Textual Information Research Articles

Denoising Alignment with Large Language Model for Recommendation

The mainstream approach of GNN-based recommendation aggregates high-order ID information associated with the node in the user-item graph. The aggregation pattern using ID as signal has two disadvantages: lack of textual semantics and the impact of interaction noise. These disadvantages pose a threat to effectively learn user preferences, especially in capturing intricate user-item semantic relationships. Although large language models (LLMs) allow the integration of rich textual information into recommenders and have had groundbreaking applications in recommender systems, current works need to bridge the gap between different representation spaces. This is because LLMs-based methods align the representations of GNN-based models only by using text embedding of LLM, leading to unsatisfactory results. To address this challenge, we propose a D enoising A lignment framework with L LMs for GNN-based R ecommenders (DALR), which aims to align structural representation with textual representation and mitigate the effects of noise. Specifically, We propose a modeling framework that integrates the representation of graph structure with textual information from LLMs to capture intricate user-item interactions. We also suggest an alignment paradigm to enhance representation performance by aligning semantic signals from LLMs and structural features from GNN models. Additionally, we introduce a contrastive learning scheme to relieve the impact of noise and improve model performance. Extensive experiments on public datasets demonstrate that our model consistently outperforms the state-of-the-art methods. DALR achieves improvements ranging from 2.82% to 12.20% in Recall@5 and from 1.04% to 3.48% in NDCG@5 compared to the strongest baseline model, using the Steam dataset as an example.

Early diagnosis of Alzheimer's Disease based on multi-attention mechanism.

Alzheimer's Disease is a neurodegenerative disorder, and one of its common and prominent early symptoms is language impairment. Therefore, early diagnosis of Alzheimer's Disease through speech and text information is of significant importance. However, the multimodal data is often complex and inconsistent, which leads to inadequate feature extraction. To address the problem, We propose a model for early diagnosis of Alzheimer's Disease based on multimodal attention(EDAMM). Specifically, we first evaluate and select three optimal feature extraction methods, Wav2Vec2.0, TF-IDF and Word2Vec, to extract acoustic and linguistic features. Next, by leveraging self-attention mechanism and cross-modal attention mechanisms, we generate fused features to enhance and capture the inter-modal correlation information. Finally, we concatenate the multimodal features into a composite feature vector and employ a Neural Network(NN) classifier to diagnose Alzheimer's Disease. To evaluate EDAMM, we perform experiments on two public datasets, i.e., NCMMSC2021 and ADReSSo. The results show that EDAMM improves the performance of Alzheimer's Disease diagnosis over state-of-the-art baseline approaches on both datasets.

FuseLinker: Leveraging LLM’s pre-trained text embeddings and domain knowledge to enhance GNN-based link prediction on biomedical knowledge graphs

ObjectiveTo develop the FuseLinker, a novel link prediction framework for biomedical knowledge graphs (BKGs), which fully exploits the graph’s structural, textual and domain knowledge information. We evaluated the utility of FuseLinker in the graph-based drug repurposing task through detailed case studies. MethodsFuseLinker leverages fused pre-trained text embedding and domain knowledge embedding to enhance the graph neural network (GNN)-based link prediction model tailored for BKGs. This framework includes three parts: a) obtain text embeddings for BKGs using embedding-visible large language models (LLMs), b) learn the representations of medical ontology as domain knowledge information by employing the Poincaré graph embedding method, and c) fuse these embeddings and further learn the graph structure representations of BKGs by applying a GNN-based link prediction model. We evaluated FuseLinker against traditional knowledge graph embedding models and a conventional GNN-based link prediction model across four public BKG datasets. Additionally, we examined the impact of using different embedding-visible LLMs on FuseLinker’s performance. Finally, we investigated FuseLinker’s ability to generate medical hypotheses through two drug repurposing case studies for Sorafenib and Parkinson’s disease. ResultsBy comparing FuseLinker with baseline models on four BKGs, our method demonstrates superior performance. The Mean Reciprocal Rank (MRR) and Area Under receiver operating characteristic Curve (AUROC) for KEGG50k, Hetionet, SuppKG and ADInt are 0.969 and 0.987, 0.548 and 0.903, 0.739 and 0.928, and 0.831 and 0.890, respectively. ConclusionOur study demonstrates that FuseLinker is an effective novel link prediction framework that integrates multiple graph information and shows significant potential for practical applications in biomedical and clinical tasks. Source code and data are available at https://github.com/YKXia0/FuseLinker.

