Text-based Methods Research Articles

Unravelling the knowledge matrix: exploring knowledge-sharing behaviours on market-based platforms using regression tree analysis

PurposeThis study aims to enhance the effectiveness of knowledge markets and overall knowledge management (KM) practices within organisations. By addressing the challenge of internal knowledge stickiness, it seeks to demonstrate how machine learning and AI approaches, specifically a text-based AI method for personality assessment and regression trees for behavioural analysis, can automate and personalise knowledge market incentivisation mechanisms.Design/methodology/approachThe research employs a novel approach by integrating machine learning methodologies to overcome the limitations of traditional statistical methods. A natural language processing (NLP)-based AI tool is used to assess employees’ personalities, and regression tree analysis is applied to predict and categorise behavioural patterns in knowledge-sharing contexts. This approach is designed to capture the complex interplay between individual personality traits and environmental factors, which traditional methods often fail to adequately address.FindingsCognitive style was confirmed as a key predictor of knowledge-sharing, with extrinsic motivators outweighing intrinsic ones in market-based platforms. These findings underscore the significance of diverse combinations of environmental and individual factors in promoting knowledge sharing, offering key insights that can inform the automatic design of personalised interventions for community managers of such platforms.Originality/valueThis research stands out as it is the first to empirically explore the interaction between the individual and the environment in shaping actual knowledge-sharing behaviours, using advanced methodologies. The increased automation in the process extends the practical contribution of this study, enabling a more efficient, automated assessment process, and thus making critical theoretical and practical advancements in understanding and enhancing knowledge-sharing behaviours.

Donning and doffing training using augmented reality and text-based educational methods.

Donning and doffing training using augmented reality and text-based educational methods.

Clearing Up Confusion: The Effect of Outlier Similarity on IPO Underpricing

Prior research demonstrates that audiences tend to converge in their valuations of firms similar to preexisting category prototypes or exemplars. Much less is known of the influence of salient outliers, specific firms that receive market-wide attention due to their extreme, ambiguous performance, on audiences’ valuations. We argue that outlier similarity, by contrast with prototype similarity, leads to divergent valuations among individual investors. We explore this insight in the context of initial public offerings (IPOs). In this context, converging valuations among investors lead to limited information asymmetry concerns and hence reduced underpricing on the first day of trading of an issuing firm. Hence, we expect that prototype similarity leads to lower underpricing while outlier similarity leads to higher underpricing. We test our hypotheses using a sample of 2,488 U.S. IPOs from 1996 to 2015, measuring prototype and outlier similarity through a natural language processing technique applied to nearly 160,000 financial documents. We find that in low-tech industries, where prototypes are informative about category members, prototype similarity reduces underpricing, but not in high-tech industries. Additionally, we find that outlier similarity increases underpricing, especially for more recent outliers. This paper contributes to the literature on market valuation and market categories, and advances research on meaning and culture using new text-based computational methods.

A Empirical Study on Critical Thinking Ability in College English Reading Teaching under CBI

In the context of the new liberal arts, the ordinary English teaching model is difficult to adapt to the development of college English teaching, and English reading, as a necessary part of cultivating students’ comprehensive language ability and critical thinking ability, has played an important role in college English teaching. As a text-based teaching method, Content-Based Instruction (CBI) is more in line with the learning of English reading. Based on Krashen’s input hypothesis and Swain’s output hypothesis theory, it selected 66 English major seniors from Y University as the research subjects, who subsequently are divided into control classes and experimental classes. Data was collected by survey questionnaires and reading tests. Based on the analysis and discussion of relevant data, it concludes that: firstly, the critical thinking ability of college students is at a moderate level. Secondly, the CBI teaching method is significantly positively correlated with students’ critical thinking ability. Thirdly, the CBI teaching method has a positive impact on students’ critical English reading skills.

