Semantic Associations Research Articles

A knowledge graph-supported information fusion approach for multi-faceted conceptual modelling

It has become progressively more evident that a single data source is unable to comprehensively capture the variability of a multi-faceted concept, such as product design, driving behaviour or human trust, which has diverse semantic orientations. Therefore, multi-faceted conceptual modelling is often conducted based on multi-sourced data covering indispensable aspects, and information fusion is frequently applied to cope with the high dimensionality and data heterogeneity. The consideration of intra-facets relationships is also indispensable. In this context, a knowledge graph (KG), which can aggregate the relationships of multiple aspects by semantic associations, was exploited to facilitate the multi-faceted conceptual modelling based on heterogeneous and semantic-rich data. Firstly, rules of fault mechanism are extracted from the existing domain knowledge repository, and node attributes are extracted from multi-sourced data. Through abstraction and tokenisation of existing knowledge repository and concept-centric data, rules of fault mechanism were symbolised and integrated with the node attributes, which served as the entities for the concept-centric knowledge graph (CKG). Subsequently, the transformation of process data to a stack of temporal graphs was conducted under the CKG backbone. Lastly, the graph convolutional network (GCN) model was applied to extract temporal and attribute correlation features from the graphs, and a temporal convolution network (TCN) was built for conceptual modelling using these features. The effectiveness of the proposed approach and the close synergy between the KG-supported approach and multi-faceted conceptual modelling is demonstrated and substantiated in a case study using real-world data.

Hierarchical Feature Association and Global Correction Network for Change Detection

Optical satellite image change detection has attracted extensive research due to its comprehensive application in earth observation. Recently, deep learning (DL)-based methods have become dominant in change detection due to their outstanding performance. Remote sensing (RS) images contain different sizes of ground objects, so the information at different scales is crucial for change detection. However, the existing DL-based methods only employ summation or concatenation to aggregate several layers of features, lacking the semantic association of different layers. On the other hand, the UNet-like backbone is favored by deep learning algorithms, but the gradual downscaling and upscaling operation in the mainstream UNet-like backbone has the problem of misalignment, which further affects the accuracy of change detection. In this paper, we innovatively propose a hierarchical feature association and global correction network (HFA-GCN) for change detection. Specifically, a hierarchical feature association module is meticulously designed to model the correlation relationship among different scale features due to the redundant but complementary information among them. Moreover, a global correction module on Transformer is proposed to alleviate the feature misalignment in the UNet-like backbone, which, through feature reuse, extracts global information to reduce false alarms and missed alarms. Experiments were conducted on several publicly available databases, and the experimental results show the proposed method is superior to the existing state-of-the-art change detection models.

Exploiting Functional Discourse Grammar to Enhance Complex Arabic Relation Extraction using a Hybrid Semantic Knowledge Base - Machine Learning Approach

Relation extraction from unstructured Arabic text is especially challenging due to the Arabic language complex morphology and the variation in word semantics and lexical categories. The research documented in this paper presents a hybrid Semantic Knowledge base - Machine Learning (SKML) approach for extracting complex Arabic relations from unstructured Arabic documents; the proposed approach exploits the principles of Functional Discourse Grammar (FDG) to emphasise the semantic and pragmatic properties of the language and facilitate the identification of relation elements. At the initial phase, the novel FDG-SKML relation extraction approach deploys a lexical-based mechanism that utilises a purposely built domain-specific Semantic Knowledge to encode the semantic association between the identified relations’ elements. The evaluation of the initial stage evidenced improved accuracy for extracting most complex Arabic relations. The initial relation extraction mechanism was further extended by integrating its output into a Machine Learning classifier that facilitated extracting especially complex relations with significant disparity in the relation elements’ presence, order, and correlation. Using Economics as the problem domain, experimental evaluation evidenced the high accuracy of our FDG-SKML approach in complex Arabic relation extraction task and demonstrated its further improvement upon integration with machine learning classifiers.

