A Dataset for Information Extraction from News Web Pages

Peas, as a globally significant legume crop, play a crucial role in food production and human nutrition. However, field pests and diseases, drought, high temperature and nutrient deficiency and other stresses seriously affect the yield and quality of peas. Timely and efficient stress monitoring is of great significance for ensuring pea production. This study reviews the advantages of unmanned aerial vehicle (UAV) remote sensing platforms in farmland stress monitoring, including high-throughput phenotypic acquisition and the roles of multispectral, RGB, and thermal infrared sensors. It also analyzes the applicability of commonly used deep learning models (CNN, RNN, Transformer) in crop stress detection. And the application progress of image classification, object detection, and semantic segmentation technologies in identifying crop stress types, a technical framework combining unmanned aerial vehicles and deep learning was constructed. The data collection and annotation process, data preprocessing and enhancement methods, as well as the pipeline for fusing multi-source images with deep learning models were expounded. The identification, degree classification and spatial distribution visualization of physiological stress (drought, nutrient deficiency, high temperature) and biological stress (diseases, pests) were mainly discussed. In the case of field stress detection of peas, the recognition performance of the deep learning model and its application value in field management decisions were verified. The combination of unmanned aerial vehicle (UAV) remote sensing and deep learning technology can achieve precise detection of field stress in peas. This study aims to provide strong support for precision agricultural management.

Hate speech-related lexicons have been proved to be useful for many tasks such as data collection and classification. However, existing Portuguese lexicons do not distinguish between European and Brazilian Portuguese, and do not include neutral terms that are potentially useful to detect a broader spectrum of content referring to minorities. In this work, we present MIN_PT, a new European Portuguese Lexicon for Minorities-Related Terms specifically designed to tackle the limitations of existing resources. We describe the data collection and annotation process, discuss the limitation and ethical concerns, and prove the utility of the resource by applying it to a use case for the Portuguese 2021 presidential elections.

IntroductionPatients increasingly rely on online health information to understand and manage their diseases. Concerns about the quality and readability of these materials have been reported in the literature. Poor quality and difficult-to-read information lead to delayed diagnoses and adverse outcomes. We assessed the quality and readability of online health information about iron deficiency anemia (IDA) on Google.MethodWe searched for online web pages using the term "iron deficiency anemia"on Google. One hundred and twelve out of 200 web pages were included. We assessed web page typology, readability, the Journal of the American Medical Association (JAMA) score, the DISCERN score, and the Health on the Net Foundation certification (HONcode). Statistical analysis was performed with R version 4.2.2.ResultNon-profit and scientific journal web pages were the most common. Scientific journal web pages were of the highest quality. News web pages were the most readable. The first Google Page web pages did not have greater JAMA scores or lower Flesch-Kinkaid Reading Grade Level (FKGL) and Simple Measure of Gobbledygook (SMOG) scores. Forty-six percent of all web pages were high-quality. Web pages on the first Google page were more likely to have HONCode certification.ConclusionWe highlight gaps in the readability and quality of online information about IDA. Online web pages exceeded the recommended reading level for patients. Most web pages were low quality; only a quarter were HONcode-certified; and the first Google page web pages were not higher in quality than the later web pages on search.

As the web endures to develop, there is an enormous amount of information which is typically designed for its users, which makes it difficult to extract relevant data from numerous sources. In this paper, we propose an approach for extracting the main content from news web pages. Our approach is based on the concept of tokenisation of HTML page, these tokens construct the tag tree; web pages from different websites are parsed into Tag tree and generated a template from each web pages and discover matching patterns and multiple sequence alignment. It finds and removed shared token sequences from the web pages until the relevant information is extracted from them. We perform experiments on 500 web pages from ten different news websites. Experimental results show that our approach efficiently extracts the relevant information.

<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">_</sup> One of the biggest challenges faced in deep learning community is the image data annotation, we know how data is essential for any machine learning problems but when it comes to object detection and segmentation (ODS) problems the data preparation process becomes extremely difficult and time-consuming. The data preparation becomes two steps process, first the data collection itself and second the data annotation. In this paper I present AADGen, a novel way of automatic data collection and annotation (DCA) for text detection and segmentation (TDS) problems. This method reduces the time taken for DCA to almost zero and provides error-free data annotation process. The DCA process will help us in building online learning model training by providing on demand continuous stream of training data. I have done extensive experiments on the data annotation process and achieved more than 99% annotation accuracy. I present a fast, scalable, customizable and error free data annotation approach.

