Metadata Features Research Articles

An Efficient Algorithm for Exploitative Monetization Scam Video Detection Over Social Media Platforms

ABSTRACTIn recent years, there has been a surge in scam videos on YouTube which make false assurances of monetary benefits to the viewers by performing trivial tasks such as watching videos, downloading dubious apps, and sharing personal identifiable information (PII). The present work aims to detect such videos referred to as exploitative monetization scam videos. In the first step of the proposed two‐step algorithm, two new features are derived, named as scam–parent comment ratio features and nonscam parent comment ratio features using Word2Vec‐based comment classifier. The next step combines these features and other video metadata features to perform the final labeled video classification using machine learning algorithms. Five different algorithms, namely Gaussian Naive Bayes, logistic regression, support vector machine, adaboost and random forest have been used in the video classification task. The results illustrate that the area‐under‐curve (AUC) score of 91% is obtained for the video classification. A comparative analysis of the proposed model with the other state‐of‐the‐art methods shows that our model outperforms others in the exploitative monetization scam video classification task.

MeSH2Matrix: combining MeSH keywords and machine learning for biomedical relation classification based on PubMed

Biomedical relation classification has been significantly improved by the application of advanced machine learning techniques on the raw texts of scholarly publications. Despite this improvement, the reliance on large chunks of raw text makes these algorithms suffer in terms of generalization, precision, and reliability. The use of the distinctive characteristics of bibliographic metadata can prove effective in achieving better performance for this challenging task. In this research paper, we introduce an approach for biomedical relation classification using the qualifiers of co-occurring Medical Subject Headings (MeSH). First of all, we introduce MeSH2Matrix, our dataset consisting of 46,469 biomedical relations curated from PubMed publications using our approach. Our dataset includes a matrix that maps associations between the qualifiers of subject MeSH keywords and those of object MeSH keywords. It also specifies the corresponding Wikidata relation type and the superclass of semantic relations for each relation. Using MeSH2Matrix, we build and train three machine learning models (Support Vector Machine [SVM], a dense model [D-Model], and a convolutional neural network [C-Net]) to evaluate the efficiency of our approach for biomedical relation classification. Our best model achieves an accuracy of 70.78% for 195 classes and 83.09% for five superclasses. Finally, we provide confusion matrix and extensive feature analyses to better examine the relationship between the MeSH qualifiers and the biomedical relations being classified. Our results will hopefully shed light on developing better algorithms for biomedical ontology classification based on the MeSH keywords of PubMed publications. For reproducibility purposes, MeSH2Matrix, as well as all our source codes, are made publicly accessible at https://github.com/SisonkeBiotik-Africa/MeSH2Matrix.

Serum miRNA improves the accuracy of a multivariate index assay for triage of an adnexal mass

ObjectiveTo determine whether a multimodal assay combining serum microRNA with protein biomarkers and metadata improves triage assessment of an adnexal mass. MethodsSerum samples from 468 training subjects (191 cancer cases and 277 benign adnexal mass controls or healthy controls) were analyzed for seven protein biomarkers and 180 miRNA. Circulating analyte data were combined with age and menopausal status (metadata) into a neural network model to classify samples as cases or controls. Forward regression with ten-fold cross-validation minimized the dimensionality of the model while maximizing linear separation between cases and controls. Model validation proceeded using both internal (44 cases and 56 controls) and external validation sets (51 cases and 59 controls). ResultsThe total study population comprised 678 subjects, including 286 cases and 392 controls. Overall, 290 (43%) of the subjects were premenopausal. A panel of 10 miRNA delivered optimal performance when combined with protein and metadata features. The combined model improved the Receiver Operator Characteristic Area Under the Curve (ROC AUC) on the internal (AUC = 0.9; 95% CI 0.81–0.95) and external validation sets (AUC = 0.95; 95% CI 0.90–0.98) compared to miRNA alone or proteins plus metadata (without miRNA). On external validation, the combined model offered 92% sensitivity at 80% specificity overall, with 80% and 100% sensitivity for early and late-stage cancers, respectively, including 78% sensitivity for early-stage, serous ovarian cancers and 82% sensitivity for early-stage, non-serous cancers. ConclusionsA multimodal assay combining miRNA with protein biomarkers, age, and menopausal status improves surgical triage of an adnexal mass.