Subjective Experience of Word Production Difficulties in Aphasia: a Metaphor Analysis of Autobiographical Accounts

ABSTRACT Background and aims The subjective experience of neurological symptoms provides useful information for assessment, intervention and care. However, research in the subjective experience of aphasia is limited. Word production difficulties are universal to aphasia, and interdisciplinary research has produced sophisticated models of the multiple stages and processes involved. Critically, this word-production research does not incorporate the subjective experience of symptoms. We carried out a metaphor-led discourse analysis on autobiographical accounts written by people with aphasia, to determine whether subjective descriptions of word finding difficulties are consistent with the stages and processes of psycholinguistic models. Method Metaphor-led discourse analysis was used to identify, code and interpret metaphorical expressions describing word production difficulties in 12 English-language autobiographical accounts written by people with aphasia. These expressions were then analysed to determine the systematic metaphors (i.e., the related concepts which are used consistently to describe a particular topic). Two distinct types of systematic metaphor emerged in the analysis: conventional systematic metaphors frequently recurring throughout all or most of the accounts; novel systematic metaphors used in one or two extended passages in an overlapping subset of the accounts. Results and discussion 4020 metaphorical expressions described word production, predominantly using conventional metaphors about communication and cognition. The conventional metaphor WORD-PRODUCTION AS MOVING OBJECTS OUT OF A CONTAINER was the most prevalent, with elaborations and variations allowing mapping of different symptoms. Other conventional metaphors included: WORD PRODUCTION AS A JOURNEY/HUNTING/HERDING THROUGH A LANDSCAPE, allowing description of effortful or partial retrieval, neuroplastic recovery, and internal strategies; APHASIA AS BODILY IMPAIRMENT, which described various symptoms in terms of different body parts, including self-monitoring difficulties; and APHASIA AS FRAGMENTATION AND PERSONIFICATION OF SELF and SELF AS MACHINE/COMPUTER to describe a disrupted sense of agency and attention. Novel systematic metaphors were used to describe certain symptoms: APHASIA AS SILENCE and APHASIA AS SPIRITUAL EXPERIENCE were used to describe a lack of ‘inner speech’, and APHASIA AS A DISMEMBERED TREE to describe problems making semantic associations. Conclusions and implications This research demonstrates the many consistencies of subjective descriptions of word production difficulties in aphasia with theoretical models, but also shows that some subjectively salient symptoms, in particular attentional and self-monitoring difficulties, and a lack of inner speech, are not captured by all theoretical models. Careful attention to the way that people with aphasia describe their symptoms can provide a valuable source of information to be integrated with objective measures.

Probabilistic Semantic Data Association for Collaborative Human-Robot Sensing

Humans cannot always be treated as oracles for collaborative sensing. Robots thus need to maintain beliefs over unknown world states when receiving semantic data from humans, as well as account for possible discrepancies between human-provided data and these beliefs. To this end, this paper introduces the problem of semantic data association (SDA) in relation to conventional data association problems for sensor fusion. It then develops a novel probabilistic semantic data association (PSDA) algorithm to rigorously address SDA in general settings, unlike previous work on semantic data fusion which developed heuristic techniques for specific settings. PSDA is further incorporated into a recursive hybrid Bayesian data fusion scheme which uses Gaussian mixture priors for object states and softmax functions for semantic human sensor data likelihoods. Simulations of a multi-object search task show that PSDA enables robust collaborative state estimation under a wide range of conditions where semantic human sensor data can be erroneous or contain significant reference ambiguities.

Attention-Guided Pyramid Context Networks for Detecting Infrared Small Target Under Complex Background

Infrared small target detection techniques remain a challenging task due to the complex background. To overcome this problem, by exploring context information, this research presents a data-driven approach called Attention-Guided Pyramid Context Network (AGPCNet). Specifically, we design Attention-Guided Context Block (AGCB) and perceive pixel correlations within and between patches at specific scales via Local Semantic Association (LSA) and Global Context Attention (GCA) respectively. Then the contextual information from multiple scales is fused by Context Pyramid Module (CPM) to achieve better feature representation. In the upsampling stage, we fuse the low and deep semantics through Asymmetric Fusion Module (AFM) to retain more information about small targets. The experimental results illustrate that AGPCNet has achieved state-of-the-art performance on three available infrared small target datasets. The source codes are available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Tianfang-Zhang/AGPCNet</uri> .

