Textual Features Research Articles

COMICS AS A MEANS OF PRESERVING CLASSICAL LITERATURE IN MODERN PEOPLE’S READING (The case of W. Shakespeare)

The article focuses on adaptations of the tragedy “Hamlet” and the comedy “A Midsummer Night’s Dream” by W. Shakespeare in the form of comic books. The object of study is the peculiarities of comics as a text in demand by the representatives of clip consciousness, which is formed under the influence of the IT. The subject of study is the peculiarities of the source text transformation in comic adaptations of the 20th–21st centuries. The study is based on two adaptations of “Hamlet” and three adaptations of “A Midsummer Night’s Dream” by W. Shakespeare. The tasks of the study were as follows: to determine the connection between the comic book genre and clip consciousness; to highlight the characteristic features of the comics genre; to analyse the reflection of genre features in concrete texts; to describe changes in the source text depending on the nature of the audience. The study showed that the popularity of comics as a genre is inextricably linked with the emergence of clip consciousness. Certain features of the comics genre correlate with those of clip consciousness: fragmentation corresponds to mastering information in small portions, the emphasis on the visual presentation of information is associated with the perception of the world through visual images.

All Research Summaries Are Scientific, but Some Are More Scientific Than Others?—Indications for a Mediation of the Relationship Between Author and Text Scientificness Through Perceived Scientificness

Especially in times characterized by abundant misinformation, knowledge on fostering trust as a scientist is crucial. Previous studies have found that readers rate texts displaying a scientific style as more credible and show higher agreement with their message, a phenomenon called “scientificness effect”. However, it is unclear how strongly the relationship between scientific text features and trust is influenced by readers’ subjective scientificness perceptions, and if such an influence can also be found on an author level. We thus conducted an experimental online study with N = 838 German-speaking layreaders. Participants read two research summaries with randomized levels of author and text scientificness (low vs. high). After each text, they rated the scientificness and the trustworthiness of the author and the text. Mediation analyses provided preliminary evidence for our hypotheses. Under the assumptions of our model, perceived scientificness was found to be a mediator of the scientificness-trustworthiness relationship on an author and text level. Limitations regarding our cross-sectional mediation design are discussed. Based on our results, we offer suggestions for future research and perspectives for future theory-building.

“Making Sure the Books Check Every Box”: Preservice Teachers Learning to Select Texts for Reading Instruction

ABSTRACT A key consideration in providing effective reading instruction is the selection of a text. However, little is known about how preservice teachers are prepared to understand text complexity and how they leverage that knowledge in their text selection practices. Accordingly, this design/development study explored how preservice teachers (n = 81) gained declarative knowledge of text complexity, procedural knowledge of text selection practices, and conditional knowledge of activity, reader, and text considerations in their teacher education program. Findings indicate that following a two-course reading methods sequence in which they selected texts for small group instruction, typically performing preservice teachers gave the most attention to activity considerations (e.g. lesson-to-text match, instructional sequence) and the least attention to reader considerations (e.g. student interest). When noting text features teachers considered important, they described word-level features such as multisyllabic words, vocabulary, and decodability as most influencing text complexity. Furthermore, preservice teachers who held more declarative knowledge of text complexity tended to be more critical of text leveling systems whereas those with less knowledge of text complexity tended to view levels as an acceptable method of selecting texts. Together these findings suggest an emergent learning continuum on the topic of text selection for teacher education.

A transfer learning based patent transfer prediction model: Discussion about the condition of China and U.S.

In the same patent classification, there are significant differences in the patent transfer quantity for different countries. For the countries with small patent transfer quantities, the performance of the patent transfer prediction is worse than that of high-transfer countries. To improve the performance of the patent prediction model learned for the low-transfer countries, we propose a transfer learning based patent transfer prediction model, for which the high-transfer countries are the source domain and the low-transfer countries are the target domain. In the model, the textual feature of each patent text is embedded using BERT-CNN with the parameters fine-tuned on the target domain from the source domain. The textual features are optimized to minimize the domain loss of Maximum Mean Discrepancy between source and target domain. What is more, we extract graph feature vectors by embedding the patent node in a three-layer heterogeneous network which consists of patent nodes, International Patent Classification nodes, consumer nodes, consumer location nodes and their relations. Finally, the patent text vectors, structural feature vectors, and graph feature vectors are spliced together to predict target domain patents transfer or not. The performance of the model is verified by some comparative experiments. The experimental results show that the proposed model outperforms significantly the baseline models in the values of Precision, Recall, and F1 of patent data improve by 6.47%, 6.54%, and 6.58% on average, respectively.

