Implications of the Improvement in Atmospheric Fine Particles: A Case Study of COVID-19 Pandemic in Northern Taiwan

The outbreak of COVID-19 pandemic in northern Taiwan led to the implementation of Level 3 alert measures during 2021 and thereby impacted the air quality significantly, which provided an unprecedented opportunity to better understand the control strategies on air pollutants in the future. This study investigated the variations in sources, chemical characteristics and human health risks of PM2.5 comprehensively. The PM2.5 mass concentrations decreased from pre-alert to Level 3 alert by 49.4%, and the inorganic ions, i.e., NH4+, NO3- and SO42-, dropped even more by 71%, 90% and 52%, respectively. Nonetheless, organic matter (OM) and elemental carbon (EC) simply decreased by 36% and 13%, which caused the chemical composition of PM2.5 to change so that the carbonaceous matter in PM2.5 dominated instead of the inorganic ions. Correlation-based hierarchical clustering analysis further showed that PM2.5 was clustered with carbonaceous matter during the Level 3 alert, while that clustered with inorganic ions during both pre-alert and post-alert periods. Moreover, 6 sources of PM2.5 were identified by positive matrix factorization (PMF), in which secondary nitrate (i.e., aging traffic aerosols) exhibited the most significant decrease and yet primary traffic-related emissions, dominated by carbonaceous matter, changed insignificantly. This implied that secondary traffic-related aerosols could be easily controlled when traffic volume declined, while primary traffic source needs more efforts in the future, especially for the reduction of carbonaceous matter. Therefore, cleaner energy for vehicles is still needed. Assessments of both carcinogenic risk and non-carcinogenic risk induced by the trace elements in PM2.5 showed insignificant decrease, which can be attributed to the factories that did not shut down during Level 3 alert. This study serves as a metric to underpin the mitigation strategies of air pollution in the future and highlights the importance of carbonaceous matter for the reduction in PM2.5.

Scalable one-stage multi-view subspace clustering with dictionary learning

The clustering of large numbers of heterogeneous features is a hot topic in multi-view communities. Most existing multi-view clustering (MvC) methods employ matrix factorization or anchor strategies to handle large-scale datasets. The former operates on the original data and is, therefore, sensitive to noise and feature redundancy, which is reflected in the final clustering performance. The latter requires post -processing steps to generate the clustering results, which may be suboptimal owing to the isolation steps. To address the above problems, we propose one-stage multi-view subspace clustering with dictionary learning (OSMvSC). Specifically, we integrate dictionary learning, representation coefficient matrix learning, and matrix factorization as a unified learning framework, which directly learns the dictionary and representation coefficient matrix to encode the original multi-view data, and obtains the clustering results with linear time complexity without any postprocessing step. By manipulating the class centroid with the nuclear norm, a more compact and discriminative class centroid representation can be obtained to further improve clustering performance. An effective optimization algorithm with guaranteed convergence is designed to solve the proposed method. Substantial experiments on various real-world multi-view datasets demonstrate the effectiveness and superiority of the proposed method. The source code is available at https://github.com/justcallmewilliam/OSMvSC.(c) 2022 Elsevier B.V. All rights reserved.

Explainability of the COVID-19 epidemiological model with nonnegative tensor factorization.

The world is witnessing the devastating effects of the COVID-19 pandemic. Each country responded to contain the spread of the virus in the early stages through diverse response measures. Interpreting these responses and their patterns globally is essential to inform future responses to COVID-19 variants and future pandemics. A stochastic epidemiological model (SEM) is a well-established mathematical tool that helps to analyse the spread of infectious diseases through communities and the effects of various response measures. However, interpreting the outcome of these models is complex and often requires manual effort. In this paper, we propose a novel method to provide the explainability of an epidemiological model. We represent the output of SEM as a tensor model. We then apply nonnegative tensor factorization (NTF) to identify patterns of global response behaviours of countries and cluster the countries based on these patterns. We interpret the patterns and clusters to understand the global response behaviour of countries in the early stages of the pandemic. Our experimental results demonstrate the advantage of clustering using NTF and provide useful insights into the characteristics of country clusters.