SketchDream: Sketch-based Text-To-3D Generation and Editing

Existing text-based 3D generation methods generate attractive results but lack detailed geometry control. Sketches, known for their conciseness and expressiveness, have contributed to intuitive 3D modeling but are confined to producing texture-less mesh models within predefined categories. Integrating sketch and text simultaneously for 3D generation promises enhanced control over geometry and appearance but faces challenges from 2D-to-3D translation ambiguity and multi-modal condition integration. Moreover, further editing of 3D models in arbitrary views will give users more freedom to customize their models. However, it is difficult to achieve high generation quality, preserve unedited regions, and manage proper interactions between shape components. To solve the above issues, we propose a text-driven 3D content generation and editing method, SketchDream, which supports NeRF generation from given hand-drawn sketches and achieves free-view sketch-based local editing. To tackle the 2D-to-3D ambiguity challenge, we introduce a sketch-based multi-view image generation diffusion model, which leverages depth guidance to establish spatial correspondence. A 3D ControlNet with a 3D attention module is utilized to control multi-view images and ensure their 3D consistency. To support local editing, we further propose a coarse-to-fine editing approach: the coarse phase analyzes component interactions and provides 3D masks to label edited regions, while the fine stage generates realistic results with refined details by local enhancement. Extensive experiments validate that our method generates higher-quality results compared with a combination of 2D ControlNet and image-to-3D generation techniques and achieves detailed control compared with existing diffusion-based 3D editing approaches.

Enhancing text-based knowledge graph completion with zero-shot large language models: A focus on semantic enhancement

The design and development of text-based knowledge graph completion (KGC) methods leveraging textual entity descriptions are at the forefront of research. These methods involve advanced optimization techniques such as soft prompts and contrastive learning to enhance KGC models. The effectiveness of text-based methods largely hinges on the quality and richness of the training data. Large language models (LLMs) can utilize straightforward prompts to alter text data, thereby enabling data augmentation for KGC. Nevertheless, LLMs typically demand substantial computational resources. To address these issues, we introduce a framework termed constrained prompts for KGC (CP-KGC). This CP-KGC framework designs prompts that adapt to different datasets to enhance semantic richness. Additionally, CP-KGC employs a context constraint strategy to effectively identify polysemous entities within KGC datasets. Through extensive experimentation, we have verified the effectiveness of this framework. Even after quantization, the LLM (Qwen-7B-Chat-int4) still enhances the performance of text-based KGC methods.11Code and datasets are available at https://github.com/sjlmg/CP-KGC. This study extends the performance limits of existing models and promotes further integration of KGC with LLMs.

A review of deep learning based malware detection techniques

With the popularization of computer technology, the number of malware has increased dramatically in recent years. Some malware can threaten the network security of users by downloading and installing, and even spreading widely on the Internet, causing consequences such as private data leakage in the operating system, extortion, and network paralysis. In order to deal with these threats, researchers analyze malicious samples through various analysis techniques, which are usually divided into static and dynamic analysis based on the principle of whether the code needs to be executed or not. This paper analyzes in detail several classical methods of feature extraction in malware detection techniques. With the technological development of artificial intelligence, deep learning is gradually being introduced into malware detection, which does not require the identification of professional security personnel and greatly improves the generalization ability of detection. In the paper, text-based detection methods, image visualization-based detection, and graph structure-based detection techniques are reviewed according to different feature extraction methods. In addition, the paper compares 26 datasets that have been commonly used in recent years applied in the research field and explains the main contents and specifications of the datasets. Finally, a summary and outlook of the malware research field is given.

Engineering Approach to Explore Language Reflecting Well-Being

Although well-being is helpful in measuring the state of society from various perspectives, past research has been limited to (1) questionnaire surveys, which make it difficult to target a large number of people, and (2) the major indices focus on individual factors and do not incorporate group factors. To tackle these issues, we collected daily reports from the company employees that included text, their individual subjective well-being, and team subjective well-being. By using the collected data, we constructed a well-being estimation model based on the Large Language Model and examined an indicator called ``sharedness index'', as a state of the team that influences an individual well-being, measured using both score- and text-based methods.