In addition to the news content, most news web pages also contain various noises, such as advertisements, recommendations, and navigation panels. These noises may hamper the studies and applications which require pre-processing to extract the news content accurately. Existing methods of news content extraction mostly rely on non-content features, such as tag path, text layout, and DOM structure. However, without considering topics of the news content, these methods are difficult to recognize noises whose external characteristics are similar to those of the news content. In this paper, we propose a method that combines non-content features and a topic feature based on Dempster-Shafer (DS) theory to increase the recognition accuracy. We use maximal compatibility blocks to generate topics from text nodes and then obtain feature values of topics. Each feature is converted into evidence for the DS theory which can be utilized in the uncertain information fusion. Experimental results on English and Chinese web pages show that combining the topic feature by DS theory can improve the extraction performance obviously.

This paper presents a multilayered garment animation generation method. Generating realistic dynamics in 3D garment animations is a challenging task due to the complex nature of multilayered garments and the variety of outer forces involved. Existing data-driven approaches have mainly focused on the study of static draping deformation of multilayer garments, with less consideration for the temporal deformation of garments, such as the time-varying motion behaviors of individual layers and their continuous interactions during motion. In addition, these methods require a substantial amount of high-quality paired garment datasets for network training, leading to a costly data acquisition and annotation process. To address these challenges, we propose a multilayered garment animation generation method that explicitly models different garment layers as separate meshes, and employs a combination of unsupervised and temporally supervised learning strategies to analyze and model the behavior of individual garment layers and their interactions. Our primary contribution lies in introducing a two-stage network architecture for layered garment processing, which decomposes multilayer garment deformation prediction into single-layer garment generation and interlayer garment interaction deformation. We focus more on generating two-layered clothing animations. Of course, our two-layered approach can be used iteratively to support more layers by using the current outer layer as the inner layer for the next iteration. This approach achieves dynamic simulation of multilayer garments, and experimental results demonstrate that our method can generate realistic multilayer garment deformation effects, outperforming existing methods both visually and in terms of evaluation metrics.

Healthcare-associated infections (HAIs) are associated with increases in mortality, morbidity, and overall cost. To guide patient care, clinicians desire accurate, rapid and cost-effective tools that enable them to effectively detect, control, and prevent HAI while managing limited resources. Without any doubts, molecular methods, especially real-time PCR, have been playing an important role in infection control strategies and the clinical management of HAIs. PCR and electrospray ionization mass spectrometry (PCR/ESI-MS) is an attractive methodology; it enables the identification and typing of microorganisms, as well as the detection of resistance and virulence genes because it combines the advantages of real-time PCR with those of mass spectrometry detection. However, this methodology will be probably surpassed by whole genome sequence (WGS), since it already has a competitive price and a relative rapid turnaround. The data assembly, annotation process, and data analysis still constitute obstacles to the wider use of WGS, as well as DNA obtainment directly from clinical specimens. The development of automated tools for analysis, compilation of WGS databases, and development of specific genetic criteria may enable a wider use of these techniques by routine clinical microbiology laboratory.

The process of collecting and annotating training data may introduce distribution artifacts which may limit the ability of models to learn correct generalization behavior. We identify failure modes of SOTA relation extraction (RE) models trained on TACRED, which we attribute to limitations in the data annotation process. We collect and annotate a challenge-set we call Challenging RE (CRE), based on naturally occurring corpus examples, to benchmark this behavior. Our experiments with four state-of-the-art RE models show that they have indeed adopted shallow heuristics that do not generalize to the challenge-set data. Further, we find that alternative question answering modeling performs significantly better than the SOTA models on the challenge-set, despite worse overall TACRED performance. By adding some of the challenge data as training examples, the performance of the model improves. Finally, we provide concrete suggestion on how to improve RE data collection to alleviate this behavior.

Over the years, wireless communication networks have been widely used by a large community of users in wide variety of applications such as intelligent transportation systems, energy management, safety, and security etc. But, during this era, due to large number of user's request, there may be a performance bottleneck in some part of the network with respect to various QoS parameters such as congestion and network delay. Hence, there is a requirement of an efficient classification technique to reduce congestion in the network so that throughput of various applications can be increased. Classification helps in searching, sorting, retrieval, and querying of a document for the wireless networks. World Wide Web (WWW) contains huge repository of information in the form of web pages. However, size of Internet is growing day-by-day. The huge repository of information poses challenge to collect and process the relevant related information of a particular domain. So, traditional text classification techniques are difficult to apply on the rapidly growing web-based contents. Hence, novel approaches and techniques need to be devised to reduce the manual efforts in web page classification. Keeping focus on these points, this paper proposes a novel approach for multiclass classifier based on unique personality features of the web page of particular domain category for the next generation wireless networks. Personality features are collected and assigned weights in the proposed scheme. Then, the proposed classifier is trained based on these special features. Results obtained depict that proposed classifier successfully classified news domain pages, education, resume, online shopping, and research web pages from large database repository. Accuracy of the proposed classifier is found to be satisfactory from a large data set of different categories. Also, there is a 10---15 % overall performance gain using the proposed scheme in comparison to the other existing schemes.