Consistency Prototype Module and Motion Compensation for few-shot action recognition (CLIP-CPM[formula omitted]C)

Recently, few-shot action recognition has progressed significantly, as it has learned the feature discriminability and designed suitable comparison methods. Still, there are the following restrictions. (a) Previous works are mainly based on visual mono-modal. Although some multi-modal works use labels as supplementary to construct prototypes of support videos, they cannot use this information for query videos. The labels are not used efficiently. (b) Most of the works ignore the motion feature of video, although the motion features are essential for distinguishing. We proposed a Consistency Prototype and Motion Compensation Network(CLIP-CPM2C) based on CLIP to address these issues. Firstly, in order to make the amount of information between the prototype and the query more similar, we propose a novel method to compensate for the text(prompt) information of query videos when text(prompt) does not exist, which depends on a Consistency Loss. Secondly, we use the differential features of the adjacent frames in two directions as the motion features, which explicitly embeds the network with motion dynamics, and then apply the Consistency Loss to the motion features. Extensive experiments on standard benchmark datasets demonstrate that the proposed method can compete with state-of-the-art results. Our code is available at the URL: https://github.com/cofly2014/CPM-MC.git.

Harnessing text information for enhanced hurricane resilience and public engagement: Unveiling disaster perspectives through social media

The recent disaster by hurricane Otis in Acapulco, Mexico, exemplifies how natural disasters, particularly hurricanes, can often arrive with a force that defies prediction despite technological advancements. Their impact, coupled with secondary disasters, underscores the urgency for efficient preemptive measures. Resilience research in the face of such calamities necessitates the creation of a safeguard before incurring irreversible losses. This research leverages big data, explicitly harnessing information obtained from social media about hurricanes, to bolster the efficacy of hurricane response efforts. The focus lies in amplifying the dissemination of adequate warnings and enhancing rescue operations. Concurrently, the research aims to delve into potential disaster rescue suggestions by tracking and analyzing high-interaction discussions on social media — X. This exploration seeks to unveil public focal points during hurricane disasters and identify valuable perspectives for future disaster response strategies. The research findings show that dynamic public discussions on X failed to adequately focus on crucial aspects such as hurricane rescue and warnings. Instead, certain discussions tended to target specific groups and individuals. By unveiling these trends, this research highlights the need to develop a nuanced understanding of interpreting and engaging with social media discussions during significant hurricane events.

Evaluation of Patient Information Texts About Bruxism on the Internet in Terms of Readability and Content

Aim: Bruxism is a common health problem in our society. The treatment method to be preferred should be determined by the decision of the patient and the physician together. Individuals' knowledge of the course of the disease, symptoms, and treatment options will contribute to the treatment process. There are many websites on the Internet that can guide patients about health and illness. The aim of this study is to evaluate the patient information texts about bruxism on web sites in terms of content and readability. Material and Method: Websites reached in a search using the keyword "bruxism" on Google were evaluated in terms of readability and content. The readability assessment was classified according to the level of education. Text content accuracy was also evaluated according to information level separately. Results: Forty-four of the 100 websites examined were included in the study. The average grade level is 11.41±0.82. The readability level of 79% (n=35) of the 44 texts examined is at a difficult level. It is seen that the definition and etiology of the disease are mentioned in 68% (n=30) of the texts examined in the study. Conclusion: Although the content of the English texts prepared to inform patients about bruxism on the Internet is sufficient, it has been found that the readability level is low.

Verb Form Recognition and Error Detection in English Articles Using Long Short-Term Memory and Grammar Checks

Error checking of verb forms in English articles is beneficial for learning English and improving the fluency of English texts. In this study, long shortterm memory (LSTM) was used to recognize the types of errors in verb forms. To maximize the utilization of textual context information, a bidirectional LSTM algorithm was employed. Simulation experiments were then conducted, and the algorithm was evaluated against the support vector machine (SVM) algorithm and the grammar rules-based algorithm. The bidirectional LSTM method demonstrated higher accuracy in recognizing the parts of speech of words and the types of verb form errors in the text. Additionally, the accuracy was more stable when faced with different types of verb form errors.