Emotional Experiences in Online, Text-Based Tabletop Role-Playing Games

Background This study explores the emotional experiences of players in online, text-based tabletop role-playing games (TTRPGs) and their potential impact on players. The goal is to understand the therapeutic and educational possibilities arising from these emotional experiences. Methods Through qualitative analysis, themes were identified based on participants' narratives. Eight participants who engaged in online, text-based TTRPGs were interviewed to gather data on their emotional experiences and effects outside the game. Result The study identified three main categories of emotions experienced by players: emotions related to the community, emotions associated with the game world and characters, and emotions linked to game mechanics. These emotions were found to impact players beyond the gaming environment, influenced by the players' emotional immersion. Additionally, the study revealed two distinct approaches to immersion among players and highlighted emotions as motivating factors for gameplay. Discussion The findings emphasize the role of TTRPGs in providing creative outlets, social interaction, and recreation, facilitating skill development, and broadening perspectives based on players' emotional experiences. Additionally, this study highlights the potential for immersion in online, text-based TTRPGs linked to therapeutic and educational uses. Limitations This study employs an in-depth phenomenological qualitative interview method. While this approach yields detailed and nuanced data, it may not provide extensive generalizability. Future research endeavors could benefit from more significant and diverse participant pools to enhance the understanding of emotional experiences in TTRPG players. Furthermore, exploring additional factors such as genres, narrative structures, and player-character relationships could offer more profound insights. Conclusion This study demonstrates that online, text-based TTRPGs allow players to experience various emotions with implications beyond the game's boundaries. The findings suggest that interventions based on role-playing activities may benefit from the online text-based method due to its accessibility and lower barriers to entry.

Convolutional Neural Networks-Based For Predicting Aerodynamic Coefficient Of Airfoils At Ultra-Low Reynolds Number

Many applications, including airplane design, wind turbines, and heat transmission, use symmetric or asymmetric airfoils. Engineers employ these airfoil shapes to optimize performance and efficiency. Each airfoil has a unique set of aerodynamic coefficients that must be calculated to maximize the airfoil design. Engineers utilize numerous ways to calculate coefficients, such as lift and drag. One of the methods is the prediction method, which effectively reduces time and cost. This study's training dataset is obtained from particle-based numerical computation using the Lattice Boltzmann Method (LBM). Then, Convolutional Neural Networks (CNN) are used as a prediction method to get the aerodynamic coefficients of airfoils for lift and drag based on two different Reynolds numbers. In CNN, airfoil geometry representation is essential. The Signed Distance Function (SDF) was used to convert airfoil geometry into RGB pictures. On the other hand, the SDF method cannot explain different flow conditions; in this case, it is represented by the Reynolds number (Re). Therefore, we propose a Text-based Watermarking Method (TWM) to differentiate between Re = 500 and Re = 1000. Each airfoil representation was trained and tested to generate each prediction model using a modified LeNet-5. The computation results show that using CNN with TWM on SDF to define the Reynolds numbers could predict the lift and drag coefficients with varying angles of attack. Future research can focus on generalizations to different aerodynamic aspects and practical applications in complex scenarios.

Assessing Training Methods for Advanced Driver Assistance Systems and Autonomous Vehicle Functions: Impact on User Mental Models and Performance

Understanding the complexities of Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicle (AV) technologies is critical for road safety, especially concerning their adoption by drivers. Effective training is a crucial element in ensuring the safe and competent operation of these technologies. This study emphasises the critical role of training methodologies in shaping drivers’ mental models, defined as an individual’s cognitive frameworks for understanding and interacting with ADAS and AV systems. Their mental models substantially influence their interactions with those technologies. A comparative analysis of text-based and video-based training methods has been conducted to assess their influence on participants’ performance and the development of their mental models of ADAS and AV functionalities. Performance is evaluated in terms of the accuracy and reaction time of the participants as they interacted with ADAS and AV functions in a driving simulation. The findings reveal that video-based training yielded better performance outcomes, more accurate mental models, and a deeper understanding of ADAS functionalities among participants. These findings are crucial for policy makers, automotive manufacturers, and educational institutions involved in driver training. They underscore the necessity of developing tailored training programs to facilitate the proficient and safe operation of increasingly complex automotive technologies.

