Informative Instances Research Articles

Distantly supervised relation extraction with a Meta-Relation enhanced Contrastive learning framework

Distantly supervised relation extraction employs the alignment of unstructured corpora with knowledge bases to automatically generate labeled data. This method, however, often introduces significant label noise. To address this, multi-instance learning has been widely utilized over the past decade, aiming to extract reliable features from a bag of sentences. Yet, multi-instance learning struggles to effectively distinguish between clean and noisy instances within a bag, thereby hindering the full utilization of informative instances and the reduction of the impact of incorrectly labeled instances. In this paper, we propose a new Meta-Relation enhanced Contrastive learning based method for distantly supervised Relation Extraction named MRConRE. Specifically, we generate a “meta relation pattern” (MRP) for each bag, based on its semantic content, to differentiate between clean and noisy instances. Noisy instances are then transformed into beneficial bag-level instances through relabeling. Subsequently, contrastive learning is employed to develop precise sentence representations, forming the overall representation of the bag. Finally, we utilize a mixup strategy to integrate bag-level information for model training. Our method’s effectiveness is validated through experiments on various benchmarks.

Neurolinguistic Priming and Gender Stereotype Effects in the Ratings of Justice vs. Authority Moral Violations: Republicans and Democrats

ABSTRACT An experimental method for assessing gender biases was used to compare Republicans’ and Democrats’ ratings of moral violations in the domains of justice vs. respect for authority. Four experimental conditions of a text-based survey instrument manipulated the gender of the protagonist and the location of the first instance of gender information in single-sentence moral violation vignettes. Results were consistent with the theoretical time course of neurolinguistic gender priming and the hypothesized influence of implicit stereotypes on moral judgments. Republicans demonstrated a gender bias in ratings of authority violations by rating violations committed by girls and women as worse when compared to a pronoun only condition. Democrats demonstrated the opposite bias by rating authority violations committed by boys and men as worse when compared to violations committed by girls and women. No significant bias was found for any of the justice violation conditions for either Republicans or Democrats.

The utility of ChatGPT in gender-affirming mastectomy education

IntroductionThe integration of AI such as ChatGPT in medicine has been showing promise in enhancing patient education. Gender-affirming mastectomy (GAM) is a surgical procedure designed to help individuals transition to their self-identified gender, playing a crucial role in mitigating psychological distress for many transmasculine and non-binary (TNB) patients. With increased demand and attention towards GAM, plastic and reconstructive surgeons may rely on AI-driven chatbots as an accessible, accurate, and patient-driven model for relevant details on this procedure. Specific Aim(s)This study aimed to assess the quality and readability of information provided by ChatGPT in response to frequently asked questions (FAQs) related to GAM. MethodsInspired by online forums and physician websites, 10 frequently asked questions (FAQs) about pre- and postoperative topics were submitted to ChatGPT and assessed using validated readability score measures and expert interpretation. ResultsThe average readability score was 16.0 ± 1, indicating a college or graduate reading level. Mean accuracy, comprehensiveness, and danger scores were 8.8 ± 0.5, 7.8 ± 0.7, and 2.2 ± 0.4, respectively. Although physicians appreciated ChatGPT's tone, patient autonomy, and advice to seek professional medical and mental help, they also cited instance of generic information, misinformation, support of debated techniques, and pathologization of gender dysphoria. ConclusionEven with its promise in providing accurate and comprehensive information on GAM, ChatGPT’s current limitations suggest caution as a supplementary tool to physician consultation.

