Web-scale Data Research Articles

Enriching the Fake News Process of Classifying Model using Deep Learning and Machine Learning Algorithm

Fake news detection has become a pressing issue due to the rapid propagation of misinformation through various online platforms. In recent years, due to the booming development of online social networks, fake news for various commercial and political purposes has been appearing in large numbers and widespread in the online world. An important goal in improving the trustworthiness of information in online social networks is to identify the fake news timely. This paper aims at investigating the principles, methodologies and algorithms for detecting fake news articles, creators and subjects from online social networks and evaluating the corresponding performance. Information carefulness on Internet, especially on social media, is an increasingly important concern, but web-scale data hampers, ability to identify, evaluate and correct such data, or so called "fake news," present in these platforms. In this paper, we propose a method for "fake news" detection and ways to apply it on Facebook, one of the most popular online social media platforms. This method uses Naive Bayes classification model to predict whether a post on Facebook will be labelled as real or fake.

Self-supervised multi-modal training from uncurated images and reports enables monitoring AI in radiology

The escalating demand for artificial intelligence (AI) systems that can monitor and supervise human errors and abnormalities in healthcare presents unique challenges. Recent advances in vision-language models reveal the challenges of monitoring AI by understanding both visual and textual concepts and their semantic correspondences. However, there has been limited success in the application of vision-language models in the medical domain. Current vision-language models and learning strategies for photographic images and captions call for a web-scale data corpus of image and text pairs which is not often feasible in the medical domain. To address this, we present a model named medical cross-attention vision-language model (Medical X-VL), which leverages key components to be tailored for the medical domain. The model is based on the following components: self-supervised unimodal models in medical domain and a fusion encoder to bridge them, momentum distillation, sentencewise contrastive learning for medical reports, and sentence similarity-adjusted hard negative mining. We experimentally demonstrated that our model enables various zero-shot tasks for monitoring AI, ranging from the zero-shot classification to zero-shot error correction. Our model outperformed current state-of-the-art models in two medical image datasets, suggesting a novel clinical application of our monitoring AI model to alleviate human errors. Our method demonstrates a more specialized capacity for fine-grained understanding, which presents a distinct advantage particularly applicable to the medical domain.

LiteWSEC: A Lightweight Framework for Web-Scale Spectral Ensemble Clustering

Spectral Clustering (SC) is an effective clustering method for its excellent performance in partitioning non-linearly distributed data. On the other hand, Ensemble Clustering (EC), a different clustering technology, can promote cluster quality by ensembling the results of base clusterings. In this work, we concentrate on an EC framework that utilizes SC as the base method. Nevertheless, SC suffers from scalability due to its high computational complexity in constructing the Laplacian graph and computing the corresponding eigendecomposition. In the past decades, many efforts have been made to it. However, SC suffers from the scalability issue in processing extensive data, especially in web-scale scenarios. Additionally, EC requires multiple clustering results as the ensemble bases, which further aggravates resource consumption. To address this issue, LiteWSEC, a simple yet efficient Lightweight Framework for Web-scale Spectral Ensemble Clustering, is proposed to cluster web-scale data with limited resource requirements. It adopts the Web-scale Spectral Clustering (WSC) as the base method, which has minimal space overhead without computing overall embedding explicitly. LiteWSEC is highly flexible in the memory requirement, which is adaptive to the available resource. It can partition web-scale data (e.g., <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$n = 8,000~k$</tex-math></inline-formula> ) in an resource-limited host (e.g., memory is restricted to 1 GB). Experiments on real-world, large-scale, and web-scale datasets demonstrate both the efficiency and effectiveness of LiteWSEC over state-of-the-art SC and EC methods.