Foreground and background separated image style transfer with a single text condition

Traditional image-based style transfer requires additional reference style images, making it less user-friendly. Text-based methods are more convenient but suffer from issues like slow generation, unclear content, and poor quality. In this work, we propose a new style transfer method SA2-CS (means Semantic-Aware and Salient Attention CLIPStyler), which is based on the Comparative Language Image Pretraining (CLIP) model and a salient object detection network. Masks obtained from the salient object detection network are utilized to guide the style transfer process, and various strategies are employed to optimize according to different masks. Adequate experiments with diverse content images and style text descriptions were conducted, demonstrating our method's advantages: the network is easily trainable and converges rapidly; it achieves stable, superior generation results compared to other methods. Our approach addresses over-stylization issues in the foreground, enhances foreground-background contrast, and enables precise control over style transfer in various semantic regions.

Identification of Social Media Posts Containing Self-reported COVID-19 Symptoms using Triple Word Embeddings and Long Short-Term Memory

The COVID-19 pandemic has permeated the global sphere and influenced nearly all nations and regions. Common symptoms of this pandemic include fever, cough, fatigue, and loss of sense of smell. The impact of COVID-19 on public health and the economy has made it a significant global concern. It has caused economic contraction in Indonesia, particularly in face-to-face interaction and mobility sectors, such as transportation, warehousing, construction, and food and beverages. Since the pandemic began, Twitter users have shared symptoms in their tweets. However, they couldn't confirm their concerns due to testing limitations, reporting delays, and pre-registration requirements in healthcare. The classification of text from Twitter data about COVID-19 topics has predominantly focused on sentiment analysis regarding the pandemic or vaccination. Research on identifying COVID-19 symptoms through social media messages is limited in the literature. The main objective of this study is to identify symptoms using word embedding techniques and the LSTM algorithm. Various techniques such as Word2Vec, GloVe, FastText, and a composite approach are used. LSTM is used for classification, improving upon the RNN technique. Evaluation criteria include accuracy, precision, and recall. The model with an input dimension of 147x100 achieves the highest accuracy at 89%. This study aims to find the best LSTM model for detecting COVID-19 symptoms in social media tweets. It evaluates LSTM models with different word embedding techniques and input dimensions, providing insights into the optimal text-based method for COVID-19 detection through social media texts.

SimRE: Simple contrastive learning with soft logical rule for knowledge graph embedding

Knowledge graphs serve as a pivotal framework for the structured representation of information regarding entities and relations. However, in the real world, these knowledge graphs are often incomplete and harboring missing facts. Knowledge graph completion (KGC) has emerged as a central research focus, entailing the automated prediction of these missing facts and garnering substantial scholarly attention in recent years. Text-based knowledge graph embedding methods have demonstrated considerable potential for tackling the challenges associated with KGC by employing pre-trained language models. However, their limitation lies in the lack of logical features, which constrains the efficacy of capturing intricate patterns within knowledge graphs. This paper proposed SimRE, a straightforward contrastive learning framework augmented with soft logic rules. SimRE introduces a self-supervised framework that leverages the input rule bodies to predict the corresponding rule heads through a contrastive objective. We introduced two rule sampling techniques to enhance the efficiency and accuracy of the model: in-batch rule negatives and pre-batch rule negatives. SimRE employs a simple method for integrating logical features with the text-based model. The experimental results on benchmark datasets demonstrate that the proposed approach outperforms state-of-the-art methods.