Using Health IT Standards to Facilitate Metadata Exchange Between Research Data Management Systems.

The National Research Data Infrastructure for Personal Health Data (NFDI4Health) uses Local Data Hubs (LDHs) to manage locally research studies, documents and sensitive personal data to support controlled data sharing. While research data management (RDM) systems facilitate the storage and preparation of data and metadata as well as organizational access, they often lack support for interoperability standards of the application domain. To support the exchange with external registries of research studies, we chose 17 attributes to characterize the most relevant aspects of clinical trials (in the following named "metadata profile"). We implemented the metadata profile in the RDM system FAIRDOM SEEK using core attributes and SEEK's extended metadata feature and created a mapping conforming to the Health Level 7 Fast Healthcare Interoperability Resources (FHIR) standard version R4. Finally, we implemented a prototype application interface for exports in FHIR-JSON format. We plan to extend the interface to serve central registries and support specific FHIR Implementation Guides from various use cases.

CF-MGAN: Collaborative filtering with metadata-aware generative adversarial networks for top-N recommendation

Generative Adversarial Networks (GANs) have several advantages over latent variable and autoregressive models, such as directly computing the input data distribution, fast sampling, and best samples generated. Furthermore, the metadata information of users and items plays a crucial role in effectively learning the user-item latent representations. Existing neural network approaches are deterministic, incapable of accounting for user-item latent representations, and the rating prediction process is calculative rather than generative. However, only some papers have utilized GANs for information retrieval tasks. However, they have several limitations, such as the inability to abstract data effectively, the limitation of the model's ability to learn accurate latent representations, and the absence of metadata information. To address these issues, we propose a GAN-based recommendation framework named Collaborative Filtering with Metadata-awareGenerative Adversarial Networks (CF-MGAN) to learn complex rating distribution for collaborative filtering (CF) tasks. We use a negative item generator for the implicit user purchase vector to prevent GAN from generating biased ratings. We exploited metadata information to overcome data sparsity issues. We use Variational Autoencoder to abstract user-item metadata features, a regularizer for GAN-generated rating. Extensive experiments on datasets reveal the effectiveness of the proposed model, showing an improvement in the top-N recommendation task.

Predicting adherence to ecological momentary assessments

Smartphones allow for prompting users with a short questionnaire about their current subjective experience, a technique often called Ecological Momentary Assessments. One of the biggest challenges for such studies is a lack of adherence, diminishing the benefits for both user and researcher. Being able to predict if a user is going to stop answering the questionnaire prompts would be beneficial for researchers and developers. This would allow for, for example, specifically addressing those users, or for over-sampling populations at higher risk of dropping out of a study. In this work, based on an observational study of the general population, we analyzed data from almost 1,000 users. The data include a large variety of sensor data from the users’ smartphones. We utilized machine learning to predict adherence on a day-to-day level, as well as predict adherence based on participant data after on-boarding. For day-to-day prediction, the best performing model was a model based on metadata features (days since first questionnaire was filled out, days since the last questionnaire was filled out, number of filled-out questionnaires, days since app installation), yielding an area under the precision–recall curve of 0.89. The inclusion of sensor data did not improve the model’s performance, indicating that the high cost of collecting and processing sensor data is not worth the benefits for predicting fill-out behavior. Predicting at sign-up if a user will adhere to a questionnaire prompt at least once was better than chance, but further studies are needed.

