Semantic Nodes Research Articles

Relation-aware heterogeneous graph neural network for entity alignment

Entity alignment refers to finding equivalent entities from different knowledge graphs. Most of the existing entity alignment methods are studied based on homogeneous graphs. However, the knowledge graph is a heterogeneous graph containing many types of nodes, such as entities, relations, and attributes. Therefore, we propose introducing a heterogeneous graph neural network to model entities and relations simultaneously and propose an iterative fusion method to enhance the interaction between these two semantic nodes. Since not all datasets contain relation information, this paper does not directly introduce a feature representation of relations. The generalizability of the approach is improved by utilizing a relation-aware strategy to obtain information about the relation. Specifically, the information propagation of the head and tail entities in the triplet is utilized to obtain the feature representation of the relation. Experimental results show that the present method performs better on three cross-lingual datasets DBP15K and two large-scale datasets DWY100K.

Religion in modern projects: between alarmism, political expediency and utopia

The permanent shift of modernity, the epistemological crisis, and the politicization of all social spheres determine an atmosphere of global political anxiety. The ever-expanding security deficit becomes an attribute of modernity, as a fragmentation as a response to globality and the emergence of alternative ways of political genesis reflects new socio-cultural semantic nodes. Existential threats raise a persistent appeal to religion. The loss of epistemological certainty in political constructions contributes to the theoretical confusion about the future. The diffusion of political and value contexts forces people to turn to religion while developing projects and setting the goals of modernity. This creates cognitive and political- instrumental difficulties in using legitimizing and, at the same time, mutually exclusive religious meanings. The simultaneous truth in different types of discourses and in theory makes it impossible to find a common denominator. We argue that the projects and goal settings of modernity, claiming to be universal, but not achieving it, are structured primarily based on the logic of alarmism and eschatology. A positive project, trying to overcome the limitations of politicization, moves to the value plane. We consider three perspectives of the teleology of modernity, including religious meanings: theoretical, metaphysical, and instrumental. The research methodology is based on the problems of political goal setting in the logic of, traditional/ modern, explaining the religious coloring of political strategies and the splitting of future visions. The othering based on religious differences as a tool of political goal setting, the political instrumentalization of religious identities, simultaneously with religious creation of the world, puts two possible projects of modernity at opposite poles - the World of Religions and the World of Faith. Among the possible projects arising at the intersection of religion and politics, the author suggests three large clusters: alarmism, political expediency, and utopia.

Knowledge fusion method of power grid model based on Seq2seq half pointer and half label method

Due to the complexity of the calculation process of the existing methods, the efficiency of data fusion of the power grid model is low. In order to improve the knowledge fusion effect of power grid model, this paper studied the knowledge fusion method of power grid model based on Seq2seq half pointer and half label method. The Text Rank algorithm is used to calculate the weight of semantic nodes of each grid model, and combined with the topological potential method, the semantic information of the grid model is extracted according to the final weight value, and the Seq2Seq semi-pointer semi-label model framework is constructed. The data of the scheduling automation system OMS and the production management system PMS are used as input. The extracted candidate mesh model semantics and the original mesh model semantics are encoded by Seq2Seq half-pointer half-label model. The semantic data of the power grid model is fused and sent to the Seq2Seq encoder. After the training is completed, the effective information is extracted from the power grid model through the Seq2Seq model to complete the knowledge fusion of the power grid model. Experimental results show that this method eliminates the redundant part of the basic attributes of each data source in the substation grid model after knowledge fusion, and the description of each basic attribute is more standardized, unified and perfect. Under different mesh model data dimensions, the support of the proposed method is all above 98%. The model trained by the proposed method tends to be stable after 120 iterations, and the precision, recall and F1 of the test set are 0.98, 0.93 and 0.91, respectively. At the same time, this method has high efficiency in the knowledge fusion processing of the power grid model, and its data processing speed is less than 160 s. The average integrity of the private data of the power grid model is 98.86%, indicating that the proposed method can better ensure the integrity of the data. Finally, compared with the application of other methods under different data amounts, the mean square error obtained by the proposed method is the smallest, indicating that the proposed method effectively improves the fusion accuracy.

