Semantic Vector Research Articles

Improving Ranking Using Hybrid Custom Embedding Models on Persian Web

Ranking plays a crucial role in information retrieval systems, especially in the context of web search engines. This article presents a new ranking approach that utilizes semantic vectors and embedding models to enhance the accuracy of web document ranking, particularly in languages with complex structures like Persian. The article utilizes two real-world datasets, one obtained through web crawling to collect a large-scale Persian web corpus, and the other consisting of real user queries and web documents labeled with a relevancy score. The datasets are used to train embedding models using a combination of static Word2Vec and dynamic BERT algorithms. The proposed hybrid ranking formula incorporates these semantic vectors and presents a novel approach to document ranking called HybridMaxSim. Experiments conducted indicate that the HybridMaxSim formula is effective in enhancing the precision of web document ranking up to 0.87 according to the nDCG criterion.

Semantic feature norms: a cross-method and cross-language comparison

The ability to assign meaning to perceptual stimuli forms the basis of human behavior and the ability to use language. The meanings of things have primarily been probed using behavioral production norms and corpus-derived statistical methods. However, it is not known to what extent the collection method and the language being probed influence the resulting semantic feature vectors. In this study, we compare behavioral with corpus-based norms, across Finnish and English, using an all-to-all approach. To complete the set of norms required for this study, we present a new set of Finnish behavioral production norms, containing both abstract and concrete concepts. We found that all the norms provide largely similar information about the relationships of concrete objects and allow item-level mapping across norms sets. This validates the use of the corpus-derived norms which are easier to obtain than behavioral norms, which are labor-intensive to collect, for studies that do not depend on subtle differences in meaning between close semantic neighbors.

Modeling personality language use with small semantic vector subspaces

Modeling personality language use with small semantic vector subspaces

КАТЕГОРИЯ «ОППОЗИЦИЯ» В ТЕРМИНОЛОГИЧЕСКОМ ПОЛЕ «ХИРУРГИЧЕСКИЙ ИНСТРУМЕНТАРИЙ» (НА ПРИМЕРЕ ТЕРМИНОВ НЕМЕЦКОГО ЯЗЫКА)

In this article, the category “opposition” in the terminological field “surgical instruments” of the German language is examined by cognitive approach. Antonyms are analyzed in accordance with the semantic and structural classification developed by L.A. Novikov. The purpose of this work is to analyze and describe the terms of the studied terminological field, in the semantics of which there are indications of the category “opposition”. The materials for the study were compound terms (774 terms) and terminological word-combinations (524 terms) in the German language. The results of the analysis showed that the category “opposition” is characteristic for the analyzed terminological field. Using semantic classification contrarian, complementary and vector antonyms were identified in the studied terminological field. According to their structure antonyms of the terminological field “surgical instruments” in the German language are represented by compound terms and terminological word-combinations.

A multi-message passing framework based on heterogeneous graphs in conversational emotion recognition

As an important development direction of natural language processing, emotion recognition in conversation (ERC) remains a challenge in sentiment analysis. Given the large-scale dialogue datasets and their wide application in the fields of recommendation systems and human–machine dialogue systems, researchers have begun to pay more attention to the issue of ERC. In recent research, the task of ERC has been largely based on the graph structure to model the speaker level. However, most existing studies simply splice multimodal features, and the heterogeneity of multimodal features tends to be overlooked. Hence, this paper proposes a multivariate messaging framework to embed heterogeneous information into multimodal relational graphs. In the process of aggregating graph node information, we take into account the homogeneity of nodes and assign different weights to different nodes so as to better aggregate semantic information. In order to improve the robustness of the model, we utilize the mechanism of sharing weights among neighbors to reduce the number of network parameters and improve the generalization ability of the model. In so doing, the node information is aggregated through the constructed graph network, and the final semantic vector representation is obtained. Experiments over two benchmark datasets for ERC show that our proposed model achieves improved performance in accuracy and F1 value.