Triple-Stream Commonsense Circulation Transformer Network for Image Captioning

Traditional image captioning methods only have a local perspective at the dataset level, allowing them to explore dispersed information within individual images. However, the lack of a global perspective prevents them from capturing common characteristics among similar images. To address the limitation, this paper introduces a novel Triple-stream Commonsense Circulating Transformer Network (TCCTN). It incorporates contextual stream into the encoder, combining enhanced channel stream and spatial stream for comprehensive feature learning. The proposed commonsense-aware contextual attention (CCA) module queries commonsense contextual features from the dataset, obtaining global contextual association information by projecting grid features into the contextual space. The pure semantic channel attention (PSCA) module leverages compressed spatial domain for channel pooling, focusing on attention weights of pure channel features to capture inherent semantic features. The region spatial attention (RSA) module enhances spatial concepts in semantic learning by incorporating region position information. Furthermore, leveraging the complementary differences among the three features, TCCTN introduces the mixture of experts strategy to enhance the unique discriminative ability of features and promote their integration in textual feature learning. Extensive experiments on the MS-COCO dataset demonstrate the effectiveness of contextual commonsense stream and the superior performance of TCCTN.

Joint multimodal entity-relation extraction based on temporal enhancement and similarity-gated attention

Joint Multimodal Entity and Relation Extraction (JMERE), which needs to combine complex image information to extract entity-relation quintuples from text sequences, posts higher requirements of the model’s multimodal feature fusion and selection capabilities. With the advancement of large pre-trained language models, existing studies focus on improving the feature alignments between textual and visual modalities. However, there remains a noticeable gap in capturing the temporal information present in textual sequences. In addition, these methods exhibit a certain deficiency in distinguishing irrelevant images when integrating image and text features, making them susceptible to interference from image information unrelated to the text. To address these challenges, we propose a temporally enhanced and similarity-gated attention network (TESGA) for joint multimodal entity relation extraction. Specifically, we first incorporate an LSTM-based Text Temporal Enhancement module to enhance the model’s ability to capture temporal information from the text. Next, we introduce a Text-Image Similarity-Gated Attention mechanism, which controls the degree of incorporating image information based on the consistency between image and text features. Subsequently, We design the entity and relation prediction module using a form-filling approach based on entity and relation types, and conduct prediction of entity-relation quintuples. Notably, apart from the JMERE task, our approach can also be applied to other tasks involving text-visual enhancement, such as Multimodal Named Entity Recognition (MNER) and Multimodal Relation Extraction (MRE). To demonstrate the effectiveness of our approach, our model is extensively experimented on three benchmark datasets and achieves state-of-the-art performance. Our code will be available upon paper acceptance.11https://github.com/vacuum-cup/TESGA.

Region-guided transformer for remote sensing image captioning

ABSTRACT Remote sensing image acquisition is an essential way to obtain information. However, research on remote sensing images mainly focuses on object detection or image classification. The emergence of remote sensing image captioning (RSIC) has enabled understanding and inference of remote sensing images, thus attracting considerable attention. There are still challenges in RSIC: the features used in RSIC are mostly based on grid features, and this form of features makes it difficult for the model to determine the main description targets. Hence, a more effective cross-modal matching method is needed for better text generation. Thus, we propose a region-guided transformer in response to the aforementioned issues. We extracted region features to enhance the ability of the model to focus on the main targets. To address the issue of information loss caused by region feature extraction, we proposed environment features to supplement background information. To improve the matching between text and image features, we propose a region-guided decoder that enhances the model's perception of different features through a weighted cross-attention mechanism. Meanwhile, we introduce region-guided information to guide the text-generation process. The effectiveness and superiority of our model have been demonstrated through extensive experiments.

TFG-Net: A Text Feature-Guided Network for Small Traffic Sign Detection.

Detecting small signs in complex real-world environments remains challenging due to limited feature information and interference from other objects. In this article, we propose a novel text feature-guided network (TFG-Net) to improve the performance of the small signs detection not only enhancing the feature information of small signs but also avoiding the influence of other objects. As the name suggests, TFG-Net incorporates a text detection branch, which extracts additional textual features from the signs and supplies them to the object detection branch. Furthermore, the object detection branch of TFG-Net optimizes the backbone network's output structure by merging deep features and introducing a high-resolution feature layer. Finally, a fusion method that enhances both overall and local features is proposed to fully integrate detailed and semantic information. Experimental results display that our TFG-Net reaches the highest mean average precision (mAP) of 92.5% on the public datasets Tsinghua-Tencent 100K (TT100K), 83.7% on CCTSDB2021, and 79.1% on DFG, surpassing current state-of-the-art object detectors.