Multi-granularity adaptive extractive document summarization with heterogeneous graph neural networks.

The crucial aspect of extractive document summarization lies in understanding the interrelations between sentences. Documents inherently comprise a multitude of sentences, and sentence-level models frequently fail to consider the relationships between distantly-placed sentences, resulting in the omission of significant information in the summary. Moreover, information within documents tends to be distributed sparsely, challenging the efficacy of sentence-level models. In the realm of heterogeneous graph neural networks, it has been observed that semantic nodes with varying levels of granularity encapsulate distinct semantic connections. Initially, the incorporation of edge features into the computation of dynamic graph attention networks is performed to account for node relationships. Subsequently, given the multiplicity of topics in a document or a set of documents, a topic model is employed to extract topic-specific features and the probability distribution linking these topics with sentence nodes. Last but not least, the model defines nodes with different levels of granularity-ranging from documents and topics to sentences-and these various nodes necessitate different propagation widths and depths for capturing intricate relationships in the information being disseminated. Adaptive measures are taken to learn the importance and correlation between nodes of different granularities in terms of both width and depth. Experimental evidence from two benchmark datasets highlights the superior performance of the proposed model, as assessed by ROUGE metrics, in comparison to existing approaches, even in the absence of pre-trained language models. Additionally, an ablation study confirms the positive impact of each individual module on the model's ROUGE scores.

Efficient motion capture data recovery via relationship-aggregated graph network and temporal pattern reasoning.

Human motion capture (mocap) data is of crucial importance to the realistic character animation, and the missing optical marker problem caused by marker falling off or occlusions often limit its performance in real-world applications. Although great progress has been made in mocap data recovery, it is still a challenging task primarily due to the articulated complexity and long-term dependencies in movements. To tackle these concerns, this paper proposes an efficient mocap data recovery approach by using Relationship-aggregated Graph Network and Temporal Pattern Reasoning (RGN-TPR). The RGN is comprised of two tailored graph encoders, local graph encoder (LGE) and global graph encoder (GGE). By dividing the human skeletal structure into several parts, LGE encodes the high-level semantic node features and their semantic relationships in each local part, while the GGE aggregates the structural relationships between different parts for whole skeletal data representation. Further, TPR utilizes self-attention mechanism to exploit the intra-frame interactions, and employs temporal transformer to capture long-term dependencies, whereby the discriminative spatio-temporal features can be reasonably obtained for efficient motion recovery. Extensive experiments tested on public datasets qualitatively and quantitatively verify the superiorities of the proposed learning framework for mocap data recovery, and show its improved performance with the state-of-the-arts.

Loop Closure Detection for Mobile Robot based on Multidimensional Image Feature Fusion

Loop closure detection is a crucial part of VSLAM. However, the traditional loop closure detection algorithms are difficult to adapt to complex and changeable scenes. In this paper, we fuse Gist features, semantic features and appearance features of the image to detect the loop closures quickly and accurately. Firstly, we take advantage of the fast extraction speed of the Gist feature by using it to screen the loop closure candidate frames. Then, the current frame and the candidate frame are semantically segmented to obtain the mask blocks of various types of objects, and the semantic nodes are constructed to calculate the semantic similarity between them. Next, the appearance similarity between the images is calculated according to the shape of the mask blocks. Finally, based on Gist similarity, semantic similarity and appearance similarity, the image similarity calculation model can be built as the basis for loop closure detection. Experiments are carried out on both public and self-filmed datasets. The results show that our proposed algorithm can detect the loop closure in the scene quickly and accurately when the illumination, viewpoint and object change.