Cohort-Based Analysis of Foreign Residents' Growth in Japan

In Japan, the cohort structure of foreign residents and its spatial distribution significantly change along with the acceptance of a specialized and technically skilled workforce. This study aims to analyze the transition of foreign residents' characteristics with time series with statistical viewpoints, in order to clarify the policy building for foreign migrants by local government. A nonnegative matrix factorization model (NMF) is applied to the cohort data of foreign residents in 47 Japanese prefectures in 2010, 2015 and 2020. By applying NMF to the ratio by cohort shifting, the common basis of foreign residents among the prefectures and its structures are obtained. The results show the cohort transition for foreign migrants with infants or children were significantly different, especially between Tohoku and Kyusyu regions from 2010 to 2020. The elderly cohort also had a significant change in Tohoku region from 2015 to 2020. Since the regions highlighted in this analysis include many depopulated areas, and the capacity of those local governments for the policy building would not be enough, they should be supported well by the national government.

A Matrix Factorization-Based Drug-Virus Link Prediction Method for SARS-CoV-2 Drug Prioritization

Matrix factorization (MF) has been widely used in drug discovery for link prediction, which aims to reveal new drug-target links by integrating drug-drug and target-target similarity information with a drug-target interaction matrix. The MF method is based on the assumption that similar drugs share similar targets and vice versa. However, one major disadvantage is that only one similarity metric is used in MF models, which is not enough to represent the similarity between drugs or targets. In this work, we develop a similarity fusion enhanced MF model to incorporate different types of similarity for novel drug-target link prediction. We apply the proposed model on a drug-virus association dataset for anti-COVID drug prioritization, and compare the performance with other existing MF models developed for COVID. The results show that the similarity fusion method can provide more useful information for drug-drug and virus-virus similarity and hence improve the performance of MF models. The top 10 drugs as prioritized by our model are provided, together with supporting evidence from literature. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Privacy preserving via multi-key homomorphic encryption in cloud computing

As the world grapples with the COVID-19 and its variants, multi-user collaboration by means of cloud computing is ubiquitous. How to make better use of cloud resources while preventing user privacy leakage has become particularly important. Multi-key homomorphic encryption(MKHE) can effectively deal with the privacy disclosure issue during the multi-user collaboration in the cloud computing setting. Firstly, we improve the DGHV homomorphic scheme by modifying the selection of key and the coefficients in encryption, so as to eliminate the restriction on the parity of the ciphertext modulus in the public key. On this basis, we further propose a DGHV-type MKHE scheme based on the number theory. In our scheme, an extended key is introduced for ciphertext extension, and we prove that it is efficient in performance analysis. The semantic security of our schemes is proved under the assumption of error-free approximate greatest common divisor and the difficulty of large integer factorization. Furthermore, the simulation experiments show the availability and computational efficiency of our MKHE scheme. Therefore, our scheme is suitable for the multi-user scenario in cloud environment. © 2023 Elsevier Ltd

Positive Matrix Factorization as Source Apportionment of Paddy Soil Heavy Metals in Black Shale Areas in Western Zhejiang Province, China

The source apportionment of pollutants is the key to preventing and controlling the pollution caused by heavy metals in soils. The aim of this study was to investigate the main sources of heavy metals in the soils of black shale areas in western Zhejiang, China. Based on geostatistical spatial analysis, this research employed positive matrix factorization (PMF) for the source apportionment of heavy metals in paddy soil. The results showed that contaminated arable soils were concentrated in the western and southern study areas. At least five major sources of heavy metals were screened in this study: natural sources (39.66%), traffic emissions (32.85%), industrial emissions (9.23%), agricultural activities (9.17%), and mining (9.10%). To be specific, Cd was mainly from mining;As originated from agricultural inputs such as fertilizers and pesticides;and Hg, as an industrial pollutant, was transported by atmospheric deposition in the study area. The accumulation of Pb, Zn, and Cu was mainly influenced by natural sources and anthropogenic sources, i.e., traffic emissions, while that of Cr and Ni was controlled by natural sources.

DRaW: prediction of COVID-19 antivirals by deep learning-an objection on using matrix factorization.