HASHTAG-BASED TWEET EXPANSION FOR IMPROVED TOPIC MODELLING

In this paper author employing BILSTM, BERT and GRAPHCNN algorithms to predict topics from heterogeneous and homogenous tweets where heterogeneous refers to unrelated tweets and homogenous refers to related tweets. To predict topics from text many existing algorithms are available such as tweet aggregation, where related tweets are combined into a single document or 2) tweet expansion with related text from external sources. The first approach faces the problem of losing the topic distribution in individual tweets. While finding a relevant text from the external source for a random tweet in the second approach is challenging for various reasons like differences in writing styles, multilingual content, and informal text. In contrast to adding context from external resources or combining related tweets into a pool, this study uses the internal vocabulary (hashtags) to counter under-specificity and sparsity in tweets. Earlier studies have indicated hashtags to be an important feature for representing the underlying context present in the tweet. Sequential models like Bidirectional Long Short Term Memory (BiLSTM) and Convolution Neural Network (CNN) over distributed representation of words have shown promising results in capturing semantic relationships between words of a tweet in the past. Motivated by the above, this article proposes a unified framework of hashtag-based tweet expansion exploiting text-based and network-based representation learning methods such as BiLSTM, BERT, and Graph Convolution Network (GCN). The hashtag-based expanded tweets using the proposed framework have significantly improved topic modelling performance compared to un-expanded (raw) tweets and hashtag-poolingbased approaches over two real-world tweet datasets of different nature.Propose paper will expand or identify TWEETS from given Hashtags and then identify topics from expanded HashtagsBILSTM, GRAPHCNN and BERT are based on Text processing which can be used to efficiently process and predict topics from the tweets. It’s difficult to train all 3 algorithms so we are training BERT and BILSTM. To train above algorithms author is using tweets from various HASHTAGS and not publish those tweets on internet so we have collected some tweets from various hashtags and below are the dataset details.

Macroprudential Policies in the Light of the Development of Information Technologies: A Synthesis on the Effective Early Warning Signals

In response to recent recurrent crises, innovative macroprudential policies (MaPs) have been framed and implemented to address the weaknesses of market-led microprudential mechanisms and enhance the stabili­ty of financial systems. However, the effectiveness of the tools used to implement MaPs remains a critical research question. Early warning signals (EWS) serve as indicators of potential future crises. This paper explores approaches for identifying EWS to optimize the impact of MaPs, particularly in light of advances in information technology. It provides a comprehensive review of academic studies that identify effective EWS by analyzing numerical data through econometric and machine learning (ML) methods or by extracting economic insights from text using deep learning (DL) techniques — innovative methods for financial supervision and regu­lation. The findings, considering current regulatory practices, highlight the benefits of ML-based approaches for processing large sets of numerical data and the growing potential of text-based methods for assessing economic expectations.

Text-Guided Synthesis of Eulerian Cinemagraphs

We introduce Text2Cinemagraph, a fully automated method for creating cinemagraphs from text descriptions --- an especially challenging task when prompts feature imaginary elements and artistic styles, given the complexity of interpreting the semantics and motions of these images. We focus on cinemagraphs of fluid elements, such as flowing rivers, and drifting clouds, which exhibit continuous motion and repetitive textures. Existing single-image animation methods fall short on artistic inputs, and recent text-based video methods frequently introduce temporal inconsistencies, struggling to keep certain regions static. To address these challenges, we propose an idea of synthesizing image twins from a single text prompt --- a pair of an artistic image and its pixel-aligned corresponding natural-looking twin. While the artistic image depicts the style and appearance detailed in our text prompt, the realistic counterpart greatly simplifies layout and motion analysis. Leveraging existing natural image and video datasets, we can accurately segment the realistic image and predict plausible motion given the semantic information. The predicted motion can then be transferred to the artistic image to create the final cinemagraph. Our method outperforms existing approaches in creating cinemagraphs for natural landscapes as well as artistic and other-worldly scenes, as validated by automated metrics and user studies. Finally, we demonstrate two extensions: animating existing paintings and controlling motion directions using text.