The use of lexical semantics for processing face-masked speech in monolinguals and bilinguals.

Face masks impede visual and acoustic cues that help make speech processing and language comprehension more efficient. Many studies report this phenomenon, but few examined how listeners utilize semantic information to overcome the challenges posed by face masks. Fewer still investigated this impact on bilinguals' processing of face-masked speech [Smiljanic, Keerstock, Meemann, and Ransom, S. M. (2021). J. Acoust. Soc. Am. 149(6), 4013-4023; Truong, Beck, and Weber (2021). J. Acoust. Soc. Am. 149(1), 142-144]. Therefore, this study aims to determine how monolingual and bilingual listeners use semantic information to compensate for the loss of visual and acoustic information when the speaker is wearing a mask. A lexical priming experiment tested how monolingual listeners and early-acquiring simultaneous bilingual listeners responded to video of English word pairs. The prime-target pairs were either strongly related, weakly related, or unrelated and were both either masked or unmasked. Analyses of reaction time results showed an overall effect of masking in both groups and an effect of semantic association strength on processing masked and unmasked speech. However, speaker groups were not different; subsequent analyses of difference values showed no effect of semantic context. These results illustrate the limited role of word-level semantic information on processing in adverse listening conditions. Results are discussed in light of semantic processing at the sentence level.

Feature Fusion Based on Transformer for Cross-modal Retrieval

With the popularity of the Internet and the rapid growth of multimodal data, multimodal retrieval has gradually become a hot area of research. As one of the important branches of multimodal retrieval, image-text retrieval aims to design a model to learn and align two modal data, image and text, in order to build a bridge of semantic association between the two heterogeneous data, so as to achieve unified alignment and retrieval. The current mainstream image-text cross-modal retrieval approaches have made good progress by designing a deep learning-based model to find potential associations between different modal data. In this paper, we design a transformer-based feature fusion network to fuse the information of two modalities in the feature extraction process, which can enrich the semantic connection between the modalities. Meanwhile, we conduct experiments on the benchmark dataset Flickr30k and get competitive results, where recall at 10 achieves 96.2% accuracy in image-to-text retrieval.

WHGDroid: Effective android malware detection based on weighted heterogeneous graph

The growing Android malware is seriously threatening the privacy and property security of Android users. However, the existing detection methods are often unable to maintain sustainability as Android malwares evolve. To address this issue, instead of directly using the intra-App feature, we exploit diverse inter-App relations to build a higher-level semantic association, making it more difficult for malware to evade detection. In this paper, we propose WHGDroid, a new malware detection framework based on weighted heterogeneous graph, which helps detect malware by implicit higher-level semantic connectivity across Apps. To comprehensively analyze Apps, we first extract five different Android entities and five relations, and then model the entities and relations among them into a weighted heterogeneous graph (WHG), in which weights are used to represent the importance of entities. Rich-semantic metapaths are proposed to establish the implicit associations between App nodes and derive homogeneous graphs containing only App nodes. Finally, graph neural network is used to learn the numerical embedding representations of Apps. We make a comprehensive comparison with five baseline methods on large datasets in different read scenarios. The experimental results show that WHGDroid is superior to two state-of-the-art methods in all cases.