BACKGROUND: Due to the high resource consumption of introducing a new drug, drug repurposing plays an essential role in drug discovery. To do this, researchers examine the current drug-target interaction (DTI) to predict new interactions for the approved drugs. Matrix factorization methods have much attention and utilization in DTIs. However, they suffer from some drawbacks. METHODS: We explain why matrix factorization is not the best for DTI prediction. Then, we propose a deep learning model (DRaW) to predict DTIs without having input data leakage. We compare our model with several matrix factorization methods and a deep model on three COVID-19 datasets. In addition, to ensure the validation of DRaW, we evaluate it on benchmark datasets. Furthermore, as an external validation, we conduct a docking study on the COVID-19 recommended drugs. RESULTS: In all cases, the results confirm that DRaW outperforms matrix factorization and deep models. The docking results approve the top-ranked recommended drugs for COVID-19. CONCLUSIONS: In this paper, we show that it may not be the best choice to use matrix factorization in the DTI prediction. Matrix factorization methods suffer from some intrinsic issues, e.g., sparsity in the domain of bioinformatics applications and fixed-unchanged size of the matrix-related paradigm. Therefore, we propose an alternative method (DRaW) that uses feature vectors rather than matrix factorization and demonstrates better performance than other famous methods on three COVID-19 and four benchmark datasets.

Understanding components of mobility during the COVID-19 pandemic.

Travel restrictions have proven to be an effective strategy to control the spread of the COVID-19 epidemics, in part because they help delay disease propagation across territories. The question, however, as to how different types of travel behaviour, from commuting to holiday-related travel, contribute to the spread of infectious diseases remains open. Here, we address this issue by using factorization techniques to decompose the temporal network describing mobility flows throughout 2020 into interpretable components. Our results are based on two mobility datasets: the first is gathered from Danish mobile network operators; the second originates from the Facebook Data-For-Good project. We find that mobility patterns can be described as the aggregation of three mobility network components roughly corresponding to travel during workdays, weekends and holidays, respectively. We show that, across datasets, in periods of strict travel restrictions the component corresponding to workday travel decreases dramatically. Instead, the weekend component, increases. Finally, we study how each type of mobility (workday, weekend and holiday) contributes to epidemics spreading, by measuring how the effective distance, which quantifies how quickly a disease can travel between any two municipalities, changes across network components. This article is part of the theme issue 'Data science approaches to infectious disease surveillance'.

Multi-label Emotion Classification of COVID-19 Tweets with Deep Learning and Topic Modelling

The COVID-19 pandemic has become one of the severe diseases in recent years. As it majorly affects the common livelihood of people across the universe, it is essential for administrators and healthcare professionals to be aware of the views of the community so as to monitor the severity of the spread of the outbreak. The public opinions are been shared enormously in microblogging media like twitter and is considered as one of the popular sources to collect public opinions in any topic like politics, sports, entertainment etc., This work presents a combination of Intensity Based Emotion Classification Convolution Neural Network (IBEC-CNN) model and Non-negative Matrix Factorization (NMF) for detecting and analyzing the different topics discussed in the COVID-19 tweets as well the intensity of the emotional content of those tweets. The topics were identified using NMF and the emotions are classified using pretrained IBEC-CNN, based on predefined intensity scores. The research aimed at identifying the emotions in the Indian tweets related to COVID-19 and producing a list of topics discussed by the users during the COVID-19 pandemic. Using the Twitter Application Programming Interface (Twitter API), huge numbers of COVID-19 tweets are retrieved during January and July 2020. The extracted tweets are analyzed for emotions fear, joy, sadness and trust with proposed Intensity Based Emotion Classification Convolution Neural Network (IBEC-CNN) model which is pretrained. The classified tweets are given an intensity score varies from 1 to 3, with 1 being low intensity for the emotion, 2 being the moderate and 3 being the high intensity. To identify the topics in the tweets and the themes of those topics, Non-negative Matrix Factorization (NMF) has been employed. Analysis of emotions of COVID-19 tweets has identified, that the count of positive tweets is more than that of count of negative tweets during the period considered and the negative tweets related to COVID-19 is less than 5%. Also, more than 75% negative tweets expressed sadness, fear are of low intensity. A qualitative analysis has also been conducted and the topics detected are grouped into themes such as economic impacts, case reports, treatments, entertainment and vaccination. The results of analysis show that the issues related to the pandemic are expressed different emotions in twitter which helps in interpreting the public insights during the pandemic and these results are beneficial for planning the dissemination of factual health statistics to build the trust of the people. The performance comparison shows that the proposed IBEC-CNN model outperforms the conventional models and achieved 83.71% accuracy. The % of COVID-19 tweets that discussed the different topics vary from 7.45% to 26.43% on topics economy, Statistics on cases, Government/Politics, Entertainment, Lockdown, Treatments and Virtual Events. The least number of tweets discussed on politics/government on the other hand the tweets discussed most about treatments. © 2023 CRL Publishing. All rights reserved.

Causal Discovery in High-dimensional, Multicollinear Datasets.