Multiplex graph aggregation and feature refinement for unsupervised incomplete multimodal emotion recognition

Multimodal Emotion Recognition (MER) involves integrating information of various modalities, including audio, visual, text and physiological signals, to comprehensively grasp human sentiments, which has emerged as a vibrant area within human–computer interaction. Researchers have developed many methods for this task, but many of these methods rely on labeled supervised learning and struggle to address the issue of missing some modalities of data. To address these issues, we propose a Multiplex Graph Aggregation and Feature Refinement framework for unsupervised incomplete MER, comprising four modules: Completion, Aggregation, Refinement, and Embedding. Specifically, we first capture the correlation information between samples using the graph structures, which aids in the completion of missing data and the multiplex aggregation of multimodal data. Then, we perform refinement operations on the aggregated features as well as alignment and enhancement operations on the embedding features to obtain the fused feature representations, which are consistent, highly separable and conducive to emotion recognition. Experimental results on multimodal emotion recognition datasets demonstrate that our method achieves state-of-the-art performance among unsupervised methods, validating its effectiveness.

An implicit aspect-based sentiment analysis method using supervised contrastive learning and knowledge embedding

Aspect-based sentiment analysis aims to analyze and understand people’s opinions from different aspects. Some comments do not contain explicit opinion words but still convey a clear human-perceived emotional orientation, which is known as implicit sentiment. Most previous research relies on contextual information from a text for implicit aspect-based sentiment analysis. However, little work has integrated external knowledge with contextual information. This paper proposes an implicit aspect-based sentiment analysis model combining supervised contrastive learning with knowledge-enhanced fine-tuning on BERT (BERT-SCL+KEFT). In the pre-training phase, the model utilizes supervised contrastive learning (SCL) on large-scale sentiment-annotated corpora to acquire sentiment knowledge. In the fine-tuning phase, the model uses a knowledge-enhanced fine-tuning (KEFT) method to capture explicit and implicit aspect-based sentiments. Specifically, the model utilizes knowledge embedding to embed external general knowledge information into textual entities by using knowledge graphs, enriching textual information. Finally, the model combines external knowledge and contextual features to predict the implicit sentiment in a text. The experimental results demonstrate that the proposed BERT-SCL+KEFT model outperforms other baselines on the general implicit sentiment analysis and implicit aspect-based sentiment analysis tasks. In addition, ablation experimental results show that the proposed BERT-SCL+KEFT model without the knowledge embedding module or supervised contrastive learning module significantly decreases performance, indicating the importance of these modules. All experiments validate that the proposed BERT-SCL+KEFT model effectively achieves implicit aspect-based sentiment classification.

Media attention, negative reports and textual information of banks’ ESG disclosure

This paper applies text analysis method and Baidu sentiment tendency analysis model to identify the text characteristics of bank ESG information disclosure, and explores the impact of media attention and negative reports on the text of bank ESG as well as the influence mechanism. The study finds that the higher the media attention, the stronger the positive sentiment of bank ESG information disclosure text, and the lower the readability. Similarly, negative media coverage is positively related to the sentiment of bank ESG text, and negatively related to the readability of the text. Mechanism analysis shows that media attention and negative reports can inhibit banks' active risk-taking, increase analysts' attention to banks as well, and then affect banks' ESG disclosure behavior. Heterogeneity analysis shows that the impact of media attention and negative reports on ESG disclosure text is more obvious for banks with higher capital adequacy ratio and lower weighted risk assets ratio. This paper examines the impact of media governance on bank ESG disclosure behavior from the perspective of text big data analysis, which is a useful supplement to the research on ESG disclosure.