A patent recommendation method based on KG representation learning

The number of patents is constantly increasing, providing an invaluable knowledge base. But retrieving this knowledge becomes ever more challenging with the increase in data. In this context, this study proposes a patent recommendation method based on knowledge graph (KG) representation. Although KG-based recommendation systems received broad attention in the literature, there is still a lack of literature that explores how it can be used for patent recommendations. First, and in terms of KG construction, the existing methods encounter issues when it comes to simultaneously accounting for accuracy and recall rate. This study proposes a combination of rule-based and deep learning methods for KG construction to solve this problem. Second, existing methods have performance and semantic association limitations when learning KG representations. This study therefore uses high-performance graph neural networks for encoding and models the semantic association limitations in graphs using traditional KG representation learning methods. Finally, existing recommendation systems usually focus on the semantic correlation between interaction items in recommendations, ignoring their direct relationships. This study therefore applies the KG representation learning method for the patent recommendation task, using KG representation learning to model the semantic association and constraints in the product–patent heterogeneous graph. Experiments based on a practical case example demonstrate that the proposed method outperforms baseline methods, and that it can guide innovative design processes. Findings have important implications for research and practice.

Otherness and suspiciousness: a comparative study of public opinions between the Confucius Institute and Goethe-Institut in developing countries

This study aims to examine the differences in lexical priming features between Confucius Institutes and Goethe-Instituts in developing countries using lexical priming theory and natural language processing techniques. By collecting news media coverage from 2014 to 2023, this study analyses the corpus through collocation, colligation, semantic association, and semantic prosody. In this study, it was found that the Goethe-Institut has a more stable institutional identity, while the Confucius Institute is still largely recognized for language teaching and cultural dissemination activities. The association of the Confucius Institute with China and its government creates a stronger sense of "otherness" and leads to negative perceptions. Overall, this study contributes to a better understanding of public perception and institutional image in developing countries.

Experiencer-Driven and Knowledge-Aware Graph Model for emotion–cause pair extraction

Numerous approaches have been explored to learn relations between emotion and cause clauses as a key step in the extraction of emotion–cause pairs. Despite their effectiveness, there are some limitations of previous studies: (1) they ignored that the emotion experiencer is an important clue implicating causality between clauses and (2) ignored that causal commonsense as prior knowledge can enhance semantic associations between clauses. In this paper, we propose a novel Experiencer-Driven and Knowledge-Aware Graph Model (EDKA-GM). For the first limitation, we introduce an experiencer identification task and present a document-level heterogeneous graph network for capturing global experiencer information to enrich experiencer-based cross-clause association. For the second, we retrieve the causal commonsense from the ATOMIC knowledge base for each clause and establish a knowledge-aware graph network to further enhance the inter-clause relationship by modeling a full connection graph of clauses with commonsense knowledge. We are the first to explore the influence of experiencer on emotion–cause pair extraction. On a benchmark dataset, our approach performs better than competitive baselines, achieving new state-of-the-art performance.

Posverdad y fake news en revistas científicas de comunicación de Iberoamérica

El artículo presenta un análisis de contenido con enfoque bibliométrico de la posverdad y las noticias falsas para determinar su presencia temática en las revistas científicas de comunicación de Iberoamérica. Se analizan los aportes teóricos, metodológicos y conceptuales de la posverdad y las noticias falsas con una muestra de N = 46 artículos publicados en 17 revistas, especializadas en comunicación (WoS, ESCI y Scopus) entre 2000 y 2019. El país con mayor publicación sobre posverdad y noticias falsas es España, el enfoque de investigación predominante es cuantitativo, estableciendo asociaciones semánticas con el ecosistema mediático, tecnológico e informacional y la desinformación.

Learning graph structures with transformer for weakly supervised semantic segmentation

Weakly supervised semantic segmentation (WSSS) is a challenging task of computer vision. The state-of-the-art semantic segmentation methods are usually based on the convolutional neural network (CNN), which mainly have the drawbacks of inability to explore the global information correctly and failure to activate potential object regions. To avoid such drawbacks, the transformer approach is explored in the WSSS task, but no effective semantic association between different patch tokens can be determined in the transformer. To address this issue, inspired by the graph convolutional network (GCN), this paper proposes a graph structure to learn the semantic category relationships between different blocks in the vector sequence. To verify the effectiveness of the proposed method in this paper, a large number of experiments were conducted on the publicly available PASCAL VOC2012 dataset. The experimental results show that our proposed method achieves significant performance improvement in the WSSS task and outperforms other state-of-the-art transformer-based methods.