As the cost of high-throughput genomic sequencing technology declines, its application in clinical research becomes increasingly popular. The collected datasets often contain tens or hundreds of thousands of biological features that need to be mined to extract meaningful information. One area of particular interest is discovering underlying causal mechanisms of disease outcomes. Over the past few decades, causal discovery algorithms have been developed and expanded to infer such relationships. However, these algorithms suffer from the curse of dimensionality and multicollinearity. A recently introduced, non-orthogonal, general empirical Bayes approach to matrix factorization has been demonstrated to successfully infer latent factors with interpretable structures from observed variables. We hypothesize that applying this strategy to causal discovery algorithms can solve both the high dimensionality and collinearity problems, inherent to most biomedical datasets. We evaluate this strategy on simulated data and apply it to two real-world datasets. In a breast cancer dataset, we identified important survival-associated latent factors and biologically meaningful enriched pathways within factors related to important clinical features. In a SARS-CoV-2 dataset, we were able to predict whether a patient (1) had Covid-19 and (2) would enter the ICU. Furthermore, we were able to associate factors with known Covid-19 related biological pathways.

Integrated Data Analysis Uncovers New COVID-19 Related Genes and Potential Drug Re-Purposing Candidates.

The COVID-19 pandemic is an acute and rapidly evolving global health crisis. To better understand this disease's molecular basis and design therapeutic strategies, we built upon the recently proposed concept of an integrated cell, iCell, fusing three omics, tissue-specific human molecular interaction networks. We applied this methodology to construct infected and control iCells using gene expression data from patient samples and three cell lines. We found large differences between patient-based and cell line-based iCells (both infected and control), suggesting that cell lines are ill-suited to studying this disease. We compared patient-based infected and control iCells and uncovered genes whose functioning (wiring patterns in iCells) is altered by the disease. We validated in the literature that 18 out of the top 20 of the most rewired genes are indeed COVID-19-related. Since only three of these genes are targets of approved drugs, we applied another data fusion step to predict drugs for re-purposing. We confirmed with molecular docking that the predicted drugs can bind to their predicted targets. Our most interesting prediction is artenimol, an antimalarial agent targeting ZFP62, one of our newly identified COVID-19-related genes. This drug is a derivative of artemisinin drugs that are already under clinical investigation for their potential role in the treatment of COVID-19. Our results demonstrate further applicability of the iCell framework for integrative comparative studies of human diseases.

SmartTensors: Unsupervised and Physics-informed Machine Learning Framework for the Geoscience Applications

SmartTensors (https://github.com/SmartTensors) is a novel framework for unsupervised and physics-informed machine learning for geoscience applications. The methods in SmartTensors AI platform are developed using advanced matrix/tensor factorization constrained by penalties enforcing robustness and interpretability (e.g., nonnegativity, sparsity, physics, and mathematical constraints;etc.). This framework has been applied to analyze diverse datasets related to a wide range of problems: from COVID-19 to wildfires and climate. Here, we will focus on the analysis of geothermal prospectivity of the Great Basin, U.S. The basin covers a vast area that is yet to be thoroughly explored to discover new geothermal resources. The available regional geochemical data are expected to provide critical information about the geothermal reservoir properties in the basin, including temperature, fluid/heat flow, boundary conditions, and spatial extent. The geochemical data may also include hidden (latent) information that is a proxy for geothermal prospectivity. We processed the sparse geochemical dataset of 18 geochemical attributes observed at 14,341 locations. The data are analyzed using our GeoThermalCloud toolbox for geothermal exploration (https://github.com/SmartTensors/GeoThermalCloud.jl) whichis also a part of the SmartTensors framework. An unsupervised machine learning using non-negative matrix factorization with customized k-means clustering (NMFk) as implemented in SmartTensors identified three hidden geothermal signatures representing low-, medium-, and high-temperature reservoirs, respectively (Fig). NMFk also evaluated the probability of occurrence of these types of resources through the studied region. NMFk also reconstructed attributes from sparse into continuous over the study domain. Future work will add in the ML analyses other regional- and site-scale datasets including geological, geophysical, and geothermal attributes. © 2022 Society of Exploration Geophysicists and the American Association of Petroleum Geologists.