Novel hippocampus-centered methodology for informative instance selection in Alzheimer's disease data

The quantity and quality of a dataset play a crucial role in the performance of prediction models. Increasing the amount of data increases the computational requirements and can introduce negligible variations, outliers, and noise. These significantly impact the model performance. Thus, instance selection techniques are crucial for building prediction models with informative data, reducing the dataset size, improving performance, and minimizing computational costs. This study proposed a novel methodology for identifying the most informative two-dimensional slices derived from magnetic resonance imaging (MRI) to study Alzheimer's disease. The efficacy of our methodology was attributable to a hippocampus-centered analysis using data from multiple atlases. The methodology was evaluated by constructing convolutional neural networks to identify Alzheimer's disease, using a consolidated dataset constructed from three standard datasets: Alzheimer's Disease Neuroimaging Initiative, Australian Imaging, Biomarker & Lifestyle Flagship Study of Ageing, and Open Access Series of Imaging Studies. The proposed methodology demonstrated a commendable subject-level classification accuracy of approximately (95.00%) when distinguishing between normal cognition and Alzheimer's.

How do people predict a random walk? Lessons for models of human cognition.

Repeated forecasts of changing values are common in many everyday tasks, from predicting the weather to financial markets. A particularly simple and informative instance of such fluctuating values are random walks: Sequences in which each point is a random movement from only its preceding value, unaffected by any previous points. Moreover, random walks often yield basic rational forecasting solutions in which predictions of new values should repeat the most recent value, and hence replicate the properties of the original series. In previous experiments, however, we have found that human forecasters do not adhere to this standard, showing systematic deviations from the properties of a random walk such as excessive volatility and extreme movements between subsequent predictions. We suggest that such deviations reflect general statistical signatures of cognition displayed across multiple tasks, offering a window into underlying mechanisms. Using these deviations as new criteria, we here explore several cognitive models of forecasting drawn from various approaches developed in the existing literature, including Bayesian, error-based learning, autoregressive, and sampling mechanisms. These models are contrasted with human data from two experiments to determine which best accounts for the particular statistical features displayed by participants. We find support for sampling models in both aggregate and individual fits, suggesting that these variations are attributable to the use of inherently stochastic prediction systems. We thus argue that variability in predictions is strongly influenced by computational noise within the decision making process, with less influence from "late" noise at the output stage. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

그린 투어리즘 관점에서 본 스페인의 ‘실버 루트’

This essay examines Spain's Silver Route, also known as the “Vía de la Plata”, from the standpoint of ecotourism. Originally used as a logistical route for the transfer of minerals, primarily gold, the Silver Route is now well-known as a pilgrimage path and attracts travelers from all over the world. From the standpoint of ecotourism and green tourism in particular, the Silver Route is a remarkable subject of study since it allows visitors to “walk with nature” and enjoy natural heritage. Furthermore, it is a noteworthy example of green tourism due to its affiliation with Caminos Naturales, the premier ecotourism initiative in Spain. The purpose of this study is to highlight how the Silver Route relates to the ideas of sustainable and green tourism. Above all, a group of municipal governments effectively oversees the Silver Route. Additionally, we discovered instances of community renewal and information and communication technology-based “carbon footprint” reduction initiatives. Enhancing sustainability and securing competitiveness as a tourist destination will be achievable by applying the Silver Route's benefits to Korea's travel sector.

MIST: Multi-instance selective transformer for histopathological subtype prediction

Accurate histopathological subtype prediction is clinically significant for cancer diagnosis and tumor microenvironment analysis. However, achieving accurate histopathological subtype prediction is a challenging task due to (1) instance-level discrimination of histopathological images, (2) low inter-class and large intra-class variances among histopathological images in their shape and chromatin texture, and (3) heterogeneous feature distribution over different images. In this paper, we formulate subtype prediction as fine-grained representation learning and propose a novel multi-instance selective transformer (MIST) framework, effectively achieving accurate histopathological subtype prediction. The proposed MIST designs an effective selective self-attention mechanism with multi-instance learning (MIL) and vision transformer (ViT) to adaptive identify informative instances for fine-grained representation. Innovatively, the MIST entrusts each instance with different contributions to the bag representation based on its interactions with instances and bags. Specifically, a SiT module with selective multi-head self-attention (S-MSA) is well-designed to identify the representative instances by modeling the instance-to-instance interactions. On the contrary, a MIFD module with the information bottleneck is proposed to learn the discriminative fine-grained representation for histopathological images by modeling instance-to-bag interactions with the selected instances. Substantial experiments on five clinical benchmarks demonstrate that the MIST achieves accurate histopathological subtype prediction and obtains state-of-the-art performance with an accuracy of 0.936. The MIST shows great potential to handle fine-grained medical image analysis, such as histopathological subtype prediction in clinical applications.