Pre-trained Language Model-based Retrieval and Ranking for Web Search

Pre-trained language representation models (PLMs) such as BERT and Enhanced Representation through kNowledge IntEgration (ERNIE) have been integral to achieving recent improvements on various downstream tasks, including information retrieval. However, it is nontrivial to directly utilize these models for the large-scale web search due to the following challenging issues: (1) the prohibitively expensive computations of massive neural PLMs, especially for long texts in the web document, prohibit their deployments in the web search system that demands extremely low latency; (2) the discrepancy between existing task-agnostic pre-training objectives and the ad hoc retrieval scenarios that demand comprehensive relevance modeling is another main barrier for improving the online retrieval and ranking effectiveness; and (3) to create a significant impact on real-world applications, it also calls for practical solutions to seamlessly interweave the resultant PLM and other components into a cooperative system to serve web-scale data. Accordingly, we contribute a series of successfully applied techniques in tackling these exposed issues in this work when deploying the state-of-the-art Chinese pre-trained language model, i.e., ERNIE, in the online search engine system. We first present novel practices to perform expressive PLM-based semantic retrieval with a flexible poly-interaction scheme and cost-efficiently contextualize and rank web documents with a cheap yet powerful Pyramid-ERNIE architecture. We then endow innovative pre-training and fine-tuning paradigms to explicitly incentivize the query-document relevance modeling in PLM-based retrieval and ranking with the large-scale noisy and biased post-click behavioral data. We also introduce a series of effective strategies to seamlessly interwoven the designed PLM-based models with other conventional components into a cooperative system. Extensive offline and online experimental results show that our proposed techniques are crucial to achieving more effective search performance. We also provide a thorough analysis of our methodology and experimental results.

Investigating the Bioactive Conformation of Angiotensin II Using Markov State Modeling Revisited with Web-Scale Clustering.

Molecular dynamics simulation is a powerful technique for studying the structure and dynamics of biomolecules in atomic-level detail by sampling their various conformations in real time. Because of the long timescales that need to be sampled to study biomolecular processes and the big and complex nature of the corresponding data, relevant analyses of important biophysical phenomena are challenging. Clustering and Markov state models (MSMs) are efficient computational techniques that can be used to extract dominant conformational states and to connect those with kinetic information. In this work, we perform Molecular Dynamics simulations to investigate the free energy landscape of Angiotensin II (AngII) in order to unravel its bioactive conformations using different clustering techniques and Markov state modeling. AngII is an octapeptide hormone, which binds to the AT1 transmembrane receptor, and plays a vital role in the regulation of blood pressure, conservation of total blood volume, and salt homeostasis. To mimic the water-membrane interface as AngII approaches the AT1 receptor and to compare our findings with available experimental results, the simulations were performed in water as well as in water-ethanol mixtures. Our results show that in the water-ethanol environment, AngII adopts more compact U-shaped (folded) conformations than in water, which resembles its structure when bound to the AT1 receptor. For clustering of the conformations, we validate the efficiency of an inverted-quantized k-means algorithm, as a fast approximate clustering technique for web-scale data (millions of points into thousands or millions of clusters) compared to k-means, on data from trajectories of molecular dynamics simulations with reasonable trade-offs between time and accuracy. Finally, we extract MSMs using various clustering techniques for the generation of microstates and macrostates and for the selection of the macrostate representatives.

Fake News Detection Using Machine Learning

Abstract: The role of social media in our day to day life has increased rapidly in recent years. Information quality in social media is an increasingly important issue, but web-scale data hinders experts’ ability to assess and correct much of the inaccurate content, or “fake news”, present in these platforms. It is now used not only for social interaction, but also as an important platform for exchanging information and news. Twitter, Facebook a micro-blogging service, connects millions of users around the world and allows for the real-time propagation of information and news. The fake news on social media and various other media is wide spreading and is a matter of serious concern due to its ability to cause a lot of social and national damage with destruction impacts. A lot of research is already focused on detecting it. A human being is unable to detect all these fake news. Detecting fake news is an important step. This process will result in feature extraction and vectorization; we propose using Python scikit-learn library to perform tokenization and feature extraction of text data, because this library contains useful tools like Count Vectorizer and Tiff Vectorizer. Then, we will perform feature selection methods, to experiment and choose the best fit features to obtain the highest precision, according to confusion matrix results. A feature analysis then identifies features that are most predictive for crowdsourced and journalistic accuracy assessments, results of which are consistent with prior work. We aim to provide the user with the ability to classify the news as “fake” or “real”.