ExaAUAC: Arabic Twitter user age prediction corpus based on language and metadata features

Twitter is a rich resource for analyzing the contents of social media and extracting the age groups of users can be beneficial for recommender systems, marketing and advertising. Age detection task is an aspect of demographic information of users. In this study a large-scale corpus of Arabic Twitter users including 181k user profiles with diverse age groups consisting of −18, 18–24, 25–34, 35–49, 50–64, +65 is presented. The corpus is created by four methods: (1) collecting publicly available birthday announcement tweets using the Twitter Search application programming interface, (2) augmenting data, (3) fetching verified accounts, and (4) manual annotation. To have a best age detection model on the presented corpus, different evaluations are tested to find the model with highest accuracy and efficiency. Number of tweets, regression vs. classification, using metadata of users and tweets, using LSTM+CNN model vs. BERT are some parts of examinations done. Presented methodology is based on language and metadata features and final model is fine-tuned with BERT on 70k users and evaluated on 8200 manually annotated users. We show that our best model, compared with LSTM+CNN model and BERT-based similar model yields an improvement of up to 9% in F1-score and increment of 5% in accuracy, respectively. The model achieved macro-averaged F1-score of 44 on six age groups, and F1-score of 58 on three age groups of −25, 25–34, +35. The link of our proposed data is provided here: www.github.com/exaco/ExaAUAC.

Interoperability of Oral History Metadata: An Ontological Model

ABSTRACT This article presents the metadata characteristics of oral history collections and the challenges faced when attempting to provide long-term access to audio or video recordings. We propose an ontological model combining the International Committee for Documentation Conceptual Reference Model (CIDOC CRM) from the cultural heritage field with European Broadcasting Union Core (EBUCore) from the European audiovisual community. We form a mapping methodology with the goal of combining the strongest attributes from these two standards. The goal of this research is to generate a conceptual model for the description of oral history interviews in order to reveal the entities participating in the interviews, their semantics, and their relationships. The proposed model enables the meaningful exchange of information of audiovisual collections and their integration with other cultural heritage collections in the online environment.

DeepMetaDroid: Real-Time Android Malware Detection Using Deep Learning and Metadata Features

The increasing prevalence of Android malware poses significant risks to mobile devices and user privacy. The traditional detection methods have limitations in keeping up with the evolving landscape of malware attacks, necessitating the development of more effective solutions. In this paper, we present DeepMetaDroid, a real-time detection approach for Android malware that leverages metadata features. By analyzing crucial metadata, including APK size, download size, permissions, certificates, and DEX files, the proposed method enables effective identification of malware and enhances mobile security. Using deep learning techniques, a lightweight Android real-time monitoring system is equipped with the trained model. These methods include long short-term memory (LSTM), gated recurrent units (GRU), convolutional neural networks (CNN), deep neural networks (DNN), and other ensemble models. Utilizing the rectified linear unit (ReLU) as the activation function, the DNN model is constructed with 32 neurons in the input layer. A one-dimensional convolutional layer with 32 neurons and a filter size of three is used as the input layer in the CNN model. The LSTM model is designed with an input layer consisting of 16 neurons. The GRU model with 32 neurons is employed in the input layer. Additionally, ensemble models that combined several architectures were developed. The proposed method offers a faster and more scalable solution for malware detection by consuming fewer resources like memory and CPU. This work ensures device security by providing real-time monitoring on Android devices to prevent users from installing malicious applications and, thus, enhance user privacy and security.

Joint-individual fusion structure with fusion attention module for multi-modal skin cancer classification

Many convolutional neural network (CNN) based approaches for skin cancer classification primarily rely on dermatological images, yielding commendable results in classification accuracy. However, leveraging patient metadata, a crucial source of clinical information for dermatologists, can further enhance accuracy. Current methodologies predominantly employ basic joint fusion structures (FS) and fusion modules (FMs) for multi-modal classification, leaving room for advancement in enhancing accuracy through exploration of more sophisticated FS and FM architectures. Thus, this paper introduces a novel fusion method that integrates dermatological images (dermoscopy images or clinical images) with patient metadata for skin cancer classification, focusing on enhancing FS and FM components. Initially, we propose a joint-individual fusion (JIF) structure that simultaneously learns shared features across multi-modality data while preserving specific characteristics. Subsequently, we introduce a multi-modal fusion attention (MMFA) module designed to amplify the most relevant image and metadata features through a combination of self and mutual attention mechanisms, thereby bolstering the decision-making pipeline. Our study compares the efficacy of the proposed JIF-MMFA method with other state-of-the-art fusion techniques across three distinct public datasets. Results demonstrate that the JIF-MMFA method consistently enhances classification outcomes across various CNN backbones, outperforming alternative fusion methodologies on all three datasets. These findings underscore the effectiveness and robustness of our proposed approach in skin cancer classification.