The Russian Language Personality in the Current Diachrony: a New Associative-Verbal Database (2014-2021)

The study presents a new Russian regional associative database (Siberia and the Far East) - SIBAS 2, obtained in a mass associative experiment (2014-2021) in the Asian regions of the Russian Federation. The author describes the main parameters of the scientifically significant information provided by the electronic resource SIBAS, the characteristics of the data in SIBAS 2 in comparison with the database SIBAS 1 reflecting the dimensions of the Russian language personality in the Asian regions of Russia, recorded in the period from 2008 to 2013. The associative dominants identified with the help of the SIBAS 1 resource are compared with the corresponding dimensions of the RLP, explicated by psycholinguists in the European part of the Russian Federation on the basis of the EURAS reverse dictionary (2008-2013); the dominants of SIBAS 1 and SIBAS 2 are compared with the all-Russian data of the reverse dictionaries of RAS (Russian Associative Thesaurus): the first (1988-1991) and second (1992-1995) stages of the all-Russian experiment. Against this background, the author presents new results of a general analysis of the semantic nodes of the SIBAS associative-verbal network, statistical ordering of the entire array of verbal associates in the current diachrony (2008-2021). Such statistical organization of the associativeverbal material can be viewed as a methodological technique, a preliminary step in the study of the psychologically relevant fluctuations of the system-forming semantic accentuations of the Russian language personality, their historical and cultural roots and meaningful transformations referring to a particular period of time (here mostly according to the data of SIBAS 1 and SIBAS 2).

Multi-granularity heterogeneous graph attention networks for extractive document summarization

Extractive document summarization is a fundamental task in natural language processing (NLP). Recently, several Graph Neural Networks (GNNs) are proposed for this task. However, most existing GNN-based models can neither effectively encode semantic nodes of multiple granularity level apart from sentences nor substantially capture different cross-sentence meta-paths. To address these issues, we propose MHgatSum, a novel Multi-granularity Heterogeneous Graph ATtention networks for extractive document SUMmarization. Specifically, we first build a multi-granularity heterogeneous graph (HetG) for each document, which is better to represent the semantic meaning of the document. The HetG contains not only sentence nodes but also multiple other granularity effective semantic units with different semantic levels, including keyphrases and topics. These additional nodes act as the intermediary between sentences to build the meta-paths involved in sentence node (i.e., Sentence-Keyphrase-Sentence and Sentence-Topic-Sentence). Then, we propose a heterogeneous graph attention networks to embed the constructed HetG for extractive summarization, which enjoys multi-granularity semantic representations. The model is based on a hierarchical attention mechanism, including node-level and semantic-level attentions. The node-level attention can learn the importance between a node and its meta-path based neighbors, while the semantic-level attention is able to learn the importance of different meta-paths. Moreover, to better integrate sentence global knowledge, we further incorporate sentence node global importance in local node-level attention. We conduct empirical experiments on two benchmark datasets, which demonstrates the superiority of MHgatSum over previous SOTA models on the task of extractive summarization.

ICGPN: Interaction-centric graph parsing network for human-object interaction detection

Human-Object Interaction (HOI) detection aims to infer different interactions, which occur between humans and related objects of images. HOI is usually represented by a triplet human,action,object and can be modeled as a graph. Thus, with global structural information of images, graph-based methods can detect interactions. However, in existing graph networks, although different fully-connected graphs are built, all detected bounding boxes are regarded as graph nodes equally or different types of nodes according to the category, thereby the dominant role of humans in HOI is ignored. In addition, object node representations mainly focus on appearance features, contributing little to HOI inference. To address these issues, a novel graph-based HOI detection model, named interaction-centric graph parsing network (iCGPN), models one human node as a central node, and other nodes as semantic nodes. Firstly, for each detected human instance, a human-centric fully-connected graph is constructed to learn related HOIs. Secondly, in order to reflect the difference between central nodes and semantic nodes, we design different feature representations and model different edge relationships. Through introducing the attention mechanism, global information related to human-object interaction is explored to enrich the semantic node representation, in which spatial layout, relative locations and object categories information are also combined. Finally, a multi-relation graph convolutional network is applied to update the node feature and infer the HOI. Furthermore, a multi-IOU random shift scheme is proposed to augment the data of the training set to fit the object detection deviation and enhance the generalization ability of our network. Extensive experimental results show that iCGPN achieves very competitive results in comparison with state-of-the-arraph-based methods on the V-COCO and HICO-DET datasets, which demonstrate the effectiveness of the proposed method.