Exploring the Effectiveness of Shallow and L2 Learner-Suitable Textual Features for Supervised and Unsupervised Sentence-Based Readability Assessment

Simplicity in information found online is in demand from diverse user groups seeking better text comprehension and consumption of information in an easy and timely manner. Readability assessment, particularly at the sentence level, plays a vital role in aiding specific demographics, such as language learners. In this paper, we research model evaluation metrics, strategies for model creation, and the predictive capacity of features and feature sets in assessing readability based on sentence complexity. Our primary objective is to classify sentences as either simple or complex, shifting the focus from entire paragraphs or texts to individual sentences. We approach this challenge as both a classification and clustering task. Additionally, we emphasize our tests on shallow features that, despite their simplistic nature and ease of use, seem to yield decent results. Leveraging the TextStat Python library and the WEKA toolkit, we employ a wide variety of shallow features and classifiers. By comparing the outcomes across different models, algorithms, and feature sets, we aim to offer valuable insights into optimizing the setup. We draw our data from sentences sourced from Wikipedia’s corpus, a widely accessed online encyclopedia catering to a broad audience. We strive to take a deeper look at what leads to greater readability classification in datasets that appeal to audiences such as Wikipedia’s, assisting in the development of improved models and new features for future applications with low feature extraction/processing times.

Less is more: A closer look at semantic-based few-shot learning

Few-shot Learning (FSL) aims to learn and distinguish new categories from a scant number of available samples, presenting a significant challenge in the realm of deep learning. Recent researchers have sought to leverage the additional semantic or linguistic information of scarce categories with a pre-trained language model to facilitate learning, thus partially alleviating the problem of insufficient supervision signals. Nonetheless, the full potential of the semantic information and pre-trained language model have been underestimated in the few-shot learning till now, resulting in limited performance enhancements. To address this, we propose a straightforward and efficacious framework for few-shot learning tasks, specifically designed to exploit the semantic information and language model. Specifically, we explicitly harness the zero-shot capability of the pre-trained language model with learnable prompts. And we directly add the visual feature with the textual feature for inference without the intricate designed fusion modules as in prior studies. Additionally, we apply the self-ensemble and distillation to further enhance performance. Extensive experiments conducted across four widely used few-shot datasets demonstrate that our simple framework achieves impressive results. Particularly noteworthy is its outstanding performance in the 1-shot learning task, surpassing the current state-of-the-art by an average of 3.3% in classification accuracy. Our code will be available at https://github.com/zhouchunpong/SimpleFewShot.

Advancing sentiment classification through a population game model approach

Computational Sentiment Analysis involves the automation of human emotion comprehension by categorizing sentiments as positive, negative, or neutral. In the contemporary digital environment, the extensive volume of social media content presents significant challenges for manual analysis, thereby necessitating the development and implementation of automated analytical tools. To address the limitations of existing techniques, which heavily rely on machine learning and extensive dataset pre-training, we propose an innovative unsupervised approach for sentiment classification. This novel methodology is grounded in game theory concepts, particularly the population game model, offering a promising solution by circumventing the need for extensive training procedures. We extract two textual features from review comments, namely context score and emotion score. Leveraging lexicon databases and numeric scores, this cognitive mathematical framework is language-independent. Competitive results are demonstrated across various domains (hotels, restaurants, electronic devices, etc.), and the efficacy of the proposed work is validated in two languages (English and Hindi). The highest accuracy recorded for the English domain dataset is 89%, while electronic Hindi reviews attain an 84% accuracy rate. The proposed model exhibits domain and language independence, validated through statistical analyses confirming the significance of the findings. The framework demonstrates noteworthy rationality and coherence in its outcomes.

Exploring the synergy between textual identity and visual signals in human-object interaction

Human-Object Interaction (HOI) detection task aims to recognize and understand interactions between humans and objects depicted in images. Unlike instance recognition tasks, which focus on isolated objects, HOI detection requires considering various explanatory factors, such as instance identity, spatial relationships, and scene context. However, previous HOI detection methods have primarily relied on local visual cues, often overlooking the vital role of instance identity and thus limiting the performance of models. In this paper, we introduce textual features to expand the definition of HOI representations, incorporating instance identity into the HOI reasoning process. Drawing inspiration from the human activity perception process, we explore the synergy between textual identity and visual signals to leverage various explanatory factors more effectively and enhance HOI detection performance. Specifically, our method extracts HOI explanatory factors using both modal representations. Visual features capture interactive cues, while textual features explicitly denote instance identities within human-object pairs, delineating relevant interaction categories. Additionally, we utilize Contrastive Language-Image Pre-training (CLIP) to enhance the semantic alignment between visual and textual features and design a cross-modal learning module for integrating HOI multimodal information. Extensive experiments on several benchmarks demonstrate that our proposed framework surpasses most existing methods, achieving outstanding performance with a mean average precision (mAP) of 33.95 on the HICO-DET dataset and 63.2 mAP on the V-COCO dataset.

