Crowdsourcing Human Annotation on Web Page Structure

The distance metric is the corner stone of nearest neighbor (NN)-based methods, and therefore, of nearest prototype (NP) algorithms. That is because they classify depending on the similarity of the data. When the data is characterized by a set of features which may contribute to the classification task in different levels, feature weighting or selection is required, sometimes in a local sense. However, local weighting is typically restricted to NN approaches. In this paper, we introduce local feature weighting (LFW) in NP classification. LFW provides each prototype its own weight vector, opposite to typical global weighting methods found in the NP literature, where all the prototypes share the same one. Providing each prototype its own weight vector has a novel effect in the borders of the Voronoi regions generated: They become nonlinear. We have integrated LFW with a previously developed evolutionary nearest prototype classifier (ENPC). The experiments performed both in artificial and real data sets demonstrate that the resulting algorithm that we call LFW in nearest prototype classification (LFW-NPC) avoids overfitting on training data in domains where the features may have different contribution to the classification task in different areas of the feature space. This generalization capability is also reflected in automatically obtaining an accurate and reduced set of prototypes.

Domain-independent web information extraction can be addressed as a structured prediction problem where we learn a mapping function from an input web page to the structured and interdependent output variables, labeling each block on the page. In this paper, built upon an HTML parser of Internet Explorer that parses and renders a web page based on HTML tags and visual appearance, we propose a max margin learning approach for web information extraction. Specifically, the output of the parser is a vision tree, which is similar to a DOM tree but with visual information, i.e., how each node is displayed. Based on this hierarchical structure, we develop a max margin learning method for labeling each of its nodes. Due to the rich connections between blocks on the web page, we further introduce edges that connect spatially adjacent nodes on the vision tree, complicating the problem into a cyclic graph labeling task. A max margin learning method on cyclic graphs is developed for this problem, where loopy belief propagation is used for approximate inference. Experimental results on web data extraction show the feasibility and promise of our approach.

Crowdsourcing system, which utilizes many workers to process computer-complexity tasks, has become an effective platform in today’s online labor markets. In a crowdsourcing system, maximizing the total utility is one key design goal. This goal is extremely hard because a computer-complexity task can be a multi-dimensional large-scale task that contains thousands or millions of atomic tasks. In online situation, we need to consider both the varying service of workers and future unknown task arrivals. As we know, none of the previous work considers a multi-dimensional large-scale task assignment for utility maximization. In this paper, an online framework is proposed to solve this optimization problem by running atomic tasks in parallel on workers. To estimate worker service rates, we consider each varying worker as an arm for a multi-armed bandit in crowdsourcing system. We design the online scheduling algorithm from a bandit perspective by Online Convex Optimization (OCO) techniques. We prove that our designed algorithm can yield a sublinear regret bound. Finally, we show that our designed algorithm is better than the baseline algorithms by nearly 10% for the total utility achieved.

This is an accepted manuscript of an article published by ACM in W4A '20: Proceedings of the 17th International Web for All Conference on 20/04/2020, available online: https://doi.org/10.1145/3371300.3383340 
\nThe accepted version of the publication may differ from the final published version.

C 3H 8 separation from CH 4 and H 2 using a synthesized PDMS membrane: Experimental and neural network modeling

Although several deepfake detection technologies have achieved great detection accuracy inside the data domain in recent years, there are still limitations in cross-domain generalization. This is due to the model’s ease of fitting the data sample distribution in the training data domain and its tendency to detect a specific forgery trace in order to reach a judgment rather than catching generalized forgery traces. In this paper, we propose to learn Local Reconstruction Errors for face forgery detection. The local anomaly traces of the fake face are often mapped using the original real face as a reference; however, the original real face of the fake face cannot be acquired in the real scenario. Therefore, this solution designs a local reconstruction autoencoder trained with real samples. By masking key areas of the face, the original real face can be reconstructed. Because the autoencoder only learns how to restore the essential parts of the real face using local patches of real samples, it cannot recover the forging traces or target face information in the fake face. Therefore, the reconstructed image forms a reconstructed difference with the original image. This solution aids the model in detecting local differences in fake faces by producing feature-level local difference attention mappings in the network’s middle layer. A series of experiments demonstrate that this solution has good detection and generalization performance.