A Low-Cost High-Resolution Solid-State Lidar With Wavelength Division Multiplexed Components and Interleaved Orthogonal Polarization Grating Couplers

Laser diodearray and photo diode array combined through coarse wavelength division multiplexed (CWDM) are widely used in short-reach optical communication within massive web-scale data center. Those components can be used in solid-state photonics-based Lidar built upon silicon-on-insulator (SOI) waveguide as well, achieving significant cost saving. Beam steering can be naturally achieved through diffraction in output grating couplers. However, current design is hampered by low angular resolution. We demonstrate a novel design by combining outputs of two grating couplers with orthogonal polarizations in an interleaving manner. This scheme effectively doubles the angular resolution of on-chip Lidar.

GraphScope

Due to diverse graph data and algorithms, programming and orchestration of complex computation pipelines have become the major challenges to making use of graph applications for Web-scale data analysis. GraphScope aims to provide a one-stop and efficient solution for a wide range of graph computations at scale. It extends previous systems by offering a unified and high-level programming interface and allowing the seamless integration of specialized graph engines in a general data-parallel computing environment. As we will show in this demo, GraphScope enables developers to write sequential graph programs in Python and provides automatic parallel execution on a cluster. This further allows GraphScope to seamlessly integrate with existing data processing systems in PyData ecosystem. To validate GraphScope's efficiency, we will compare a complex, multi-staged processing pipeline for a real-life fraud detection task with a manually assembled implementation comprising multiple systems. GraphScope achieves a 2.86× speedup on a trillion-scale graph in real production at Alibaba.

Building a New Semantic Social Network Using Semantic Web-Based Techniques

Most people are more or less related to the web by participating in a kind of social networking site. Semantic Web technology plays a crucial role in these sites as they contain an enormous amount of data about ‎persons, pages, events, places, corporations, etc. This research is a Semantic Web application designed to create a new ‎semantic social community called Socialpedia. It links the already existing social public information to the newly ‎public ones. This information is linked with different information on the web to construct a new immense ‎data container. The resulting data container can be processed using a variety of Semantic Web techniques to produce ‎machine-understandable content. This content shows the promise of using integrated data to improve Web search and ‎Web-scale data analysis, unlike conventional search engines or social ones. This community involves obtaining data ‎from traditional users known as contributors or participants, linking data from existing social networks, extracting ‎structured data in triples using predefined ontologies, and finally querying and inferring such data to obtain ‎meaningful pieces of information. Socailpedia supports all popular functionalities of social networking websites ‎besides the enhanced features of the Semantic Web, providing advanced semantic search that acts as a semantic ‎search engine.

Fast and memory-efficient algorithms for high-order Tucker decomposition

Multi-aspect data appear frequently in web-related applications. For example, product reviews are quadruplets of the form (user, product, keyword, timestamp), and search-engine logs are quadruplets of the form (user, keyword, location, timestamp). How can we analyze such web-scale multi-aspect data on an off-the-shelf workstation with a limited amount of memory? Tucker decomposition has been used widely for discovering patterns in such multi-aspect data, which are naturally expressed as large but sparse tensors. However, existing Tucker decomposition algorithms have limited scalability, failing to decompose large-scale high-order ($$\ge $$ 4) tensors, since they explicitly materialize intermediate data, whose size grows exponentially with the order. To address this problem, which we call “Materialization Bottleneck,” we propose S-HOT, a scalable algorithm for high-order Tucker decomposition. S-HOT minimizes materialized intermediate data by using an on-the-fly computation, and it is optimized for disk-resident tensors that are too large to fit in memory. We theoretically analyze the amount of memory and the number of data scans required by S-HOT. Moreover, we empirically show that S-HOT handles tensors with higher order, dimensionality, and rank than baselines. For example, S-HOT successfully decomposes a real-world tensor from the Microsoft Academic Graph on an off-the-shelf workstation, while all baselines fail. Especially, in terms of dimensionality, S-HOT decomposes 1000 $$\times $$ larger tensors than baselines.