AI-driven anonymization: Protecting personal data privacy while leveraging machine learning

AbstractThe development of artificial intelligence has significantly transformed people's lives. However, it has also posed a significant threat to privacy and security, with numerous instances of personal information being exposed online and reports of criminal attacks and theft. Consequently, the need to achieve intelligent protection of personal information through machine learning algorithms has become a paramount concern. Artificial intelligence leverages advanced algorithms and technologies to effectively encrypt and anonymize personal data, enabling valuable data analysis and utilization while safeguarding privacy. This paper focuses on personal data privacy protection and the promotion of anonymity as its core research objectives. It achieves personal data privacy protection and detection through the use of machine learning's differential privacy protection algorithm. The paper also addresses existing challenges in machine learning related to privacy and personal data protection, offers improvement suggestions, and analyzes factors impacting datasets to enable timely personal data privacy detection and protection.

Impact of attention on rare events across industries in Indonesia

Rare events (RE) are substantial with significant impact but are difficult to predict, often deviating from regular expectations. These events trigger psychological reactions in the market and susceptible to irrational decisions that challenge logical assumptions. The rapidity of the crisis has led to highly volatile market conditions, fostering instances of asymmetric information. Therefore, this study aimed to explore the impact of attention on market dynamics by examining diverse possibilities over time. The article focused on all publicly listed industries on the Indonesian Stock Exchange (IDX/BEI). Using time series regression data from 1997 to 2020, the article comprised 5,615 observations across nine sectors. The primary model was based on three factors originating from the Fama-French and prospect theory, with attention serving as the main risk element to assess the impact of attention on abnormal returns (AR) during RE. The results disclosed that various events showed diverse effects on attention behavior, varying across all sectors. Additionally, moderation analysis showed a correlation between attention and AR. The results signified that RE mitigates the negative relationship between attention and AR. The adverse impact of attention on AR diminishes during RE. These results contributed to the literature by providing insights into the excessive attention to specific information disrupts market mechanisms, triggers disproportionate emotional responses, and alters investor preferences. Furthermore, this study established that events prompting excessive attention have varying effects on attention behavior across all sectors.

Compliance with COVID-19 preventive policies and governmental transparency: an experimental analysis

ABSTRACT This study investigates the impact of government transparency on COVID-related policy compliance among 722 Taiwanese college students. Results reveal a positive correlation between transparency, operationalized as information disclosure, and policy compliance. However, this relationship is mediated by individuals’ trust in the government, particularly when trust levels are low. The findings challenge conventional assumptions about the direct link between transparency and trust, especially in instances of negative performance information. The study underscores the significance of transparency in fostering policy compliance while emphasizing the nuanced role of trust in shaping public behavior. Implications for policy-making are discussed in light of these findings.

Enhancing random forest predictive performance for foot and mouth disease outbreaks in Uganda: a calibrated uncertainty prediction approach for varying distributions.