A novel multi-task learning network for skin lesion classification based on multi-modal clues and label-level fusion

In this paper, we propose a multi-task learning (MTL) network based on the label-level fusion of metadata and hand-crafted features by unsupervised clustering to generate new clustering labels as an optimization goal. We propose a MTL module (MTLM) that incorporates an attention mechanism to enable the model to learn more integrated, variable information. We propose a dynamic strategy to adjust the loss weights of different tasks, and trade off the contributions of multiple branches. Instead of feature-level fusion, we propose label-level fusion and combine the results of our proposed MTLM with the results of the image classification network to achieve better lesion prediction on multiple dermatological datasets. We verify the effectiveness of the proposed model by quantitative and qualitative measures. The MTL network using multi-modal clues and label-level fusion can yield the significant performance improvement for skin lesion classification.

P238: The use of artificial intelligence in 3D imaging modalities and diabetic foot disease – a systematic review

Abstract Introduction Diabetic foot disease (DFD) is serious complication of diabetes with a multifactorial aetiology and carries a significant risk. The prevalence of DFD continues to grow globally. Artificial intelligence has been proposed in aiding early detection and risk stratification for ulceration and other major complications including sepsis, minor or major lower limb amputation and death. Aims To systematically review the literature available on the use of artificial intelligence in three-dimensional imaging modalities in diabetic foot disease. Method A literature review was conducted in accordance with PRISMA guidelines. Embase and Medline (via the Ovid interface), CINAHL (via Ebsco Host), Web of Science and Scopus databases were searched. The grey literature was also reviewed on ClinicalTrials.gov and the NIHR journals library. The medical subject headings terms “Diabetes” AND “Diabetic foot disease” AND “Artificial intelligence” and various permutations of three-dimensional imaging modalities including “Computed Tomography”, “Magnetic Resonance Imaging” and “Positron Emission Tomography” were employed in the primary search string. Articles were independently screened by two reviewers. Results 4,865 studies were identified. 102 duplicates were removed. 4,721 were excluded in the abstract screening. 42 articles underwent full text review. Existing use centers around binary use of CT based imaging in predicting risk. Conclusion The use of deep learning models is still being explored and evaluated. Current methodologies focus on wound imaging classification, plantar thermography and plantar pressures. Specialised models which evaluate 3D imaging have potential for generation of supra-human insights into extraction of novel metadata features and prediction utilising integration of multidimensional patient characteristics.

Research and implementation of multi-disease diagnosis on chest X-ray based on vision transformer.

Disease diagnosis in chest X-ray images has predominantly relied on convolutional neural networks (CNNs). However, Vision Transformer (ViT) offers several advantages over CNNs, as it excels at capturing long-term dependencies, exploring correlations, and extracting features with richer semantic information. We adapted ViT for chest X-ray image analysis by making the following three key improvements: (I) employing a sliding window approach in the image sequence feature extraction module to divide the input image into blocks to identify small and difficult-to-detect lesion areas; (II) introducing an attention region selection module in the encoder layer of the ViT model to enhance the model's ability to focus on relevant regions; and (III) constructing a parallel patient metadata feature extraction network on top of the image feature extraction network to integrate multi-modal input data, enabling the model to synergistically learn and expand image-semantic information. The experimental results showed the effectiveness of our proposed model, which had an average area under the curve value of 0.831 in diagnosing 14 common chest diseases. The metadata feature network module effectively integrated patient metadata, further enhancing the model's accuracy in diagnosis. Our ViT-based model had a sensitivity of 0.863, a specificity of 0.821, and an accuracy of 0.834 in diagnosing these common chest diseases. Our model has good general applicability and shows promise in chest X-ray image analysis, effectively integrating patient metadata and enhancing diagnostic capabilities.