IntroductionAyushman Bharat Pradhan Mantri Jan Aarogya Yojana (AB PM-JAY) has enabled the Government of India to become a strategic purchaser of health care services from private providers. To generate base cost evidence for evidence-based policymaking the Costing of Health Services in India (CHSI) study was commissioned in 2018 for the price setting of health benefit packages. This paper reports the findings of a process evaluation of the cost data collection in the private hospitals.MethodsThe process evaluation of health system costing in private hospitals was an exploratory survey with mixed methods (quantitative and qualitative). We used three approaches–an online survey using a semi-structured questionnaire, in-depth interviews, and a review of monitoring data. The process of data collection was assessed in terms of time taken for different aspects, resources used, level and nature of difficulty encountered, challenges and solutions.ResultsThe mean time taken for data collection in a private hospital was 9.31 (± 1.0) person months including time for obtaining permissions, actual data collection and entry, and addressing queries for data completeness and quality. The longest time was taken to collect data on human resources (30%), while it took the least time for collecting information on building and space (5%). On a scale of 1 (lowest) to 10 (highest) difficulty levels, the data on human resources was the most difficult to collect. This included data on salaries (8), time allocation (5.5) and leaves (5).DiscussionCost data from private hospitals is crucial for mixed health systems. Developing formal mechanisms of cost accounting data and data sharing as pre-requisites for empanelment under a national insurance scheme can significantly ease the process of cost data collection.

Information about mental health systems is essential for mental health planning to reduce the burden of neuropsychiatric disorders. Unfortunately, many low- and middle-income countries lack systematic information on their mental health systems. The objectives, scope, structure, and contents of mental health assessment and monitoring instruments commonly used in high-income countries may not be appropriate for use in middle- and low-income countries. The World Health Organization (WHO) has recently developed the WHO Assessment Instrument for Mental Health Systems (WHO-AIMS), a comprehensive assessment tool for mental health systems designed for middle- and low-income countries. WHO-AIMS was developed through an iterative process that included input from in-country and international experts on the clarity, content, validity, and feasibility of the instrument, as well as a pilot trial. The resulting instrument, WHO-AIMS 2.2, consists of six domains: policy and legislative framework, mental health services, mental health in primary care, human resources, public information and links with other sectors, and monitoring and research. These domains address the ten recommendations of the World Health Report 2001 through 28 facets and 155 items. All six domains need to be assessed to form a basic, yet broad, picture of a mental health system, with a focus on health sector activities. WHO-AIMS provides essential information for mental health policy and service delivery. Countries will be able to develop information-based mental health policy and plans with clear baseline information and targets. Moreover, they will be able to monitor progress in implementing reform policies, providing community services, and involving consumers, families, and other stakeholders in mental health promotion, prevention, care and rehabilitation. This article provides an overview of the rationale, development process, and potential uses and benefits of WHO-AIMS.

Most information in the world exists in the format of text, such as news articles and web pages. Different lines of research have been conducted to discover, understand and access knowledge about real-world entities and relations from text. However, the application of these word oriented methods to ontology integration tasks has not been yet explored. In this paper, we apply these word oriented methods to ontology integration tasks in which we analyze a noun phrase (NP) to identify its head noun, which is useful to avoid wrong relations between entities. We also propose a collaborative acquisition algorithm that combines WordNet-based and Text corpus.

In many companies, business units that aim to online sell, need every type of referential data about the market. In order to collect this data which can be group of price, content, survey etc. with a predefined format, websites which sell similar products can be used. The methods used in the data collection process are generally categorized by 3 main groups: 1- Manual 2-Half Manual 3-Auto. Statically coded data collectors (type 1 and type 2) are unable to collect healthy data in the long term and require continuous development and maintenance effort, as internet pages are dynamic and changes would happen frequently in their page designs. In this study, a data scraping application (type 3) which is not affected by structural changes in web pages was developed. This study aims to obtain data from images of web pages using Deep CNNs.

We conducted an electronic, self-administered cross-sectional survey of all ODOs in Canada from November 2021 to January 2022. We targeted key knowledge holders familiar with the data collection processes within each Canadian ODO known to Canadian Blood Services. Categorical item responses are presented as numbers and proportions. We achieved a 100% response rate from 10 Canadian ODOs. Most data were collected by organ donation coordinators. Only 2 of 10 ODOs reported using scripts explaining why sociodemographic data are being collected or incorporated training in cultural sensitivity for any given variable. A lack of cultural sensitivity training was endorsed by 50% of respondents as a barrier to the collection of sociodemographic variables by ODOs, whereas 40% of respondents identified a lack of training in sociodemographic variable collection as a significant barrier. Few programs routinely collect sufficient data to examine health inequities with an intersectional lens. Most data collection occurs midway through the ODO interaction, creating a missed opportunity to better understand differences in social identities of patients who register their intention to donate in advance or who decline the donation. National standardization of equity-relevant data collection definitions and processes of the collection is needed.