AvatarStudio: Text-Driven Editing of 3D Dynamic Human Head Avatars

Capturing and editing full-head performances enables the creation of virtual characters with various applications such as extended reality and media production. The past few years witnessed a steep rise in the photorealism of human head avatars. Such avatars can be controlled through different input data modalities, including RGB, audio, depth, IMUs, and others. While these data modalities provide effective means of control, they mostly focus on editing the head movements such as the facial expressions, head pose, and/or camera viewpoint. In this paper, we propose AvatarStudio, a text-based method for editing the appearance of a dynamic full head avatar. Our approach builds on existing work to capture dynamic performances of human heads using Neural Radiance Field (NeRF) and edits this representation with a text-to-image diffusion model. Specifically, we introduce an optimization strategy for incorporating multiple keyframes representing different camera viewpoints and time stamps of a video performance into a single diffusion model. Using this personalized diffusion model, we edit the dynamic NeRF by introducing view-and-time-aware Score Distillation Sampling (VT-SDS) following a model-based guidance approach. Our method edits the full head in a canonical space and then propagates these edits to the remaining time steps via a pre-trained deformation network. We evaluate our method visually and numerically via a user study, and results show that our method outperforms existing approaches. Our experiments validate the design choices of our method and highlight that our edits are genuine, personalized, as well as 3D- and time-consistent.

Scalable Surveillance of E-Cigarette Products on Instagram and TikTok Using Computer Vision.

Instagram and TikTok, video-based social media platforms popular among adolescents, contain tobacco-related content despite the platforms' policies prohibiting substance-related posts. Prior research identified themes in e-cigarette-related social media posts using qualitative or text-based machine learning methods. We developed an image-based computer vision model to identify e-cigarette products in social media images and videos. We created a data set of 6999 Instagram images labeled for 8 object classes: mod or pod devices, e-juice containers, packaging boxes, nicotine warning labels, e-juice flavors, e-cigarette brand names, and smoke clouds. We trained a DyHead object detection model using a Swin-Large backbone, evaluated the model's performance on 20 Instagram and TikTok videos, and applied the model to 14 072 e-cigarette-related promotional TikTok videos (2019-2022; 10 276 485 frames). The model achieved the following mean average precision scores on the image test set: e-juice container: 0.89; pod device: 0.67; mod device: 0.54; packaging box: 0.84; nicotine warning label: 0.86; e-cigarette brand name: 0.71; e-juice flavor name: 0.89; and smoke cloud: 0.46. The prevalence of pod devices in promotional TikTok videos increased by 15% from 2019 to 2022. The prevalence of e-juices increased by 33% from 2021 to 2022. The prevalence of e-juice flavor names and e-cigarette brand names increased by about 100% from 2019 to 2022. Deep learning-based object detection technology enables automated analysis of visual posts on social media. Our computer vision model can detect the presence of e-cigarettes products in images and videos, providing valuable surveillance data for tobacco regulatory science (TRS). Prior research identified themes in e-cigarette-related social media posts using qualitative or text-based machine learning methods. We developed an image-based computer vision model to identify e-cigarette products in social media images and videos. We trained a DyHead object detection model using a Swin-Large backbone, evaluated the model's performance on 20 Instagram and TikTok videos featuring at least two e-cigarette objects, and applied the model to 14 072 e-cigarette-related promotional TikTok videos (2019-2022; 10 276 485 frames). The deep learning model can be used for automated, scalable surveillance of image- and video-based e-cigarette-related promotional content on social media, providing valuable data for TRS. Social media platforms could use computer vision to identify tobacco-related imagery and remove it promptly, which could reduce adolescents' exposure to tobacco content online.

Strengthening the Role of Indonesian Language in the 2013 Curriculum According to Text-Based Language Learning

Before the 2013 Curriculum was set by the government, Indonesian only played a role as a subject and a vehicle for communication in schools. This role appears to be increasing along with the shift in the Indonesian language learning paradigm through text-based learning. This research aims to explore the role of Indonesians in the 2013 Curriculum and the implications of its methodology for learning. The method used is descriptive qualitative using library documents. The research data is in the form of Indonesian language teaching materials for junior high schools sourced from books issued by the government. The research results show that the function of the Indonesian language is confirmed by its existence as a catalyst and vehicle for science, technology, and art (science and technology). The increase in the function of the Indonesian language is also in line with the application of text-based learning methods through four learning stages, namely building context, modeling text, building text together, and building text independently. In this context, Indonesian is at the forefront of absorbing science and technology; conveying information through ideas and opinions; and binding the diversity of languages, ethnicities, and religions in the form of complete diversity.