Research on Multi-Domain Intelligent Customer Service Dialog Modeling with Integrated Transfer Learning Strategies

Abstract The intelligent customer service dialog model is centered on human-machine dialog, which has good prospects for commercial applications in multiple domains. In this paper, we use the Siamese-LSTM model to do vectorization of questions in the FAQ question and answer database to get the semantic representation vector of sentences, and then use the approximate retrieval algorithm to index the question and answer database and perform approximate nearest-neighbor retrieval of the query. After completing the question query, migration learning is employed to create a mapping between input questions and human responses, enabling the model to produce sentences that are similar to human responses. Tests show that the task success rate gradually stabilizes around 0.80 at about the 100th round and fluctuates up to around 0.986 after that. For the average number of conversation rounds, migration learning improves the conversation efficiency of intelligent customer service, and the average number of conversation rounds gradually stabilizes at about 150 rounds and eventually stabilizes at about 4.2 rounds as the number of training rounds increases. The transfer learning strategy helps machine responses to be as close to human responses as possible.

Generalized Zero-Shot Space Target Recognition Based on Global-Local Visual Feature Embedding Network

Existing deep learning-based space target recognition methods rely on abundantly labeled samples and are not capable of recognizing samples from unseen classes without training. In this article, based on generalized zero-shot learning (GZSL), we propose a space target recognition framework to simultaneously recognize space targets from both seen and unseen classes. First, we defined semantic attributes to describe the characteristics of different categories of space targets. Second, we constructed a dual-branch neural network, termed the global-local visual feature embedding network (GLVFENet), which jointly learns global and local visual features to obtain discriminative feature representations, thereby achieving GZSL for space targets with higher accuracy. Specifically, the global visual feature embedding subnetwork (GVFE-Subnet) calculates the compatibility score by measuring the cosine similarity between the projection of global visual features in the semantic space and various semantic vectors, thereby obtaining global visual embeddings. The local visual feature embedding subnetwork (LVFE-Subnet) introduces soft space attention, and an encoder discovers the semantic-guided local regions in the image to then generate local visual embeddings. Finally, the visual embeddings from both branches were combined and matched with semantics. The calibrated stacking method is introduced to achieve GZSL recognition of space targets. Extensive experiments were conducted on an electromagnetic simulation dataset of nine categories of space targets, and the effectiveness of our GLVFENet is confirmed.

Linking Allusion Words: A Method of Combining Fine‐Grained Co‐citation Relationship and Semantic Features

ABSTRACTIt is a common phenomenon for Tang poems to cite the allusions, which can generate a rich relationship network. However, insufficient attention has been paid to investigating the relationship network. To address the research gap, by employing theories and methods of information science, this study presents a method of combining fine‐grained co‐citation relationship and semantic features to link allusion words. We constructed a fine‐grained co‐citation network between allusion words by adding cited positions and sentiments. We then transformed the fine‐grained weights into relational similarities. Moreover, we also leveraged the explanatory text as semantic information for each allusion word, mapping the semantic embedding vectors and calculating the similarities as the semantic similarities. Finally, we applied the link prediction algorithm to implement the allusion word linking. Our experimental results reveal that adding the cited positions and sentiments as well as semantic similarities can improve the performance of allusion word linking, achieving 0.869 on score. Additionally, we explore the linking results from the perspective of the shortest path and find some regular knowledge. Overall, our study extends the application of information science and promotes the development of Chinese traditional cultural resources.

Towards a temporospatial framework for measurements of disorganization in speech using semantic vectors.

Incoherent speech in schizophrenia has long been described as the mind making "leaps" of large distances between thoughts and ideas. Such a view seems intuitive, and for almost two decades, attempts to operationalize these conceptual "leaps" in spoken word meanings have used language-based embedding spaces. An embedding space represents meaning of words as numerical vectors where a greater proximity between word vectors represents more shared meaning. However, there are limitations with word vector-based operationalizations of coherence which can limit their appeal and utility in clinical practice. First, the use of esoteric word embeddings can be conceptually hard to grasp, and this is complicated by several different operationalizations of incoherent speech. This problem can be overcome by a better visualization of methods. Second, temporal information from the act of speaking has been largely neglected since models have been built using written text, yet speech is spoken in real time. This issue can be resolved by leveraging time stamped transcripts of speech. Third, contextual information - namely the situation of where something is spoken - has often only been inferred and never explicitly modeled. Addressing this situational issue opens up new possibilities for models with increased temporal resolution and contextual relevance. In this paper, direct visualizations of semantic distances are used to enable the inspection of examples of incoherent speech. Some common operationalizations of incoherence are illustrated, and suggestions are made for how temporal and spatial contextual information can be integrated in future implementations of measures of incoherence.

