognition methods to extract entity information from unstructured historical text data. However, the traditional named entity recognition method has certain defects, and it is easy to ignore the association between entities. To extract entities from a large amount of historical and cultural information more accurately and efficiently, this paper proposes one named entity recognition model combining Bidirectional Encoder Representations from Transformers and Bidirectional Long Short-Term Memory-Conditional Random Field (BERT-BiLSTM-CRF). First, a BERT pre-trained language model is used to encode a single character to obtain a vector representation corresponding to each character. Then one Bidirectional Long Short-Term Memory (BiLSTM) layer is applied to semantically encode the input text. Finally, the label with the highest probability is output through the Conditional Random Field (CRF) layer to obtain each character’s category. This model uses the Bidirectional Encoder Representations from Transformers (BERT) pre-trained language model to replace the static word vectors trained in the traditional way. In comparison, the BERT pre-trained language model can dynamically generate semantic vectors according to the context of words, which improves the representation ability of word vectors. The experimental results prove that the model proposed in this paper has achieved excellent results in the task of named entity recognition in the field of historical culture. Compared with the existing named entity identification methods, the precision rate, recall rate, and $$F_1$$ value have been significantly improved.

Tree pruning is a labor and cost-intensive task. Still, it is a necessary activity that ensures a high yield of good quality products in horticulture and increases the overall health of trees in general. Extensive research has been done attempting to automate this labor-intensive procedure, lower the cost, and demand a skilled workforce. We introduce a new algorithm based on discrete differential evolution that simulates the pruning of virtual trees. Although pruning driven by differential evolution alone optimizes the overall tree light intake, it cannot maintain the distance between individual trees, nor can it shape trees into any of the growing forms. In the article, we show that adding additional steps into the pruning process, which is an initial trimming of the tree into a desired shape, can be improved significantly. We demonstrate our method by simulating the pruning of virtual trees and show that it provides results comparable to the results obtained by a human expert. By simulating the tree pruning over a few consecutive years, We show that our method is also capable of autonomous tree training toward the desired growing form. (C) 2020 The Author(s). Published by Elsevier B.V.

This paper presents the first complete approach to achieving environmental intelligence support in the management of vegetation within electrical power transmission corridors. Contrary to the related studies that focused on the mapping of power lines, together with encroaching vegetation risk assessment, we realised predictive analytics with vegetation growth simulation. This was achieved by following the JDL/DFIG data fusion model for complementary feature extraction from Light Detection and Ranging (LiDAR) derived data products and auxiliary thematic maps that feed an ensemble regression model. The results indicate that improved vegetation growth prediction accuracy is obtained by segmenting training samples according to their contextual similarities that relate to their ecological niches. Furthermore, efficient situation assessment was then performed using a rasterised parametrically defined funnel-shaped volumetric filter. In this way, RMSE≈1 m was measured when considering tree growth simulation, while a 0.37 m error was estimated in encroaching vegetation detection, demonstrating significant improvements over the field observations.

In recent years, Question Answering System has become a main focus of human machine interaction. Using the question answering system for information retrieval is convenient and efficient. Traditional question answering systems mostly use template matching. The question and answer data sets usually rely on manual design. The question and answer system implemented by this method has a quick query response and can answer relatively complex questions. But manually defining templates and rules is time-consuming and laborious. Therefore, this paper introduces the knowledge graph in the process of constructing the question answering system. The proposed method combines knowledge graph with question answering systems to realize an intelligent question answering system in the field of ethnic minorities, so as to spread ethnic minority knowledge and promote ethnic minority culture. The construction of a question answering system is mainly divided into two modules: question analysis and answer generation. The question analysis module includes named entity recognition, similarity calculation and question classification. The answer generation module includes entity mapping and graph retrieval. The experimental results show that our approach by combined with natural language processing technology and domain knowledge graph can well accurately feedback the answers to user queries. And the answer accuracy of the question answering system can reach about 81%.

The domestic population has paid increasing attention to ancient Chinese history and culture with the continuous improvement of people's living standards, the rapid economic growth, and the rapid advancement of information science and technology. The use of information technology has been proven to promote the spread and development of historical culture, and it is becoming a necessary means to promote our traditional culture. This paper will build a knowledge graph of ancient Chinese history and culture in order to facilitate the public to more quickly and accurately understand the relevant knowledge of ancient Chinese history and culture. The construction process is as follows: firstly, use crawler technology to obtain text and table data related to ancient history and culture on Baidu Encyclopedia (similar to Wikipedia) and ancient Chinese history and culture related pages. Among them, the crawler technology crawls the semi-structured data in the information box (InfoBox) in the Baidu Encyclopedia to directly construct the triples required for the knowledge graph, crawls the introductory text information of the entries in Baidu Encyclopedia, and specialized historical and cultural websites (history Chunqiu.com, On History.com) to extract unstructured entities and relationships. Secondly, entity recognition and relationship extraction are performed on an unstructured text. The entity recognition part uses the Bidirectional Long Short-Term Memory-Convolutional Neural Networks-Conditions Random Field (BiLSTM-CNN-CRF) model for entity extraction. The relationship extraction between entities is performed by using the open source tool DeepKE (information extraction tool with language recognition ability developed by Zhejiang University) to extract the relationships between entities. After obtaining the entity and the relationship between the entities, supplement it with the triple data that were constructed from the semi-structured data in the existing knowledge base and Baidu Encyclopedia information box. Subsequently, the ontology construction and the quality evaluation of the entire constructed knowledge graph are performed to form the final knowledge graph of ancient Chinese history and culture.

We present a framework for multi-objective optimization of fruit tree pruning within a simulated environment, where pruning is performed on a virtual tree model, and its effects on tree growth are observed. The proposed framework uses quantitative measures to express the short-term and long-term effects of pruning, for which potentially conflicting optimization objectives can be defined. The short-term objectives are evaluated on the pruned tree model directly, while the values of long-term objectives are estimated by executing a tree growth simulation. We demonstrate the concept by using a bi-objective case, where the estimated light interceptions of the pruned tree in the current and the next season are used to define separate optimization objectives. We compare the performance of the multi-objective simulated annealing and the NSGA-II method in building the sets of non-dominated pruning solutions. The obtained Pareto front approximations correspond to diverse pruning solutions that balance between optimizing either objective to different extents, which indicates a potential for new applications of the multi-objective pruning optimization concept.

Navidezno in obogateno resničnost lahko uporabimo tudi na področju predstavitev novih izdelkov in idej širši javnosti. Z njima lahko uporabnike in potencialne kupce seznanimo z izdelkom že v času njegovega razvoja. V članku opisujemo naše izkušnje s predstavitvijo rezultatov projekta MAHEPA. Gre za evropski projekt v okviru programa Obzorje 2020, znotraj katerega se razvijata dva hibridna pogona za letala prihodnosti. Ker so leti s tema pogonoma predvideni šele v letu 2020, je bilo možno napredek v razvoju obeh pogonskih sistemov letal na letalskem sejmu AERO 2019 predstaviti zgolj s pomočjo virtualnih modelov. V članku predstavljamo naše izkušnje s pripravo predstavitve, temelječe na navidezni in obogateni resničnosti, ter pokažemo, da so takšne predstavitve konkurenčne klasičnim in predstavljajo cenovno ugodnejšo alternativo. Predstavitev smo razdelili v tri sklope, pri čemer smo uporabljali očala HoloLens, čelado HTC Vive, s kodami QR pa smo krmilili tudi predstavitev na mobilnih napravah. Vse tri predstavitve so bile izdelane z igralnim pogonom Unity.