Structural link prediction model with multi-view text semantic feature extraction

The exponential expansion of information has made text feature extraction based on simple semantic information insufficient for the multidimensional recognition of textual data. In this study, we construct a text semantic structure graph based on various perspectives and introduce weight coefficients and node clustering coefficients of co-occurrence granularity to enhance the link prediction model, in order to comprehensively capture the structural information of the text. Firstly, we jointly build the semantic structure graph based on three proposed perspectives (i.e., scene semantics, text weight, and graph structure), and propose a candidate keyword set in conjunction with an information probability retrieval model. Subsequently, we propose weight coefficients of co-occurrence granularity and node clustering coefficients to improve the link prediction model based on the semantic structure graph, enabling a more comprehensive acquisition of textual structural information. Experimental results demonstrate that our research method can reveal potential correlations and obtain more complete semantic structure information, while the WPAA evaluation index validates the effectiveness of our model.

Handwritten text recognition and information extraction from ancient manuscripts using deep convolutional and recurrent neural network

Handwritten text recognition and information extraction from ancient manuscripts using deep convolutional and recurrent neural network

Capturing invisible health threats through augmented reality-based public health messaging: focusing on perceived realism, issue involvement, and message attitudes

ABSTRACT Augmented reality (AR) has been extensively applied to health science education. However, a question still remains as to the optimal way to design AR interfaces to enhance users’ understanding of information that is presented in text format. Utilising the context of indoor air pollution, the current study examines the effects of messages about particulate matter (PM) delivered using different modes of AR visualisation (i.e. overlaying AR visuals onto users’ faces [AR-overlay self] vs. overlaying AR visuals onto users’ surroundings [AR-overlay environment] vs. no use of AR [AR control]). A between-subjects online experiment (N = 306) demonstrated that the AR-overlay environment has a pronounced effect on perceived realism toward 3D-rendered PM, which induced powerful effects on increasing both issue involvement and positive attitudes toward the textual information being presented. Compared to the AR control, the AR-overlay environment demonstrated a serial and positive mediation effect on perceived realism, which subsequently influenced issue involvement and message attitudes. However, overlaying 3D-rendered PM onto one’s face turned out to be a less effective means of eliciting perceived realism. Implications are discussed in the context of how to design effective AR interfaces that can best help users understand health science information.

IAT/ML: a metamodel and modelling approach for discourse analysis

Language technologies are gaining momentum as textual information saturates social networks and media outlets, compounded by the growing role of fake news and disinformation. In this context, approaches to represent and analyse public speeches, news releases, social media posts and other types of discourses are becoming crucial. Although there is a large body of literature on text-based machine learning, it tends to focus on lexical and syntactical issues rather than semantic or pragmatic. Being useful, these advances cannot tackle the nuanced and highly context-dependent problems of discourse evaluation that society demands. In this paper, we present IAT/ML, a metamodel and modelling approach to represent and analyse discourses. IAT/ML focuses on semantic and pragmatic issues, thus tackling a little researched area in language technologies. It does so by combining three different modelling approaches: ontological, which focuses on what the discourse is about; argumentation, which deals with how the text justifies what it says; and agency, which provides insights into the speakers’ beliefs, desires and intentions. Together, these three modelling approaches make IAT/ML a comprehensive solution to represent and analyse complex discourses towards their understanding, evaluation and fact checking.

Gamelan Recordings As an Archive of the Gamelan Music Movement in Bali | Rekaman Gamelan sebagai Arsip Pergerakan Musik Gamelan di Bali

This article reviews the role of Gamelan recordings as archives in documenting the movement of Gamelan music in Bali. The research was conducted through observation and interviews to understand the practices and outcomes of using recordings in the Balinese cultural context. Through direct observation and interviews, this study explored various developments in playing techniques, styles, and composition forms. Recording is a process of capturing sound or images stored in digital or analogue format, serving as archives for work documentation, reproduction, and distribution. An archive consists of various documents, records, or other items stored for historical or administrative purposes. Archives may include written, audio, visual, or digital recordings encapsulating essential or valuable information in specific texts and contexts such as history, culture, and law. Thus, combining research methods has yielded a profound understanding of how recordings function as archives of the Gamelan music movement in Bali and have added value to the preservation and development of traditional culture. Furthermore, this article provides deep insights into how Gamelan recordings can be valuable in understanding the evolution of traditional Balinese music.