Феноменология рубежности. К типологии культурных интерференций

The article raises the question of the necessity to update the nomenclature of the humanities. A productive direction may be the appeal to the concept of borderline which has already been developed in a broad cultural field. However, using the concept of borderline requires adequate methodology which is not available within the modern scientific paradigm yet. Thus, the problem requires theoretical understanding from the point of view of phenomenology. What is meant by the phenomenological approach here is concentration on the anthropological, cultural (in a broad sense) reality of human existence. In this case, the phenomenology of works of art allows revealing much more meaning than traditional historical and cultural analysis. In other words, rather than study the correlation of forms, styles and movements, it is supposed to try to discover hidden narratives of the epoch of borderline as a special cultural landscape, since any borderline is, by definition, a chronotope of changes and renovations; it implies the emergence of a new quality of being, a new world relationship. The phenomenology of borderline reveals a dual cultural identity in which different semantic vectors are brought to the fore depending on the circumstances.

Variational autoencoder based on distributional semantic embedding and cross-modal reconstruction for generalized zero-shot fault diagnosis of industrial processes

The traditional fault diagnosis models cannot achieve good fault diagnosis accuracy when a new unseen fault class appears in the test set, but there is no training sample of this fault in the training set. To solve this problem, a generalized zero-shot fault diagnosis model for industrial processes based on distributional semantic embedding and cross-modal reconstruction variational autoencoder (DSECMR-VAE) is proposed. DSECMR-VAE uses two variational autocoders (VAEs) to encode the features from two different modalities of fault samples and fault attribute semantic vectors into latent variables, and then uses the latent variables to generate fault samples belonging to the seen and unseen fault classes. In addition, a Barlow matrix is designed specifically for the distribution parameters of the latent variables, this matrix is utilized to measure the consistency between the distribution of fault samples and fault attribute semantic vectors. Subsequently, cross-modal reconstruction is performed on VAE. Cross-modal reconstruction uses the input from different modalities to reconstruct the current input, which can fully combine the information from different modalities. Finally, a classifier is trained based on the obtained latent variables to realize the generalized zero-shot fault diagnosis. Two case studies demonstrate the effectiveness and superiority of the proposed model for zero-shot and generalized zero-shot fault diagnosis.

A semantic union model for open domain Chinese knowledge base question answering

In Open-domain Chinese Knowledge Base Question Answering (ODCKBQA), most common simple questions can be answered by a single relational fact in the knowledge base (KB). The abbreviations, aliases, and nesting of entities in Chinese question sentences, and the gap between them and the structured semantics in the knowledge base, make it difficult for the system to accurately return answers. This study proposes a semantic union model (SUM), which concatenates candidate entities and candidate relationships, using a contrastive learning algorithm to learn the semantic vector representation of question and candidate entity-relation pairs, and perform cosine similarity calculations to simultaneously complete entity disambiguation and relation matching tasks. It can provide information for entity disambiguation through the relationships between entities, avoid error propagation, and improve the system performance. The experimental results show that the system achieves a good average F1 of 85.94% on the dataset provided by the NLPCC-ICCPOL 2016 KBQA task.

A semantic vector map-based approach for aircraft positioning in GNSS/GPS denied large-scale environment

Accurate positioning is one of the essential requirements for numerous applications of remote sensing data, especially in the event of a noisy or unreliable satellite signal. Toward this end, we present a novel framework for aircraft geo-localization in a large range that only requires a downward-facing monocular camera, an altimeter, a compass, and an open-source Vector Map (VMAP). The algorithm combines the matching and particle filter methods. Shape vector and correlation between two building contour vectors are defined, and a coarse-to-fine building vector matching (CFBVM) method is proposed in the matching stage, for which the original matching results are described by the Gaussian mixture model (GMM). Subsequently, an improved resampling strategy is designed to reduce computing expenses with a huge number of initial particles, and a credibility indicator is designed to avoid location mistakes in the particle filter stage. An experimental evaluation of the approach based on flight data is provided. On a flight at a height of 0.2 km over a flight distance of 2 km, the aircraft is geo-localized in a reference map of 11,025 km2 using 0.09 km2 aerial images without any prior information. The absolute localization error is less than 10 m.