VTPL: Visual and text prompt learning for visual-language models

Visual-language (V-L) models have achieved remarkable success in learning combined visual–textual representations from large web datasets. Prompt learning, as a solution for downstream tasks, can address the forgetting of knowledge associated with fine-tuning. However, current methods focus on a single modality and fail to fully use multimodal information. This paper aims to address these limitations by proposing a novel approach called visual and text prompt learning (VTPL) to train the model and enhance both visual and text prompts. Visual prompts align visual features with text features, whereas text prompts enrich the semantic information of the text. Additionally, this paper introduces a poly-1 information noise contrastive estimation (InfoNCE) loss and a center loss to increase the interclass distance and decrease the intraclass distance. Experiments on 11 image datasets show that VTPL outperforms state-of-the-art methods, achieving 1.61%, 1.63%, 1.99%, 2.42%, and 2.87% performance boosts over CoOp for 1, 2, 4, 8, and 16 shots, respectively.

A Robust Solution for Tea Leaf Classification using the SMOTE and Soft Voting Classifier

Background/Objectives: The purpose of the proposed work is to identify how a tea leaf is conceptualized. This further enables the categorization of leaves into healthy and unhealthy groups, assisting both the vendor and customer in obtaining the right quality for their requirements. The primary cause of the low production of tea in various parts of the country is the late detection of diseased leaves. Literature suggests that the current AI techniques can detect but have a high level of similarities. Hence, the objective is to identify diseased leaf images at early stage. Methods: Earlier studies have used mostly CNN for classification which is more time consuming, and they create their own parameters. In this study, we extracted the textual features of images by using GLCM, and further classification was implemented with the help of SMOTE and a soft voting classifier. Findings: The proposed method resulted in optimal timeline without using CNN for classification. The accuracy is better when compared to current methods. Novelty: This work presents a novel approach in calculating textual features of leaves that can be further used for classification or tea-leaf grading purposes. Keywords: Image processing, Feature extraction, GLCM, SMOTE, Voting classifier

A Novel Approach for Stratifying Pulmonary Edema Severity on Chest X-ray via Dual-Mechanic Self-Learning and Bidirectional Multi-Modal Cross-Attention Algorithms

Abstract Accurate assessment of pulmonary edema severity in acute decompensated congestive heart failure (CHF) patients is vital for treatment decisions. Traditional methods face challenges due to the complexity of chest X-ray (CXR) and unstructured radiology reports. We proposed a method combining self-supervised learning and multimodal cross-attention to address these challenges. Dual-mechanic self-supervised pre-training enhances feature extraction using contrastive learning between text and image features, and generative learning between images. A bidirectional multi-modal cross-attention model integrates image and text information for fine-tuning, improving model performance. Four CXR datasets consisting of 519, 437 images were used for pre-training; 1200 randomly selected image-text pairs were used for fine-tuning and partitioned into train, validation, and test sets at 3: 1: 1. Ablation studies for pre-training and fine-tuning approaches indicated their practicality as evidenced by the optimal macro F1 score of 0.667 and optimal macro-AUC of 0.904. It also outperformed other state-of-the-art multi-modality methods. The novel approach could accurately assess pulmonary edema severity, offering crucial support for CHF patient management.

Integrating multimodal contrastive learning with prototypical domain alignment for unsupervised domain adaptation of time series

Unsupervised domain adaptation (UDA) addresses the challenge of transferring knowledge from a labeled source domain to an unlabeled target domain. This task is particularly critical for time series data, characterized by unique temporal dynamics. However, existing methods often fail to capture these temporal dependencies, leading to domain discrepancies and loss of semantic information. In this study, we propose a novel framework for the unsupervised domain adaptation of time series (UDATS) that integrates Multimodal Contrastive Adaptation (MCA) and Prototypical Domain Alignment (PDA). MCA leverages image encoding techniques and prompt learning to capture complex temporal patterns while preserving semantic information. PDA constructs multimodal prototypes, combining visual and textual features to align target domain samples accurately. Our framework demonstrates superior performance across various application domains, including human activity recognition, mortality prediction, and fault detection. Experiments show our method effectively addresses domain discrepancies while preserving essential semantic content, outperforming state-of-the-art models.