Foot-and-mouth disease poses a significant threat to both domestic and wild cloven-hoofed animals, leading to severe economic losses and jeopardizing food security. While machine learning models have become essential for predicting foot-and-mouth disease outbreaks, their effectiveness is often compromised by distribution shifts between training and target datasets, especially in non-stationary environments. Despite the critical impact of these shifts, their implications in foot-and-mouth disease outbreak prediction have been largely overlooked. This study introduces the Calibrated Uncertainty Prediction approach, designed to enhance the performance of Random Forest models in predicting foot-and-mouth disease outbreaks across varying distributions. The Calibrated Uncertainty Prediction approach effectively addresses distribution shifts by calibrating uncertain instances for pseudo-label annotation, allowing the active learner to generalize more effectively to the target domain. By utilizing a probabilistic calibration model, Calibrated Uncertainty Prediction pseudo-annotates the most informative instances, refining the active learner iteratively and minimizing the need for human annotation and outperforming existing methods known to mitigate distribution shifts. This reduces costs, saves time, and lessens the dependence on domain experts while achieving outstanding predictive performance. The results demonstrate that Calibrated Uncertainty Prediction significantly enhances predictive performance in non-stationary environments, achieving an accuracy of 98.5%, Area Under the Curve of 0.842, recall of 0.743, precision of 0.855, and an F1 score of 0.791. These findings underscore Calibrated Uncertainty Prediction's ability to overcome the vulnerabilities of existing ML models, offering a robust solution for foot-and-mouth disease outbreak prediction and contributing to the broader field of predictive modeling in infectious disease management.

Truth in Stories: Recognize Bias by Examining Point of View in Fiction and Nonfiction

Can students recognize bias and perspective in written works? In this study, the children focused on traditional stories presented in book or digital book format. This question is not only essential to a child’s success in school, but it is also a lifelong skill with far reaching consequences. The news has been filled with instances of inaccurate and misleading information, posted by bad actors, bots, special interest groups, or foreign interests to social media platforms. This information can lead a person to action, which can be detrimental to society and the individual. The skill of identifying misleading or inaccurate information is critical for an informed citizenry.To address this question, we must define bias and perspective. A bias is an individual’s belief, perspective is how one’s perception affects their understanding.

Representation Of Lung Cancer in Online Malaysian Newspapers

The study examined the representation of lung cancer in online Malaysian newspapers published in English. The objectives were to: (1) compare the frequency of information on severity, susceptibility, recommended actions, benefits and barriers in the newspapers; and (2) describe themes and sub-themes of these Health Belief Model constructs. The 12 online newspaper articles (2 New Strait Times, 4 The Star, 1 The Borneo Post and 5 The Sun) were analysed using the Health Belief Model. There were 89 instances of information on lung cancer. The newspaper articles crafted the threat of lung cancer through emphasis on severity (25.84%) and susceptibility (23.60%), particularly death statistics and tobacco smoking as a risk factor. The articles also recommended the public to take health protective actions (19.10%), and dealt with benefits (16.85%) and barriers (14.61%) to adoption of these behaviours. A chi-square test of independence showed significant differences among the newspapers on the frequency of information on severity, susceptibility, recommended actions, benefits, and barriers. However, the newspaper articles were similar in the highlight on the key benefits which were reducing deaths, improved health, and good treatment outcomes, and barriers which were inadequate knowledge and addiction. Half of the recommended actions to counter lung cancer were about the proposed tobacco ban for those born after 2005, and the rest were policy implementation issues, lifestyle changes and self-education. The study suggests that there is need for diversity of information on lung cancer in the media for better public health education.

Efficiently Finding Cyclical Patterns on Twitter Considering the Inherent Spatio-temporal Attributes of Data

Social networks such as Twitter provide thousands of terabytes per day, which can be exploited to find relevant information. This relevant information is used to promote marketing strategies, analyze current political issues, and track market trends, to name a few examples. One instance of relevant information is finding cyclic behavior patterns (i.e., patterns that frequently repeat themselves over time) in the population. Because trending topics on Twitter change rapidly, efficient algorithms are required, especially when considering location and time (i.e., the specific location and time) during broadcasts. This article presents an efficient algorithm based on association rules to find cyclical patterns on Twitter, considering the inherent spatio-temporal attributes of data. Using a Hash Table enhances the efficiency of this algorithm, called HashCycle. Notably, HashCycle does not use minimum support and can detect patterns in a single run over a sequence. The processing times of HashCycle were compared to the Apriori (which is a well-known and widely used on diverse platforms) and Projection-based Partial Periodic Patterns (PPA) algorithms (which is one of the most efficient algorithms in terms of processing times). Empirical results from two spatio-temporal databases (a synthetic data set and one based on Twitter) show that HashCycle has more efficient processing times than two state-of-the-art algorithms: Apriori and PPA.