Recommendations in E-Commerce Systems Based on Deep Matrix Factorization

E-commerce systems (including online shopping, entertainment, etc.) play an increasingly important role and have become popular in digital life. These systems have also become one of the cores, and vital issues for many businesses, especially from the recent COVID-19 pandemic, the importance of online e-commerce systems are very necessary. Techniques in recommendation systems are widely used to support users in finding suitable products/items in online systems. This work proposes using deep matrix factorization for recommendation in online e-commerce systems. We provide a detailed architecture of a deep matrix factorization as well as make a comparison with the standard matrix factorization model. Experimental results on ten published data sets show that the deep matrix factorization model can work well for recommendations in online e-commerce systems. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Drug repositioning for SARS-CoV-2 by Gaussian kernel similarity bilinear matrix factorization.

Coronavirus disease 2019 (COVID-19), a disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is currently spreading rapidly around the world. Since SARS-CoV-2 seriously threatens human life and health as well as the development of the world economy, it is very urgent to identify effective drugs against this virus. However, traditional methods to develop new drugs are costly and time-consuming, which makes drug repositioning a promising exploration direction for this purpose. In this study, we collected known antiviral drugs to form five virus-drug association datasets, and then explored drug repositioning for SARS-CoV-2 by Gaussian kernel similarity bilinear matrix factorization (VDA-GKSBMF). By the 5-fold cross-validation, we found that VDA-GKSBMF has an area under curve (AUC) value of 0.8851, 0.8594, 0.8807, 0.8824, and 0.8804, respectively, on the five datasets, which are higher than those of other state-of-art algorithms in four datasets. Based on known virus-drug association data, we used VDA-GKSBMF to prioritize the top-k candidate antiviral drugs that are most likely to be effective against SARS-CoV-2. We confirmed that the top-10 drugs can be molecularly docked with virus spikes protein/human ACE2 by AutoDock on five datasets. Among them, four antiviral drugs ribavirin, remdesivir, oseltamivir, and zidovudine have been under clinical trials or supported in recent literatures. The results suggest that VDA-GKSBMF is an effective algorithm for identifying potential antiviral drugs against SARS-CoV-2.

Size-resolved Compositional Analysis and Source Apportionment of Submicron Aerosol during Lockdown Period Using HR-ToF-AMS

The size-resolved compositional analysis of non-refractory submicron aerosol (NR-PM1) was conducted using the Aerodyne High-Resolution Time-of-Flight Aerosol Mass Spectrometer (HR-ToF-AMS) instrument over Pune, India during the COVID-19 lockdown period. The aerosol composition data shows the predominant presence of organics (Org) in the mass fraction followed by sulfate, ammonium, nitrate, and chloride during the pre-lockdown and lockdown periods. The size-resolved analysis showed the unimodal size distribution of organic and inorganic constituents with peaks at 550 nm, implying the dominant presence of mixed and aged aerosol species. The stoichiometric neutralization analysis showed the almost neutralized nature of submicron aerosol with an average aerosol neutralization ratio (ANR) of 0.8. The back trajectories, cluster analysis, and potential source contribution function (PSCF) showed the industrial belt located in the western part of the study location to be the potential source regions of NR-PM1. Positive matrix factorization (PMF) analyses have been applied to investigate the source apportionments of organic aerosols (OA). Four distinct OA factors, i.e., hydrocarbon-like OA (HOA), biomass burning OA (BBOA), low-volatile oxygenated OA (LVOOA), and semi-volatile oxygenated OA (SVOOA) were identified during the study period. Among these factors, HOA contributes nearly a quarter to the OA mass, and OOA accounted for nearly 60% of the total OA mass. The high-resolution positive matrix factorization (HR-PMF) analysis and the elemental ratios of H/C, O/C, and OM/OC showed distinct characteristics during different periods. The density of organic aerosol has been estimated using the elemental ratios and found to be 1.14, 1.28, and 1.35 respectively during the different lockdown periods, similar to 1.30 g cm–3 as mentioned in the literature. This study provides new insights into the chemical composition and source apportionment of the organic fraction of submicron aerosols for the first time over Pune using HR-ToF-AMS and HR-PMF.

Post COVID-19 pandemic recovery of intracity human mobility in Wuhan: Spatiotemporal characteristic and driving mechanism.