Semantic maps of causation: New hybrid approaches based on corpora and grammar descriptions

AbstractThe present paper discusses connectivity and proximity maps of causative constructions and combines them with different types of typological data. In the first case study, I show how one can create a connectivity map based on a parallel corpus. This allows us to solve many problems, such as incomplete descriptions, inconsistent terminology and the problem of determining the semantic nodes. The second part focuses on proximity maps based on Multidimensional Scaling and compares the most important semantic distinctions, which are inferred from a parallel corpus of film subtitles and from grammar descriptions. The results suggest that corpus-based maps of tokens are more sensitive to cultural and genre-related differences in the prominence of specific causation scenarios than maps based on constructional types, which are described in reference grammars. The grammar-based maps also reveal a less clear structure, which can be due to incomplete semantic descriptions in grammars. Therefore, each approach has its shortcomings, which researchers need to be aware of.

Neuronal Circuits Supporting Development of Visual Naming Revealed by Intracranial Coherence Modulations.

BackgroundImprovement in visual naming abilities throughout the childhood and adolescence supports development of higher-order linguistic skills. We investigated neuronal circuits underlying improvement in the speed of visual naming with age, and age-related dynamics of these circuits.MethodsResponse times were electronically measured during an overt visual naming task in epilepsy patients undergoing stereo-EEG monitoring. Coherence modulations among pairs of neuroanatomic parcels were computed and analyzed for relationship with response time and age.ResultsDuring the overt visual naming task, mean response time (latency) significantly decreased from 4 to 23 years of age. Coherence modulations during visual naming showed that increased connectivity between certain brain regions, particularly that between left fusiform gyrus/left parahippocampal gyrus and left frontal operculum, is associated with improvement in naming speed. Also, decreased connectivity in other brain regions, particularly between left angular and supramarginal gyri, is associated with decreased mean response time. Further, coherence modulations between left frontal operculum and both left fusiform and left posterior cingulate gyri significantly increase, while that between left angular and supramarginal gyri significantly decrease, with age.ConclusionNaming speed continues to improve from pre-school years into young adulthood. This age-related improvement in efficiency of naming environmental objects occurs likely because of strengthened direct connectivity between semantic and phonological nodes, and elimination of intermediate higher-order cognitive steps.

Effects of semantic relationship and preactivation on memory updating

Semantic relationship modulates working memory (WM) processes by promoting recall but impairing recognition. Updating is a core mechanism of WM responsible for its stability and flexibility; it allows maintenance of relevant information while removing no-longer relevant one. To our knowledge, no studies specifically investigated how WM updating may benefit from the processing of semantically related material. In the current study, two experiments were run with this aim. In Experiment 1, we found an advantage for semantically related words (vs. unrelated) regardless of their association type (i.e., taxonomic or thematic). A second experiment was run boosting semantic association through preactivation. Findings replicated those of Experiment 1 suggesting that preactivation was effective and improved semantic superiority. In sum, we demonstrated that long-term semantic associations benefitted the updating process, or more generally, overall WM function. In addition, pre-activating semantic nodes of a given word appears likely a process supporting WM and updating; thus, this may be the mechanism favoring word process and memorization in a semantically related text.