Transformer based multiple instance learning for WSI breast cancer classification

The computer-aided diagnosis method based on deep learning provides pathologists with preliminary diagnostic opinions and improves their work efficiency. Inspired by the widespread use of transformers in computer vision, we try to explore their effectiveness and potential in classifying breast cancer tissues in WSIs, and propose a hybrid multiple instance learning method called HTransMIL. Specifically, its first stage is to select informative instances based on hierarchical Swin Transformer, which can capture global and local information of pathological images and is beneficial for obtaining accurate discriminative instances. The second stage aims to strengthen the correlation between selected instances via another transformer encoder consistently and produce powerful bag-level features by aggregating interactived instances for classification. Besides, visualization analysis is utilized to better understand the weakly supervised classification model for WSIs. The extensive evaluation results on a private and two public WSI breast cancer datasets demonstrate the effectiveness and competitiveness of HTransMIL. The code and models are publicly available at https://github.com/Chengyang852/Transformer-for-WSI-classification.

Autonomic active learning strategy using cluster-based ensemble classifier for concept drifts in imbalanced data stream

Systems are becoming autonomous nowadays. In real-world scenarios, the involvement of experts in generating a new decision has to be paid a certain amount of cost. Thus, in these conditions, creating a system with the least expert involvement and costs is the main destination. As a solution, an autonomic active learning strategy with a cluster-based ensemble classifier for concept drifts in imbalanced data stream (AACE-DI) is proposed in this paper. Initially, a cluster-based initializing formula is considered to select the most informative instances for labeling a dynamically changing ensemble classifier. Next, the automatically adjusting uncertainty strategy is introduced to prioritize the most uncertain, representative, and minority class data from imbalanced data stream, which also ensures the least labeling cost. This automatic formula not only outperforms the system but also makes it autonomic. Experimental evaluations are organized using 6 real-world data streams and 16 synthetic data streams with different types of concept drift and imbalanced ratios. To measure performance, evaluators, such as the prequential area under curve, recall, and geometric mean, are considered. Moreover, the proposed AACE-DI method achieves expected outcomes with the least cost as compared with other state-of-the-art learning methods.

Genome-wide identification and expression analysis of the GRAS gene family under abiotic stresses in wheat (Triticum aestivum L.)

The GRAS transcription factors are multifunctional proteins involved in various biological processes, encompassing plant growth, metabolism, and responses to both abiotic and biotic stresses. Wheat is an important cereal crop cultivated worldwide. However, no systematic study of the GRAS gene family and their functions under heat, drought, and salt stress tolerance and molecular dynamics modeling in wheat has been reported. In the present study, we identified the GRAS gene in Triticum aestivum through systematically performing gene structure analysis, chromosomal location, conserved motif, phylogenetic relationship, and expression patterns. A total of 177 GRAS genes were identified within the wheat genome. Based on phylogenetic analysis, these genes were categorically placed into 14 distinct subfamilies. Detailed analysis of the genetic architecture revealed that the majority of TaGRAS genes had no intronic regions. The expansion of the wheat GRAS gene family was proven to be influenced by both segmental and tandem duplication events. The study of collinearity events between TaGRAS and analogous orthologs from other plant species provided valuable insights into the evolution of the GRAS gene family in wheat. It is noteworthy that the promoter regions of TaGRAS genes consistently displayed an array of cis-acting elements that are associated with stress responses and hormone regulation. Additionally, we discovered 14 miRNAs that target key genes involved in three stress-responsive pathways in our study. Moreover, an assessment of RNA-seq data and qRT-PCR results revealed a significant increase in the expression of TaGRAS genes during abiotic stress. These findings highlight the crucial role of TaGRAS genes in mediating responses to different environmental stresses. Our research delved into the molecular dynamics and structural aspects of GRAS domain-DNA interactions, marking the first instance of such information being generated. Overall, the current findings contribute to our understanding of the organization of the GRAS genes in the wheat genome. Furthermore, we identified TaGRAS27 as a candidate gene for functional research, and to improve abiotic stress tolerance in the wheat by molecular breeding.