IRC-CLVul: Cross-Programming-Language Vulnerability Detection with Intermediate Representations and Combined Features

The most severe problem in cross-programming languages is feature extraction due to different tokens in different programming languages. To solve this problem, we propose a cross-programming-language vulnerability detection method in this paper, IRC-CLVul, based on intermediate representation and combined features. Specifically, we first converted programs in different programming languages into a unified LLVM intermediate representation (LLVM-IR) to provide a classification basis for different programming languages. Afterwards, we extracted the code sequences and control flow graphs of the samples, used the semantic model to extract the program semantic information and graph structure information, and concatenated them into semantic vectors. Finally, we used Random Forest to learn the concatenated semantic vectors and obtained the classification results. We conducted experiments on 85,811 samples from the Juliet test suite in C, C++, and Java. The results show that our method improved the accuracy by 7% compared with the two baseline algorithms, and the F1 score showed a 12% increase.

Experientially-grounded and distributional semantic vectors uncover dissociable representations of conceptual categories

ABSTRACT Neuronal populations code similar concepts by similar activity patterns across the human brain's semantic networks. However, it is unclear to what extent such meaning-to-symbol mapping reflects distributional statistics, or experiential information grounded in sensorimotor and emotional knowledge. We asked whether integrating distributional and experiential data better distinguished conceptual categories than each method taken separately. We examined the similarity structure of fMRI patterns elicited by visually presented action- and object-related words using representational similarity analysis (RSA). We found that the distributional and experiential/integrative models respectively mapped the high-dimensional semantic space in left inferior frontal, anterior temporal, and in left precentral, posterior inferior/middle temporal cortex. Furthermore, results from model comparisons uncovered category-specific similarity patterns, as both distributional and experiential models matched the similarity patterns for action concepts in left fronto-temporal cortex, whilst the experiential/integrative (but not distributional) models matched the similarity patterns for object concepts in left fusiform and angular gyrus.

Sheep face image dataset and DT-YOLOv5s for sheep breed recognition

Automatic identification of breeds of sheep can be valuable to the sheep industry. Due to the lack of open source datasets of multiple breeds in some researches, which leads to poor robustness of models for automatic sheep breed recognition. This paper proposes a sheep face image dataset of 10 breeds, which consists of 5200 original images of 823 sheep from seven farms in Inner Mongolia. The dataset is expanded to 44,747 images after data enhancement. A sheep breed recognition model DT-YOLOv5s based on knowledge distillation is proposed for the dataset. The model transfers the knowledge of sheep face features learned by the teacher network YOLOv5x, which has a large number of parameters but high recognition accuracy, to the lightweight student network YOLOv5s through a 15-dimensional high-dimensional semantic feature vector, so as to improve the student network's PANet feature extraction capability and recognition accuracy, achieving the purpose of lightweight and accurate recognition of the overall network model. The experimental results show that the accuracy of our proposed model is 92.75%, which is 11.19%, 13.1% and 13.65% higher than that of SSD, Faster R-CNN and YOLOv5s models, respectively. The mean average precision (mAP@[0.5:0.95]) is 94.67%, which is 7.85%, 2.35% and 4.83% higher than that of SSD, Faster R-CNN and YOLOv5s models, respectively. Meanwhile, the inference time of the model is 12.63 ms, which is 33.9 ms, 53.13 ms and 74.61 ms less than SSD, Faster R-CNN and YOLOv5x, respectively. The proposed DT-YOLOv5s model in this paper has good robustness and generalization ability, which can provide technical support for the real-time accurate identification of the sheep breed for farmers.

Memorial GAN With Joint Semantic Optimization for Unpaired Image Captioning.