Loop Closure Detection Based on Image Semantic Segmentation in Indoor Environment

When mobile robots run in indoor environment, a large number of similar images are easy to appear in the images collected, probably causing false-positive judgment in loop closure detection based on simultaneous localization and mapping (SLAM). To solve this problem, a loop closure detection algorithm for visual SLAM based on image semantic segmentation is proposed in this paper. Specifically, the current frame is semantically segmented by optimized DeepLabv3+ model to obtain semantic labels in the image. The 3D semantic node coordinates corresponding to each semantic label are then extracted by combining mask centroid and image depth information. According to the distribution of semantic nodes, the DBSCAN density clustering algorithm is adopted to cluster densely distributed semantic nodes to avoid mismatching due to the close distance of semantic nodes in the subsequent matching process. Finally, the multidimensional similarity comparison of first rough and then fine is adopted to screen the candidate frames of loop closure from key frames and then confirm the real loop closure to complete accurate loop closure detection. Testing with public datasets and self-filmed datasets, experimental results show that being well adapted to illumination change, viewpoint deviation, and item movement or missing, the proposed algorithm can effectively improve the accuracy of loop closure detection in indoor environment.

Хосров-Мирза: историческая личность и художественный образ. К вопросу о культурном своеобразии русско-персидских связей

The image of Persian Prince Khosrow Mirza, who headed the Apology Delegation to Russia, entered the Russian artistic culture of the 19th and 20th centuries. His name was mentioned by Alexander Pushkin in A Journey to Arzrum, by Nikolai Gogol in the Petersburg stories The Nose and The Portrait. Yury Tynyanov in his novel The Death of Vazir-Mukhtar (1928) elaborated in detail characteristics of the Persian prince. In Dmitri Shostakovich’s Opera The Nose, a musical and dramatic picture was built around the image of Khosrow Mirza. Although this image is inextricably linked to the memory of the death of Alexander Griboyedov, it formed a special set of motifs “colored” by the prince’s individual and personal features. The consideration of Khosrow Mirza’s image in Gogol’s stories casts doubt on the hypothesis proposed by Natalia Seregina about the connection of this image with the satirical element of the comedy The Inspector General. The image of Khosrow Mirza turns out to be the generator of the formation of the Persian text of Gogol’s Petersburg Tales. The analysis of the draft versions of the story The Portrait allows us to argue that, at the very beginning, a special semantic node is formed, which involves the generals who participated in the wars with Turkey and Iran, and the Tehran ambassador. Thus, an orientation is created for the development of an oriental theme, which was subsequently realized in the image of the terrible portrait and the demonic lender. The study of the semantic layers associated with Khosrow Mirza makes it possible to discover the connections between the Persian images of Gogol and the adventure novel The Adventures of Haji Baba from Isfahan by James Justinian Morier. Pictures of Persian life appear in the images of the Isfahan hereditary barber and local beauties, in the motif of money, cut-off noses, a head baked in bread, and are reflected in the stories Nevsky Prospect, The Nose, The Portrait. Another perspective of analysis opens up when understanding the special mission of Khosrow Mirza, aimed at the dialogue of cultures. The prince’s desire is noted to present in Russia the works of the great Persian poets Saadi and Ferdowsi, to implement the rhetoric of dialogue in an apology to Griboyedov’s mother and in a speech addressed to the Russian emperor. The introduction to scholarly discourse of translations of the book Ruzname-ye Safari Petersbourg [The Travel Notes of Khosrow Mirza] by Mirza Mustafa Afshar provides an insight into the Persian perception of the journey to Russia. The episodes of the arrival of the Iranian embassy in the royal palace, the description of the Tauride Palace, the process of making lithographs are considered. The authors conclude that the Persian embassy in St. Petersburg in 1829, caused by the tragic events, nevertheless significantly contributed to the strengthening of Iran’s authority and the formation of its unique cultural image in the minds of the Russian public. Much credit for establishing the dialogical character of relations belongs to the head of the Persian mission, Prince Khosrow Mirza. The authors declare no conflicts of interests.