Processing of linguistic focus depends on contrastive alternatives

Readers progressed through a sentence in the Maze task (Forster et al., 2009), deciding at each word between a sensical and a non-sensical continuation. Contexts presented before these sentences manipulated whether words were linguistically focused and whether they were given or new (Experiment 1); focused targets were read more slowly even when they were given, and new targets were read slowly in general. This both replicated earlier results in which slowdowns were found in the reading of focus (Benatar and Clifton, 2014; Birch and Rayner, 1997; Lowder and Gordon, 2015), and demonstrated that focus slowdowns are not reducible to newness. To clarify earlier results in which speed-ups were found on focused words (Birch and Rayner, 2010; Morris and Folk, 1998), contexts manipulated whether contrastive alternatives to focused words were presented with a focus particle (Experiment 2) or in a cleft construction (Experiment 3). Focused targets were read less slowly when a contrastive alternative was present in the context. This effect of contrastive alternatives cannot be reduced to simple semantic associate priming: Contexts also manipulated whether a semantically associated expression was present independently of the presence of a contrastive alternative (Experiment 4). Readers slowed down less when an alternative was present in the context, even when this alternative was not semantically associated to the target. These results indicate that the processing of focus depends on contrastive alternatives, in their interaction with newness, semantic association, and focus construction.

Disambiguation of medical abbreviations for knowledge organization

The disambiguation of abbreviations is a crucial step in medical knowledge organization. In the past, most scholars have focused on the problem of disambiguating medical abbreviations in single sentences; they have not systematically considered full-article abbreviation disambiguation tasks. In this work, we present a research framework for full-article medical abbreviation disambiguation (FMADRF) based on the structural characteristics of abbreviation–definition pairs in a full scientific medical article. Our method utilizes the information including context semantic information, external linguistic features, and the mapping relationships and structural similarities between abbreviations and their expansions. The model includes a four-pronged approach, identification of abbreviations and abbreviation–definition pairs, alignment and complementation of abbreviations and abbreviation expansions. The results show that our novel BBF-BLC-R model improves the recognition and modification effects of abbreviation–definition pairs, achieving the best F1 score of 91.83%. Furthermore, our new strategy combines semantic and structural information to significantly improve the effects of term alignment, with an F1 score of 97.11%. In our test, a thesaurus of abbreviations and their expansions was constructed from 13,472 full-text medical articles, resulting in 14,742 abbreviations, with 31,327 corresponding expansions. This work enhances the semantic association of terms in full medical texts, eliminating the problems of “rich” semantics and association–relation roadblocks caused by term misalignments. It further provides technical and methodological support for the organization of medical knowledge, facilitating the deep knowledge-mining capabilities of full-text medical articles.

Heart Rate Variability Biofeedback for Mild Traumatic Brain Injury: A Randomized-Controlled Study

To determine whether heart rate variability biofeedback (HRV-BF) training, compared to a psychoeducation control condition can strengthen the integration of the central and autonomic nervous systems as measured by neuropsychological measures in patients with mild traumatic brain injury (mTBI). Participants were recruited from two university hospitals in Taipei, Taiwan. A total of 49 participants with mTBI were recruited for this study. Forty-one participants completed the study, 21 in the psychoeducation group and 20 in the HRV-BF group. Randomized controlled study. The Taiwanese Frontal Assessment Battery, the Semantic Association of Verbal Fluency Test, the Taiwanese version of the Word Sequence Learning Test, the Paced Auditory Serial Addition Test-Revised, and the Trail Making Test were used as performance-based neuropsychological functioning measures. The Checklist of Post-concussion Symptoms, the Taiwanese version of the Dysexecutive Questionnaire, the Beck Anxiety Inventory, the Beck Depression Inventory, and the National Taiwan University Irritability Scale were used as self-report neuropsychological functioning measures. Furthermore, heart rate variability pre- vs. post-training was used to measure autonomic nervous system functioning. Executive, information processing, verbal memory, emotional neuropsychological functioning, and heart rate variability (HRV) were improved significantly in the HRV-BF group at the posttest whereas the psychoeducation group showed no change. HRV biofeedback is a feasible technique following mild TBI that can improve neuropsychological and autonomic nervous system functioning. HRV-BF may be clinically feasible for the rehabilitation of patients with mTBI.