Drivers of gender differentiated self‐employment in developing countries: The instances of finance and information and communication technology

AbstractThis study unveils new insights into how financial inclusion and information and communication technology (ICT) influence self‐employment (aggregate and gender‐specific) in developing countries. The dearth of information in this perspective, particularly in the context of Africa, compels this fresh evaluation. Thus, on account of the annual panel series for 52 African countries from 2005 to 2019, the system‐generalized method‐of‐moments (sysGMM) and the novel method‐of‐moments quantile regression (MM‐QR), the following empirical narratives emerged. First, the estimates of both procedures demonstrate that self‐employment in Africa is self‐exacerbating. Second, the sysGMM demonstrates that financial inclusions could not influence self‐employment substantially; however, the estimates of the MM‐QR demonstrate varying significant positive relationships mostly at the upper quantiles for all categories of self‐employment. Third, the sysGMM reveals that ICT produced marginal positive effects on all categories except women's self‐employment. Fourth, the MM‐QR rectified a significant positive influence of ICT at the lower and middle quantiles for the aggregate, only at the lower quantiles (q10–q25) for men, and at the upper quantiles (q75–q90) for women. Lastly, it was revealed that although unemployment, GDP and inflation rates are not sufficient drivers of self‐employment, energy poverty constitutes a drag for self‐employment in the continent. Overall, the study is of immense relevance to policymakers, practicing and prospective entrepreneurs particularly in developing countries. More policy options have been provided within the study.

Global prevalence and content of information about alcohol use as a cancer risk factor on Twitter

ObjectivesAlcohol use is a major risk factor for several forms of cancer, though many people have limited knowledge of this link. Public health communicators and cancer advocates desire to increase awareness of this link with the long-term goal of reducing cancer burden. The current study is the first to examine the prevalence and content of information about alcohol use as a cancer risk on social media internationally. MethodsWe used a three-phase process (hashtag search, dictionary-based auto-identification of content, and human coding of content) to identify and evaluate information from Twitter posts between January 2019 and December 2021. ResultsOur hashtag search retrieved a large set of cancer-related tweets (N = 1,122,397). The automatic search process using an alcohol dictionary identified a small number of messages about cancer that also mentioned alcohol (n = 9061, 0.8%), a number that got small after adjusting for human coded estimates of the dictionary precision (n = 5927, 0.5%). When cancer-related messages also mentioned alcohol, 82% (n = 1003 of 1225 examined through human coding) indicated alcohol use as a risk factor. Coding found rare instances of problematic information (e.g., promotion of alcohol, misinformation) in messages about alcohol use and cancer. ConclusionsFew social media messages about cancer types that can be linked to alcohol mention alcohol as a cancer risk factor. If public health communicators and cancer advocates want to increase knowledge and understanding of alcohol use as a cancer risk factor, efforts will need to be made on social media and through other communication platforms to increase exposure to this information over time.

Applications of Active Learning in Predictive Maintenance

Nowadays, the common choice in maintenance strategies is predictive maintenance (PdM), deprecating the corrective and preventive kinds. Even with various machine learning techniques to get advanced predictive models to achieve PdM, difficulties remain in the data acquisition process. While there is a plethora of unlabeled data from sensors, most of those available techniques can only process labeled data, i.e, supervised learning. To combat the fact that the availability of the labeleddata is limited, this paper proposes the use of Active Learning to label and annotate the informative instances while minimizing overall processing time. This approach maintains high performance and decreases the number of labeled instances, with support from experimental results and a discussion of the applicability of this method.