After successfully inhibiting the first wave of COVID-19 transmission through a city lockdown, Wuhan implemented a series of policies to gradually lift restrictions and restore daily activities. Existing studies mainly focus on the intercity recovery under a macroscopic view. How does the intracity mobility return to normal? Is the recovery process consistent among different subareas, and what factor affects the post-pandemic recovery? To answer these questions, we sorted out policies adopted during the Wuhan resumption, and collected the long-time mobility big data in 1105 traffic analysis zones (TAZs) to construct an observation matrix (A). We then used the nonnegative matrix factorization (NMF) method to approximate A as the product of two condensed matrices (WH). The column vectors of W matrix were visualized as five typical recovery curves to reveal the temporal change. The row vectors of H matrix were visualized to identify the spatial distribution of each recovery type, and were analyzed with variables of population, GDP, land use, and key facility to explain the recovery driving mechanisms. We found that the "staggered time" policies implemented in Wuhan effectively staggered the peak mobility of several recovery types ("staggered peak"). Besides, different TAZs had heterogeneous response intensities to these policies ("staggered area") which were closely related to land uses and key facilities. The creative policies taken by Wuhan highlight the wisdom of public health crisis management, and could provide an empirical reference for the adjustment of post-pandemic intervention measures in other cities.

Adverse effect signature extraction and prediction for drugs treating COVID-19.

Given the considerable cost of drug discovery, drug repurposing is becoming attractive as it can effectively shorten the development timeline and reduce the development cost. However, most existing drug-repurposing methods omitted the heterogeneous health conditions of different COVID-19 patients. In this study, we evaluated the adverse effect (AE) profiles of 106 COVID-19 drugs. We extracted four AE signatures to characterize the AE distribution of 106 COVID-19 drugs by non-negative matrix factorization (NMF). By integrating the information from four distinct databases (AE, bioassay, chemical structure, and gene expression information), we predicted the AE profiles of 91 drugs with inadequate AE feedback. For each of the drug clusters, discriminant genes accounting for mechanisms of different AE signatures were identified by sparse linear discriminant analysis. Our findings can be divided into three parts. First, drugs abundant with AE-signature 1 (for example, remdesivir) should be taken with caution for patients with poor liver, renal, or cardiac functions, where the functional genes accumulate in the RHO GTPases Activate NADPH Oxidases pathway. Second, drugs featuring AE-signature 2 (for example, hydroxychloroquine) are unsuitable for patients with vascular disorders, with relevant genes enriched in signal transduction pathways. Third, drugs characterized by AE signatures 3 and 4 have relatively mild AEs. Our study showed that NMF and network-based frameworks contribute to more precise drug recommendations.

Approximating Sparse Semi-nonnegative Matrix Factorization for X-Ray Covid-19 Image Classification

Medical imaging has been intensively used to help the radiologists do the correct diagnosis for the COVID-19 disease. In particular, chest X-ray imaging is one of the prevalent information sources for COVID-19 diagnosis. The obtained images can be viewed as numerical data and processed by non-negative matrix factorization (NMF) algorithms, one of the available numerical data analysis tools. In this work, we propose a new sparse semi-NMF algorithm that can classify the patients into COVID-19 and normal patients, based on chest X-ray images. We show that the huge volume of data resulting from X-ray images can be significantly reduced without significant loss of classification accuracy. Then, we evaluate our algorithm by carrying out an experiment on a publicly available dataset, having a known chest X-ray image bi-partition. Experimental results demonstrate that the proposed sparse semi-NMF algorithm can predict COVID-19 patients with high accuracy,compared to state-of-the-art algorithms.

GOCPT: Generalized Online Canonical Polyadic Tensor Factorization and Completion

Low-rank tensor factorization or completion is well-studied and applied in various online settings, such as online tensor factorization (where the temporal mode grows) and online tensor completion (where incomplete slices arrive gradually). However, in many real-world settings, tensors may have more complex evolving patterns: (i) one or more modes can grow;(ii) missing entries may be filled;(iii) existing tensor elements can change. Existing methods cannot support such complex scenarios. To fill the gap, this paper proposes a Generalized Online Canonical Polyadic (CP) Tensor factorization and completion framework (named GOCPT) for this general setting, where we maintain the CP structure of such dynamic tensors during the evolution. We show that existing online tensor factorization and completion setups can be unified under the GOCPT framework. Furthermore, we propose a variant, named GOCPTE, to deal with cases where historical tensor elements are unavailable (e.g., privacy protection), which achieves similar fitness as GOCPT but with much less computational cost. Experimental results demonstrate that our GOCPT can improve fitness by up to 2.8% on the JHU Covid data and 9.2% on a proprietary patient claim dataset over baselines. Our variant GOCPTE shows up to 1.2% and 5.5% fitness improvement on two datasets with about 20% speedup compared to the best model. © 2022 International Joint Conferences on Artificial Intelligence. All rights reserved.