Most works of image captioning are implemented under the full supervision of paired image-caption data. Limited to expensive cost of data collection, the task of unpaired image captioning has attracted researchers' attention. In this article, we propose a novel memorial GAN (MemGAN) with the joint semantic optimization for unpaired image captioning. The core idea is to explore implicit semantic correlation between disjointed images and sentences through building a multimodal semantic-aware space (SAS). Concretely, each modality is mapped into a unified multimodal SAS, where SAS includes the semantic vectors of image I, visual concepts O, unpaired sentence S, and the generated caption C. We adopt the memory unit based on multihead attention and relational gate as a backbone to preserve and transit crucial multimodal semantics in the SAS for image caption generation and sentence reconstruction. Then, the memory unit is embedded into a GAN framework to exploit the semantic similarity and relevance in SAS, that is, imposing a joint semantic-aware optimization on SAS without supervision clues. To summarize, the proposed MemGAN learns the latent semantic relevance of SAS's multimodalities in an adversarial manner. Extensive experiments and qualitative results demonstrate the effectiveness of MemGAN, achieving improvements over state of the arts on unpaired image captioning benchmarks.

Semantic Clone Detection Based on Code Feature Fusion Learning

Code clones are duplicated code snippets that significantly threaten software maintenance and the public corpora of code representation learning. Traditionally, code context and its structure information abstract syntax tree (AST), control flow graph (CFG) are typical representations of source code, and context-based models and structure-based models contributed significantly to the development of code clone detection. In this paper, we present a hybrid embedding model for code clone detection (HEM-CCD), a fusion method of token sequential information and graph-based structure information. We insert tokens’ global context information encoded by a bi-directional recurrent neural network into the AST-based graph for comprehensive code semantic representation. Then, feeding the graph into a gated graph neural network we generate code semantic vectors for similarity evaluation. We have implemented our model on two public clone datasets (BigCloneBench and GoogleCodeJam), and the results indicate that HEM-CCD outperforms several state-of-the-art approaches.

Sentiment Analysis of Social Network Comment Text Based on LSTM and Bert

This paper has the aim of solving problems in research studies on the analysis tasks of text emotion; the problems are the low utilization of text, the difficulty of effective information extraction, the failure of recognizing word polysemy with effectiveness. Thus, based on LSTM and Bert, the method of sentiment analysis on text is adopted. To be precise, word embedding of dataset in view of the skip-gram model is used for training course. In each sample, the word embeddings combine matric with the two-dimensional feature to be neural network input. Next, construction of analysis model for text sentiment combines Bert pre-training language model and long short-term memory (LSTM) network, using the word vector pre-trained by Bert instead of that trained in the traditional way to dynamically generate the semantic vector according to the word context. Finally, the semantic representation of words from text is improved by effectively identifying the polysemy of words, and the semantic vector is input into the LSTM to capture the semantic dependencies, thereby enhancing the ability to extract valid information. The Accuracy, Precision, Recall and F-Measure for the method of Bert–LSTM based on analysis of text sentiment are 0.89, 0.9, 0.84 and 0.87, indicating high value than the compared ones. Thus, the proposed method significantly outperforms the comparison methods in text sentiment analysis.

Topic sentiment analysis based on deep neural network using document embedding technique.

Sentiment Analysis (SA) is a domain- or topic-dependent task since polarity terms convey different sentiments in various domains. Hence, machine learning models trained on a specific domain cannot be employed in other domains, and existing domain-independent lexicons cannot correctly recognize the polarity of domain-specific polarity terms. Conventional approaches of Topic Sentiment Analysis perform Topic Modeling (TM) and SA sequentially, utilizing the previously trained models on irrelevant datasets for classifying sentiments that cannot provide acceptable accuracy. However, some researchers perform TM and SA simultaneously using topic-sentiment joint models, which require a list of seeds and their sentiments from widely used domain-independent lexicons. As a result, these methods cannot find the polarity of domain-specific terms correctly. This paper proposes a novel supervised hybrid TSA approach, called Embedding Topic Sentiment Analysis using Deep Neural Networks (ETSANet), that extracts the semantic relationships between the hidden topics and the training dataset using Semantically Topic-Related Documents Finder (STRDF). STRDF discovers those training documents in the same context as the topic based on the semantic relationships between the Semantic Topic Vector, a newly introduced concept that encompasses the semantic aspects of a topic, and the training dataset. Then, a hybrid CNN-GRU model is trained by these semantically topic-related documents. Moreover, a hybrid metaheuristic method utilizing Grey Wolf Optimization and Whale Optimization Algorithm is employed to fine-tune the hyperparameters of the CNN-GRU network. The evaluation results demonstrate that ETSANet increases the accuracy of the state-of-the-art methods by 1.92%.