Metadata and Image Features Co-Aware Semi-Supervised Vertical Federated Learning With Attention Mechanism

Metadata and Image Features Co-Aware Semi-Supervised Vertical Federated Learning With Attention Mechanism

Generalization Across Experimental Parameters in Neural Network Analysis of High-Resolution Transmission Electron Microscopy Datasets.

Neural networks are promising tools for high-throughput and accurate transmission electron microscopy (TEM) analysis of nanomaterials, but are known to generalize poorly on data that is "out-of-distribution" from their training data. Given the limited set of image features typically seen in high-resolution TEM imaging, it is unclear which images are considered out-of-distribution from others. Here, we investigate how the choice of metadata features in the training dataset influences neural network performance, focusing on the example task of nanoparticle segmentation. We train and validate neural networks across curated, experimentally collected high-resolution TEM image datasets of nanoparticles under various imaging and material parameters, including magnification, dosage, nanoparticle diameter, and nanoparticle material. Overall, we find that our neural networks are not robust across microscope parameters, but do generalize across certain sample parameters. Additionally, data preprocessing can have unintended consequences on neural network generalization. Our results highlight the need to understand how dataset features affect deployment of data-driven algorithms.

The anatomy of conspiracy theorists: Unveiling traits using a comprehensive Twitter dataset

The discourse around conspiracy theories is currently thriving amidst the rampant misinformation in online environments. Research in this field has been focused on detecting conspiracy theories on social media, often relying on limited datasets. In this study, we present a novel methodology for constructing a Twitter dataset that encompasses accounts engaged in conspiracy-related activities throughout the year 2022. Our approach centers on data collection that is independent of specific conspiracy theories and information operations. Additionally, our dataset includes a control group comprising randomly selected users who can be fairly compared to the individuals involved in conspiracy activities. This comprehensive collection effort yielded a total of 15K accounts and 37M tweets extracted from their timelines. We conduct a comparative analysis of the two groups across three dimensions: topics, profiles, and behavioral characteristics. The results indicate that conspiracy and control users exhibit similarity in terms of their profile metadata characteristics. However, they diverge significantly in terms of behavior and activity, particularly regarding the discussed topics, the terminology used, and their stance on trending subjects. In addition, we find no significant disparity in the presence of bot users between the two groups. Finally, we develop a classifier to identify conspiracy users using features borrowed from bot, troll and linguistic literature. The results demonstrate a high accuracy level (with an F1 score of 0.94), enabling us to uncover the most discriminating features associated with conspiracy-related accounts.

OMEinfo: global geographic metadata for -omics experiments.

Microbiome studies increasingly associate geographical features like rurality and climate with microbiomes. It is essential to correctly integrate rich geographical metadata; and inconsistent definitions of rurality, can hinder cross-study comparisons. We address this with OMEinfo, a tool for automated retrieval of consistent geographical metadata from user-provided location data. OMEinfo leverages open data sources such as the Global Human Settlement Layer, and Open-Data Inventory for Anthropogenic Carbon dioxide. OMEinfo's web-app enables users to visualize and investigate the spatial distribution of metadata features. OMEinfo promotes reproducibility and consistency in microbiome metadata through a standardized metadata retrieval approach. To demonstrate utility, OMEinfo is used to replicate the results of a previous study linking population density to bacterial diversity. As the field explores relationships between microbiomes and geographical features, tools like OMEinfo will prove vital in developing a robust, accurate, and interconnected understanding of these interactions, whilst having applicability beyond this field to any studies utilizing location-based metadata. Finally, we release the OMEinfo annotation dataset of 5.3 million OMEinfo annotated samples from the ENA, for use in retrospective analyses of sequencing samples, and suggest several ways researchers and sequencing read repositories can improve the quality of underlying metadata submitted to these public stores. OMEinfo is freely available and released under an MIT licence. OMEinfo source code is available at https://github.com/m-crown/OMEinfo/ and https://doi.org/10.5281/zenodo.10518763.