Multi-modal transformer with language modality distillation for early pedestrian action anticipation

Language-vision integration has become an increasingly popular research direction within the computer vision field. In recent years, there has been a growing recognition of the importance of incorporating linguistic information into visual tasks, particularly in domains such as action anticipation. This integration allows anticipation models to leverage textual descriptions to gain deeper contextual understanding, leading to more accurate predictions. In this work, we focus on pedestrian action anticipation, where the objective is the early prediction of pedestrians’ future actions in urban environments. Our method relies on a multi-modal transformer model that encodes past observations and produces predictions at different anticipation times, employing a learned mask technique to filter out redundancy in the observed frames. Instead of relying solely on visual cues extracted from images or videos, we explore the impact of integrating textual information in enriching the input modalities of our pedestrian action anticipation model. We investigate various techniques for generating descriptive captions corresponding to input images, aiming to enhance the anticipation performance. Evaluation results on available public benchmarks demonstrate the effectiveness of our method in improving the prediction performance at different anticipation times compared to previous works. Additionally, incorporating the language modality in our anticipation model proved significant improvement, reaching a 29.5% increase in the F1 score at 1-second anticipation and a 16.66% increase at 4-second anticipation. These results underscore the potential of language-vision integration in advancing pedestrian action anticipation in complex urban environments.

Review of machine learning with sentimental analysis method for cross-model stock price prediction

Stock market prediction has been a popular area of research for years. Machine learning, as a fast-developing popular algorithm, is applied to stock market prediction by many previous researchers to better solve the time series involving problems over the changing prices. Different from traditional machine learning algorithms that focus on stock prices only, this paper gives a brief description and review of stock price predictions models that contain sentimental analysis over the recent paper works. Different from the prices in number format, sentimental analysis is more based on textual information extraction and mining, converting them into usable input to pass to a prediction model. This extends the prediction model's takeable input domain and strengthens the accuracy. To better classify the differences between the models, discussion and introduction are given based on different model types about whether they are traditional type or deep neural network embedded. Even though traditional types of models are more popular for sentimental analysis and neural networks perform better in prediction tasks, traditional methods are relatively easy to build or train with more explainability, compared to deep learning models suitable for larger data sets.

MBJELEL: An End-to-End Knowledge Graph Entity Linking Method Applied to Civil Aviation Emergencies

Aviation emergency management is playing a more and more important role in the aviation field. How to make effective use of massive heterogeneous and multi-source aviation accident knowledge has become a great challenge for aviation emergency management. Aiming at the problems such as too long physical length, mixed and composite entities, similar character of domain entity names, information difference between entities, separation of codes between entities, coding errors during transmission, etc., the construction method of knowledge map of civil aviation emergencies is studied. In previous research methods, entity link is always divided into two parts, that is, first detection and then disambiguation, which makes the mentioned entity and the candidate entity are encoded separately, and there is error transmission between the two parts, modules cannot communicate with each other, and the close association between entities cannot be well learned. In this paper, we proposed an end-to-end entity linking method based on two-layer BiLSTM model joint coding vectorize each word of civil aviation text information, and then concatenate feature vectors into two-layer BiLSTM model to obtain high-level context representation. Because the joint encoding of boundary information can reduce the error transmission, information is exchanged between candidate entities during the initial encoding to enhance the closeness between candidate entities and candidate entities. The experimental results show that compared with other sota models, the F1 value of the proposed model reaches 88.97%.

Problems of extracting semi-structured textual information based on Text Mining technology (using the material of the Russian and Chuvash languages)

Problems of extracting semi-structured textual information based on Text Mining technology (using the material of the Russian and Chuvash languages)

A design for neural network model of continuous reading

Cognition and learning are exceedingly modeled as an associative activity of connectionist neural networks. However, only a few such models exist for continuous reading, which involves the delicate coordination of word recognition and eye movements. Moreover, these models are limited to only orthographic level of word processing with predetermined lexicons. Here, we present a conceptual design of a developmentally plausible neural network model of reading designed to simulate word learning, parafoveal preview activation of words, their later foveal word recognition including phonological decoding, and forward saccade length as a control mechanism for intake of new textual information. We will discuss the theoretical advancements of the design and avenues for future developments.