Atom correlation based graph propagation for scene graph generation

Long-tailed distribution in the dataset is one of the major problems of the scene graph generation task. Previous methods attempt to alleviate this by introducing human commonsense knowledge in the form of statistical correlations between object pairs. However, the reasoning path they used is usually composable and the prior knowledge they employed is generally image-specific, making the knowledge learning less flexible, stable and holistic. In this paper, we propose Atom Correlation Based Graph Propagation (AC-GP) for the scene graph generation task. Specifically, diverse atom correlations between objects and their relationships are explored by separating relationships to form new semantic nodes and decomposing the compound reasoning paths. Based on these atom correlations, the knowledge graphs are introduced for the feature enhancement by information propagating in the global category space. By exploiting atom correlations, the introduced prior knowledge can be more common and easy to learn. Moreover, propagating the knowledge in the global category space enables the model aware of more comprehensive and holistic knowledge. As a result, the model capacity and stability can be effectively improved to mine infrequent and missed relationships. Experimental results on two benchmark datasets: Visual Relation Detection (VRD) and Visual Genome (VG) show the superiority of the proposed AC-GP over strong baseline methods.

Sarcastic evaluation in mass media as a way of discrediting a person: Greta Thunberg case

The article is devoted to the analysis of a sarcastic evaluation of a person, which leads to their discrediting in media texts. Sarcastic evaluation is considered in terms of linguistic praxeology: the language and compositional means of nomination, description, and actions are analyzed. In a media text, these means interact with the means of expressing the category of deviance and forming semantic nodes. The category of deviance can manifest itself, on the one hand, in exuberance or the absurd, while on the other, in simplification or insufficiency of the sign revelation. Also, specific sarcastic speech techniques are identified. They are based on the discrepancy of referent and illocutionary meanings in the person’s speech portrait. The study of Russian media discourse about Swedish eco-activist Greta Thunberg revealed the active use of linguistic means expressing sarcastic evaluation to demonstrate the opposing viewpoint in relation to the transmitted semantic position of “Other”. When the media represents Greta in the totality of her disadvantages, this enters into a polemic against those who support the ideas of this person. With the help of sarcasm, the media shows the absurdness and failure of these ideas. In this case, a sarcastic evaluation becomes an instrument of discrediting not only the person him/herself, but also his/her views and associates. Linguistic means of sarcastic evaluation are widely represented in discrediting media texts. The most important of them are means such as absurdity, hyperbole, alogism, simplification, etc.

Representation iterative fusion based on heterogeneous graph neural network for joint entity and relation extraction

Joint entity and relation extraction is an essential task in information extraction, which aims to extract all relational triples from unstructured text. However, few existing works consider possible relations information between entities before extracting them, which may lead to the fact that most of the extracted entities cannot constitute valid triples. In this paper, we propose a representation iterative fusion based on heterogeneous graph neural networks for relation extraction (RIFRE). We model relations and words as nodes on the graph and fuse the two types of semantic nodes by the message passing mechanism iteratively to obtain nodes representation that is more suitable for relation extraction tasks. The model performs relation extraction after nodes representation is updated. We evaluate RIFRE on two public relation extraction datasets: NYT and WebNLG. The results show that RIFRE can effectively extract triples and achieve state-of-the-art performance.11The code will be available at https://github.com/zhao9797/RIFRE. Moreover, RIFRE is also suitable for the relation classification task, and significantly outperforms the previous methods on SemEval 2010 Task 8 datasets.