HGATT_LR: transforming review text classification with hypergraphs attention layer and logistic regression

Text classification plays a major role in research such as sentiment analysis, opinion mining, and customer feedback analysis. Text classification using hypergraph algorithms is effective in capturing the intricate relationships between words and phrases in documents. The method entails text preprocessing, keyword extraction, feature selection, text classification, and performance metric evaluation. Here, we proposed a Hypergraph Attention Layer with Logistic Regression (HGATT_LR) for text classification in the Amazon review data set. The essential keywords are extracted by utilizing the Latent Dirichlet Allocation (LDA) technique. To build a hypergraph attention layer, feature selection based on node-level and edge-level attention is assessed. The resultant features are passed as an input of Logistic regression for text classification. Through a comparison analysis of different text classifiers on the Amazon data set, the performance metrics are assessed. Text classification using hypergraph Attention Network has been shown to achieve 88% accuracy which is better compared to other state-of-the-art algorithms. The proposed model is scalable and may be easily enhanced with more training data. The solution highlights the utility of hypergraph approaches for text classification as well as their applicability to real-world datasets with complicated interactions between text parts. This type of analysis will empower the business people will improve the quality of the product.

Text mining for disease surveillance in veterinary clinical data: part two, training computers to identify features in clinical text.

In part two of this mini-series, we evaluate the range of machine-learning tools now available for application to veterinary clinical text-mining. These tools will be vital to automate extraction of information from large datasets of veterinary clinical narratives curated by projects such as the Small Animal Veterinary Surveillance Network (SAVSNET) and VetCompass, where volumes of millions of records preclude reading records and the complexities of clinical notes limit usefulness of more "traditional" text-mining approaches. We discuss the application of various machine learning techniques ranging from simple models for identifying words and phrases with similar meanings to expand lexicons for keyword searching, to the use of more complex language models. Specifically, we describe the use of language models for record annotation, unsupervised approaches for identifying topics within large datasets, and discuss more recent developments in the area of generative models (such as ChatGPT). As these models become increasingly complex it is pertinent that researchers and clinicians work together to ensure that the outputs of these models are explainable in order to instill confidence in any conclusions drawn from them.

A text guided cross modal joint inpainting algorithm for ancient murals

Existing deep learning algorithms for ancient mural inpainting only consider prior information of mural images to guide the inpainting of damaged areas, lacking guidance from textual semantic information, which often leads to problems such as inconsistent semantics and disordered textures in the inpainting results. In this work, we propose a text guided cross modal joint inpainting algorithm for ancient murals. Firstly, a cross modal joint inpainting network for ancient murals was constructed, which introduced text semantic information as the control guidance for mural inpainting, providing contextual information for mural image inpainting and guiding the inpainting of damaged areas. Then, design a text encoder and an image encoder to extract text and image features respectively, including word and sentence features. Input the sentence features into the conditional enhancement module, resample the text sentences, and solve the problem of text sparsity. Secondly, design a text decoupling module that utilizes a multimodal attention mechanism to extract the semantic information of the missing parts of damaged murals, and design a text semantic enhancement module to improve the semantic representation ability of the text, achieve the repair of local details of murals, and solve the problem of texture disorder. Finally, the reconstruction and repair of murals are completed through a residual discriminator game and a spectral normalized Markov dual discriminator against the mural generation network. Through inpainting experiments on real Dunhuang murals, it has been shown that the proposed method is more effective than comparative methods in repairing damaged murals, achieving better visual perception and coordination in reconstruction, and outperforming comparative algorithms in both subjective and objective evaluations.

Specialized Knowledge, Text-Based Discussion, and Preparing Elementary Teachers for Disciplinary Literacy in English Language Arts Classrooms

ABSTRACT Text-based discussion is a form of disciplinary literacy that engages students in collaboratively building understanding of texts. It is a form of instruction that benefits students’ reading comprehension, but is rarely used in classrooms, despite its potential. This rarity may be because facilitating discussion requires teachers to have specialized knowledge of text. Specifically, teachers must understand how authors use discipline-specific text features and language patterns to convey meaning and how readers notice and evaluate these aspects of texts as they build textual understandings. This mixed methods study adopts a convergent-parallel design to investigate the effects of a 14-week intervention on pre-service teachers’ (PSTs) acquisition of specialized knowledge and its application to text-based discussions conducted with students in field placements. Findings suggest that the intervention improved PSTs’ specialized knowledge and that teachers felt more prepared for their discussions after applying their knowledge when planning discussions. However, even at the end of the intervention, PSTs continued to demonstrate gaps in their specialized knowledge, which influenced their ability to lead text-based discussions. The findings have implications for teacher educators attempting to prepare novice teachers for text-based discussions.