Predicting movies’ eudaimonic and hedonic scores: A machine learning approach using metadata, audio and visual features

In the task of modeling user preferences for movie recommender systems, recent research has demonstrated the benefits of describing movies with their eudaimonic and hedonic scores (E and H scores), which reflect the depth of their message and the level of fun experience they provide, respectively. So far, the labeling of movies with their E and H scores has been done manually using a dedicated instrument (a questionnaire), which is time-consuming. To address this issue, we propose an automatic approach for predicting E and H scores. Specifically, we collected E and H scores of 709 movies from 370 users (with a total of 3699 records), augmented this dataset with metadata, audio, and low-level and high-level visual features, and trained machine learning models for predicting the E and H scores of movies. This study investigates the use of machine learning models in predicting the E and H scores of movies using various feature sets, including audio, low-level and high-level visual features, and metadata. We compared the performance of predictive models using different combinations of features with the majority classifier as the baseline approach. The results demonstrate that our proposed machine learning-based models significantly outperform the baseline in predicting E and H scores, particularly when leveraging metadata features. Specifically, the random forest classifier achieved a 20% increase in ROC AUC compared to the baseline when predicting both the E score and the H score. These improvements were found to be statistically significant. Overall, our findings suggest that automated tools for predicting E and H scores in movies are promising alternatives to traditional questionnaire-based approaches.

SPDB: a comprehensive resource and knowledgebase for proteomic data at the single-cell resolution.

The single-cell proteomics enables the direct quantification of protein abundance at the single-cell resolution, providing valuable insights into cellular phenotypes beyond what can be inferred from transcriptome analysis alone. However, insufficient large-scale integrated databases hinder researchers from accessing and exploring single-cell proteomics, impeding the advancement of this field. To fill this deficiency, we present a comprehensive database, namely Single-cell Proteomic DataBase (SPDB, https://scproteomicsdb.com/), for general single-cell proteomic data, including antibody-based or mass spectrometry-based single-cell proteomics. Equipped with standardized data process and a user-friendly web interface, SPDB provides unified data formats for convenient interaction with downstream analysis, and offers not only dataset-level but also protein-level data search and exploration capabilities. To enable detailed exhibition of single-cell proteomic data, SPDB also provides a module for visualizing data from the perspectives of cell metadata or protein features. The current version of SPDB encompasses 133 antibody-based single-cell proteomic datasets involving more than 300 million cells and over 800 marker/surface proteins, and 10 mass spectrometry-based single-cell proteomic datasets involving more than 4000 cells and over 7000 proteins. Overall, SPDB is envisioned to be explored as a useful resource that will facilitate the wider research communities by providing detailed insights into proteomics from the single-cell perspective.

A blockchain based solution to improve loyalty program with NFT in agribusiness

This paper presents the design and preliminary development a new model based on blockchain for improving brand loyalty in agribusiness by customer reward. Using facilities of Ethereum Network and solidity programing, we have created an Agri non-fungible token (NFT) according to ERC-721 protocol that is most widely used protocol for creating NFTs on Ethereum. One of the main advantages of ERC-721 is its flexibility, as it allows developers to define custom metadata for each NFT and customize the attributes. The interaction between Ethereum nodes and the user is established via the frontend, which is developed in Nodejs and connects to the Ethereum network. In previous works on loyalty systems, blockchain-based solutions were proposed that resemble the system developed in this study but there are significant differences between our work and other related works such as: the use of NFT and personalization for customers, the adoption of a new version of Solidity, as well as the adoption of the metadata feature for using media to encourage and assist customer brand awareness and repurchase. We tried to provide a safe and transparent way to use NFTs in the way of increasing brand loyalty in agricultural businesses. Due to characters of NFTs as uniqueness and personalization, customers in a loyalty cycle prefer to shop more in order to earn more AgriNFT, causing retailer to sell more items and manufacturers to generate more goods.