Multimodal Fine-Grained Transformer Model for Pest Recognition

Deep learning has shown great potential in smart agriculture, especially in the field of pest recognition. However, existing methods require large datasets and do not exploit the semantic associations between multimodal data. To address these problems, this paper proposes a multimodal fine-grained transformer (MMFGT) model, a novel pest recognition method that improves three aspects of transformer architecture to meet the needs of few-shot pest recognition. On the one hand, the MMFGT uses self-supervised learning to extend the transformer structure to extract target features using contrastive learning to reduce the reliance on data volume. On the other hand, fine-grained recognition is integrated into the MMFGT to focus attention on finely differentiated areas of pest images to improve recognition accuracy. In addition, the MMFGT further improves the performance in pest recognition by using the joint multimodal information from the pest’s image and natural language description. Extensive experimental results demonstrate that the MMFGT obtains more competitive results compared to other excellent models, such as ResNet, ViT, SwinT, DINO, and EsViT, in pest recognition tasks, with recognition accuracy up to 98.12% and achieving 5.92% higher accuracy compared to the state-of-the-art DINO method for the baseline.

Multi-feature fused collaborative attention network for sequential recommendation with semantic-enriched contrastive learning

Recently, graph neural network (GNN) has been widely used in sequential recommendation because of its powerful ability to capture high-order collaborative relations, greatly promoting recommendation performance. However, some existing GNN-based methods fail to make full use of multiple relevant features of nodes and ignore the impact of semantic association between nodes on extracting user preferences. To this end, we propose a multi-feature fused collaborative attention network MASR, which sufficiently learns the temporal and positional features of nodes, and innovatively measures the importance of these two features for analyzing the nodes’ dynamic patterns. In addition, we incorporate semantic-enriched contrastive learning into collaborative filtering to enhance the semantic association between nodes and reduce the noise from the structural neighborhood, which has a positive effect on the sequential recommendation. Compared with the baseline models, the performance of MASR on MovieLens, CDs and Beauty datasets is improved by 2.0%, 2.1% and 1.7% respectively, proving its effectiveness in the sequential recommendation.

ICAD-MI: Interdisciplinary concept association discovery from the perspective of metaphor interpretation

Interdisciplinary concept association discovery is a fundamental task in interdisciplinary knowledge organization. Unlike general concept association, interdisciplinary concept association mainly manifests in the correlation between fine-grained concept properties, which requires that interdisciplinary concept association discovery be explored through a fine-grained semantic association discovery tool. Existing concept association discovery methods are limited in their ability to identify interdisciplinary concept associations at fine-grained conceptual properties because they can only identify which two concepts are associated at the coarse level. To bridge this gap, we propose a method we called interdisciplinary concept association discovery based on metaphor interpretation (ICAD-MI). First, we explored the mechanism of interdisciplinary conceptual metaphor on both the cognitive and language layers, which provides a solid foundation for our method. Second, we introduced the four-step ICAD-MI method, which integrates deep learning techniques with word semantics and multidimensional contexts. We tested the ICAD-MI framework using a dataset comprising a total of 1,915 data points of interdisciplinary metaphorical expressions (IMEs) on a typical interdisciplinary conceptual metaphor Computer is a brain. Our model achieved a precision of 94.4%, a recall of 73.9%, and an F1 score of 82.9%, which outperforms the four baseline methods. Additionally, we conducted parameter analysis to further validate the effectiveness of our proposed method. The code and datasets are publicly available at: https://github.com/haihua0913/ICADMI.