The main issue for effective Web information extraction is how to recognize similar patterns in a Web page. Traditionally, it has been shown that pattern matching by using the HTML DOM tree is more efficient than the simple string matching approach. Nonetheless, previous tree-based pattern matching methods have problems by assuming that all HTML tags have the same values, assigning the same weight to each node in HTML trees. This paper proposes an enhanced tree matching algorithm that improves the tree edit distance method by considering the characteristics of HTML features. We assign different values to different HTML tree nodes according to their weights for displaying the corresponding data objects in the browser. Pattern matching of HTML patterns is done by obtaining the maximum mapping values of two HTML trees that are constructed with weighted node values from HTML data objects. Experiments are done over several Web commerce sites to evaluate the effectiveness of the proposed HTML tree matching algorithm.

Searching for relevant information on the web is an important aspect of distance learning. This activity is a challenge for visually impaired distance learners. While sighted people have the ability to filter information in a fast and non sequential way, blind persons rely on tools that process the information in a sequential way. Learning is slowed by screen readers which do not interact well with web pages. This paper introduces WAVES, a tool for the fast retrieval of information in a web page for blind and visually impaired people. The paper describes the WAVES prototype, a system that performs a page restructuring of webpages. The system analyzes webpages, identifies elements of interests from a webpage, evaluates their importance by using semantic information and visual cues, sorts them by importance and uses them to restructure the webpage so that data from the original webpage are presented to the reader in a concise format. A preliminary evaluation test of the prototype system has been performed with a sample set of users. The results of the preliminary test show an increase in speed and accuracy when the WAVES system has been used.

Searching for relevant information on the web is an important aspect of distance learning. This activity is a challenge for visually impaired distance learners. While sighted people have the ability to filter information in a fast and non sequential way, blind persons rely on tools that process the information in a sequential way. Learning is slowed by screen readers which do not interact well with web pages. This paper introduces WAVES, a tool for the fast retrieval of information in a web page for blind and visually impaired people. The paper describes the WAVES prototype, a system that performs a page restructuring of webpages. The system analyzes webpages, identifies elements of interests from a webpage, evaluates their importance by using semantic information and visual cues, sorts them by importance and uses them to restructure the webpage so that data from the original webpage are presented to the reader in a concise format. A preliminary evaluation test of the prototype system has been performed with a sample set of users. The results of the preliminary test show an increase in speed and accuracy when the WAVES system has been used.Request access from your librarian to read this article's full text.

This paper describes a dynamic rearrangement mechanism of Web page layouts. The aim of this research is to provide webmasters or administrators with a means of determining easily the best suited content layout of their Web pages in a short period of time by automatically testing and relocating the contents with a trend of user actions. We will become able to modify dynamically the structure and layout of Web pages on the client browsers using the information of user behaviors. Our method provides the means of changing the contents inside or between Web pages and updating the contents on Web browsers, and it shorten the period of time for optimization.KeywordsUser BehaviorCellular PhoneContent DeliveryContent Management SystemPage ViewThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

A Survey on Improving the Web Search Ranking by User Behavior Information

Each web page can be segmented into semantically coherent units that fulfill specific purposes. Though the task of automatic web page segmentation was introduced two decades ago, along with several applications in web content analysis, its foundations are still lacking. Specifically, the developed evaluation methods and datasets presume a certain downstream task, which led to a variety of incompatible datasets and evaluation methods. To address this shortcoming, we contribute two resources: (1) An evaluation framework which can be adjusted to downstream tasks by measuring the segmentation similarity regarding visual, structural, and textual elements, and which includes measures for annotator agreement, segmentation quality, and an algorithm for segmentation fusion. (2) The Webis-WebSeg-20 dataset, comprising 42,450~crowdsourced segmentations for 8,490~web pages, outranging existing sources by an order of magnitude. Our results help to better understand the mental segmentation model'' of human annotators: Among other things we find that annotators mostly agree on segmentations for all kinds of web page elements (visual, structural, and textual). Disagreement exists mostly regarding the right level of granularity, indicating a general agreement on the visual structure of web pages.