The Noble Nest by I. S. Turgenev: Poetics of Costume

The costume which has recently become the focus of many scientific disciplines has hardly attracted literary critics: the methodology of its research in fiction is just beginning to take shape. The historical and cultural approach, which essentially boils down to a commentary has been replaced by a very productive structural and semiotic approach that deciphers the semantics and functions of the costume. The methodology of our research is of a complex nature, combining the structural-semiotic, mythopoetic, and motivic aspects of the analysis of literary texts. The narratological aspect in the study of costume poetics is emphasized, which, as a rule, remains outside the scope of research.The narratological aspect is aimed at identifying “flickering” meanings in the structure of the whole – “prose as poetry” (V. Schmid).The dynamics of costume descriptions, their functionality in the structure of the whole and the specifics in the organization of the narrative (taking into account the “point of view”, the motive given by costume details and based on semantic nodes that connect polar meanings, etc.) are at the center of our research.Thanks to costume inclusions, the text of the novel The Noble Nest becomes multidimensional. Thus, the characterological detail of Panshin – a screw-shaped Golden ring is situational and at the same time conceptual: it connects the “beginnings” and “ends” of his story, symbolically programming fate. Laconic sketches of Lavretsky’s clothing, scattered throughout the text, formalizing the opposition of one’s own/ another’s, prepare a motif of loneliness and homelessness. In layered narrative created by the play of the author’s and character’s points of view, Lavretsky’s point of view “migrates” to the author’s one replacing it (“poetic” sign of the optics of the hero) and then separates from it. The content of the method of crushing, which replaced the silhouette image, is an expression of the confusion of the soul, deforming the female image. The details in Lavretsky’s “split” point of view are ambiguous: on the one hand, there is alienation, on the other hand, there is a subconscious attraction to the beloved woman in the primary, unreflexed sense of a person losing happiness. The novel’s flickering meaning is created by semantic nodes that match polarities. “White” is the symbolic color of the national, rooted in the soil (the white caps of Marfa Timofeevna and Nastasya Karpovna), and at the same time the ghostly, impossible realization of happiness (the rhyming white dress of Lisa and the white dress on the portrait of Lavretsky’s mother). “Black” is also ambivalent: the elegant black silk dress of Varvara Pavlovna and the unnamed color of Lisa’s monastic clothes in the Epilogue. The scarf that Marfa Timofeevna knits is a mythologeme that encodes the story of love and failed happiness and at the same time the semantic core of the poetics of incompleteness. Keywords: costume, costume poetics, mythologeme, narrative, semantics, point of view

Linking OpenStreetMap with knowledge graphs — Link discovery for schema-agnostic volunteered geographic information

Representations of geographic entities captured in popular knowledge graphs such as Wikidata and DBpedia are often incomplete. OpenStreetMap (OSM) is a rich source of openly available, volunteered geographic information that has a high potential to complement these representations. However, identity links between the knowledge graph entities and OSM nodes are still rare. The problem of link discovery in these settings is particularly challenging due to the lack of a strict schema and heterogeneity of the user-defined node representations in OSM. In this article, we propose OSM2KG - a novel link discovery approach to predict identity links between OSM nodes and geographic entities in a knowledge graph. The core of the OSM2KG approach is a novel latent, compact representation of OSM nodes that captures semantic node similarity in an embedding. OSM2KG adopts this latent representation to train a supervised model for link prediction and utilises existing links between OSM and knowledge graphs for training. Our experiments conducted on several OSM datasets, as well as the Wikidata and DBpedia knowledge graphs, demonstrate that OSM2KG can reliably discover identity links. OSM2KG achieves an F1 score of 92.05% on Wikidata and of 94.17% on DBpedia on average, which corresponds to a 21.82 percentage points increase in F1 score on Wikidata compared to the best performing baselines.

A neuroimaging study of semantic representation in first and second languages

ABSTRACT The current study used functional magnetic resonance imaging (fMRI) to extend the embodied cognition account of language processing to second language (L2). Twenty L2 English speakers and ten native (L1) English speakers were asked to judge the semantic relatedness of English words. Behavioural data showed that L1 speakers performed the task more quickly and accurately as compared to L2 English speakers. Neurocognitive data indicated that L2 action word processing induced greater brain activation than object word processing in key language regions. In addition, although both L1 and L2 processing recruited a large brain network, significant differences were observed: L1 processing of nouns and verbs engaged a more integrated brain network connecting key language areas with sensorimotor and semantic integration nodes; for L2 processing, the connections between the semantic integration hub and sensorimotor regions were not strongly engaged. The present study sheds light on the neurocognitive representation of L2 embodied semantics.