Editor’s Note - Special Issue on Deep vs. Shallow: Learning for Emerging Web-scale Data Computing and Applications

Editor’s Note - Special Issue on Deep vs. Shallow: Learning for Emerging Web-scale Data Computing and Applications

A hybrid GA-based RS-RES model for web-multimedia data management to identify determinants of credit rating status

To address practical problems for real-life applications for benefiting large-scale data sets and resources has absolute requirement in managing web-scale data. The development of an indicator for the operating sufficiency and financial status of Asian banks will be really necessary to better understand the stability of financial markets from managing distributed large-scale data. This study proposes an effective hybrid model with two stages, including: 1) it organises random forests and key reducts; and the core concept of rough-set rule exploration system (RS-RES) with a genetic algorithm setting for the benefit of feature-selection techniques to demonstrate various combinations of extracted key features; 2) the use of the RS-decisional LEM2 algorithm as effective evaluation approaches for measuring various model combinations. To make further verification, a real dataset was practically collected from the web-multimedia data from the called BANK-CREDIT database. The target dataset contains of 1,327 samples from Asian banks for credit ratings from 1993-2007. Our research results indicate that the model proposed in this study has a more outperformance than listing models for the standard of accuracy and standard deviation, which indicates clearly model superiority and suitability.

Exploiting investors social network for stock prediction in China's market

Recent works have shown that social media platforms are able to influence the trends of stock price movements. However, existing works have major focused on the U.S. stock market and lacked attention to certain emerging countries such as China, where retail investors dominate the market. In this regard, as retail investors are prone to be influenced by news or other social media, psychological and behavioral features extracted from social media platforms are thought to well predict stock price movements in the China's market. Recent advances in the investor social network in China enables the extraction of such features from web-scale data. In this paper, on the basis of tweets from Xueqiu, a popular Chinese Twitter-like social platform specialized for investors, we analyze features with regard to collective sentiment and perception on stock relatedness and predict stock price movements by employing nonlinear models. The features of interest prove to be effective in our experiments.

STEM: a suffix tree-based method for web data records extraction

To automatically extract data records from Web pages, the data record extraction algorithm is required to be robust and efficient. However, most of existing algorithms are not robust enough to cope with rich information or noisy data. In this paper, we propose a novel suffix tree-based extraction method (STEM) for this challenging task. First, we extract a sequence of identifiers from the tag paths of Web pages. Then, a suffix tree is built on top of this sequence and four refining filters are proposed to screen out data regions that might not contain data records. To evaluate model performance, we define an evaluation metric called pattern similarity and perform rigorous experiments on five real data sets. The promising experimental results have demonstrated that the proposed STEM is superior to the state-of-the-art algorithms like MDR, TPC and CTVS with respect to precision, recall and pattern similarity. Moreover, the time complexity of STEM is linear to the total number of HTML tags contained in Web pages, which indicates the potential applicability of STEM in a wide range of Web-scale data record extraction applications.

A parallel approximate SS-ELM algorithm based on MapReduce for large-scale datasets

Extreme Learning Machine (ELM) algorithm not only has gained much attention of many scholars and researchers, but also has been widely applied in recent years especially when dealing with big data because of its better generalization performance and learning speed. The proposal of SS-ELM (semi-supervised Extreme Learning Machine) extends ELM algorithm to the area of semi-supervised learning which is an important issue of machine learning on big data. However, the original SS-ELM algorithm needs to store the data in the memory before processing it, so that it could not handle large and web-scale data sets which are of frequent appearance in the era of big data. To solve this problem, this paper firstly proposes an efficient parallel SS-ELM (PSS-ELM) algorithm on MapReduce model, adopting a series of optimizations to improve its performance. Then, a parallel approximate SS-ELM Algorithm based on MapReduce (PASS-ELM) is proposed. PASS-ELM is based on the approximate adjacent similarity matrix (AASM) algorithm, which leverages the Locality-Sensitive Hashing (LSH) scheme to calculate the approximate adjacent similarity matrix, thus greatly reducing the complexity and occupied memory. The proposed AASM algorithm is general, because the calculation of the adjacent similarity matrix is the key operation in many other machine learning algorithms. The experimental results have demonstrated that the proposed PASS-ELM algorithm can efficiently process very large-scale data sets with a good performance, without significantly impacting the accuracy of the results.