Due to the massive size of the hidden web, searching, retrieving and mining rich and high-quality data can be a daunting task. Moreover, with the presence of forms, data cannot be accessed easily. Forms are dynamic, heterogeneous and spread over trillions of web pages. Significant efforts have addressed the problem of tapping into the hidden web to integrate and mine rich data. Effective techniques, as well as application in special cases, are required to be explored to achieve an effective harvest rate. One such special area is atmospheric science, where hidden web crawling is least implemented, and crawler is required to crawl through the huge web to narrow down the search to specific data. In this study, an intelligent hidden web crawler for harvesting data in urban domains (IHWC) is implemented to address the relative problems such as classification of domains, prevention of exhaustive searching, and prioritizing the URLs. The crawler also performs well in curating pollution-related data. The crawler targets the relevant web pages and discards the irrelevant by implementing rejection rules. To achieve more accurate results for a focused crawl, ICHW crawls the websites on priority for a given topic. The crawler has fulfilled the dual objective of developing an effective hidden web crawler that can focus on diverse domains and to check its integration in searching pollution data in smart cities. One of the objectives of smart cities is to reduce pollution. Resultant crawled data can be used for finding the reason for pollution. The crawler can help the user to search the level of pollution in a specific area. The harvest rate of the crawler is compared with pioneer existing work. With an increase in the size of a dataset, the presented crawler can add significant value to emission accuracy. Our results are demonstrating the accuracy and harvest rate of the proposed framework, and it efficiently collect hidden web interfaces from large-scale sites and achieve higher rates than other crawlers.

The growing global incidence of mental health disorders underlines the urgent need for improved tools to enable early diagnosis and intervention. This study investigates the potential of machine learning models to predict mental health issues among college students by utilizing a dataset that includes a variety of demographic and behavioural characteristics. This study employs several Machine learning models, including Logistic Regression, Random Forest, Decision Tree, and XGBoost, using a dataset comprising demographic, behavioural, and self-reported mental health information. Data preprocessing involved cleansing, normalization, and feature selection to optimize model performance. Models were trained and validated using cross-validation, and their performance was measured using metrics such as accuracy, precision, and ROC-AUC scores. Machine learning models, particularly Logistic Regression, show significant potential for improving mental health assessments by providing early, accurate, and scalable predictions. This study is significant in addressing the rising mental health challenges among college students by leveraging machine learning (ML) for early detection and personalized care. Traditional diagnostic methods, often time-consuming and subjective, are enhanced by ML’s ability to process large datasets for faster, accurate, and scalable assessments. The Logistic Regression model achieved an accuracy of 85% and a precision of 81%, demonstrating its reliability for general mental health predictions. By integrating demographic, behavioural, and physiological data, the study promotes tailored interventions while emphasizing ethical considerations like privacy and transparency. Its findings can guide institutions and policymakers in developing data-driven mental health programs, fostering healthier academic environments and advancing mental health care.

Users engaged in information search often reformulate or modify their queries. This paper reports on an investigation of how task type and task situation influence users' query reformulation behavior. A controlled experiment was conducted with 48 participants, each working on six web search tasks classified into three types according to the task structure: Simple, Hierarchical and Parallel. We developed a taxonomy of query reformulation and used an automated method to detect the reformulations. Our results showed that Specialization was most frequently used in Simple tasks, and Word Substitution was most frequently used in Parallel tasks. After visiting and saving a useful web page, Generalization was less likely to be used while New query was more likely to be used. We also found that the effectiveness of each query reformulation type varied in different task types. The results of this study demonstrate the effect of task type on users' query reformulation behavior and have implications for the design of query suggestions that are offered to users during searching.