Covering the Monitoring Network: A Unified Framework to Protect E-Commerce Security

Multimedia applications in smart electronic commerce (e-commerce), such as online trading and Internet marketing, always face security in storage and transmission of digital images and videos. This study addresses the problem of security in e-commerce and proposes a unified framework to analyze the security data. First, to allocate the definite security resources optimally, we build our e-commerce monitoring model as an undirected network, where a monitored node is a vertex of the graph and a connection between vertices is an undirected edge. Moreover, we aim to find a minimal cover for the monitoring network as the optimal solution of resource allocation, which is defined as the network monitoring minimization problem (NMM). This problem is proved to be NP-hard. Second, by analyzing the latent threats, we design a novel and trusted monitoring system that can integrate incident monitoring, data analysis, risk assessment, and security warnings. This system does not touch users’ privacy data. Third, we propose a sequential model-based risk assessment method, which can predict the risk according to the text semantics. Our experimental results on web scale data demonstrate that our system is flexible enough when monitoring, which also verify the effectiveness and efficiency of our system.

A Provably Efficient Algorithm for Separable Topic Discovery

We develop necessary and sufficient conditions and a novel provably consistent and efficient algorithm for discovering topics (latent factors) from observations (documents) that are realized from a probabilistic mixture of shared latent factors that have certain properties. Our focus is on the class of topic models in which each shared latent factor contains a novel word that is unique to that factor, a property that has come to be known as separability. Our algorithm is based on the key insight that the novel words correspond to the extreme points of the convex hull formed by the row-vectors of a suitably normalized word co-occurrence matrix. We leverage this geometric insight to establish polynomial computational and sample complexity bounds based on a few isotropic random projections of the rows of the normalized word co-occurrence matrix. Our proposed random-projections-based algorithm is naturally amenable to an efficient distributed implementation and is attractive for modern web-scale distributed data mining applications.

Natural language processing and the Now-or-Never bottleneck

Researchers, motivated by the need to improve the efficiency of natural language processing tools to handle web-scale data, have recently arrived at models that remarkably match the expected features of human language processing under the Now-or-Never bottleneck framework. This provides additional support for said framework and highlights the research potential in the interaction between applied computational linguistics and cognitive science.

Efficient path query and reasoning method based on rare axis

A new concept of rare axis based on statistical facts is proposed, and an evaluation algorithm is designed thereafter. For the nested regular expressions containing rare axes, the proposed algorithm can reduce its evaluation complexity from polynomial time to nearly linear time. The distributed technique is also employed to construct the navigation axis indexes for resource description framework (RDF) graph data. Experiment results in DrugBank and BioGRID show that this method can improve the query efficiency significantly while ensuring the accuracy and meet the query requirements on Web-scale RDF graph data.

Good, Great, Excellent: Global Inference of Semantic Intensities

Adjectives like good, great, and excellent are similar in meaning, but differ in intensity. Intensity order information is very useful for language learners as well as in several NLP tasks, but is missing in most lexical resources (dictionaries, WordNet, and thesauri). In this paper, we present a primarily unsupervised approach that uses semantics from Web-scale data (e.g., phrases like good but not excellent) to rank words by assigning them positions on a continuous scale. We rely on Mixed Integer Linear Programming to jointly determine the ranks, such that individual decisions benefit from global information. When ranking English adjectives, our global algorithm achieves substantial improvements over previous work on both pairwise and rank correlation metrics (specifically, 70% pairwise accuracy as compared to only 56% by previous work). Moreover, our approach can incorporate external synonymy information (increasing its pairwise accuracy to 78%) and extends easily to new languages. We also make our code and data freely available.