A drug recommender system Based on Collaborative Filtering for Covid-19 patients

The epidemic caused by a new mutation of the coronavirus family called Covid-19 has created a global crisis involving all the world's countries. This disease has become a severe danger to everyone due to its unknown nature, high spread, and inability to detect the infected. In this regard, one of the important issues facing patients with Covid-19 is the prescription of Drugs according to the severity of the disease and considering the records of underlying diseases in people. In recent years, recommender systems have been developed significantly along with the advancement in information technology and artificial intelligence, which is one of its applications in various fields of medical sciences. Among them, we can refer to recommending systems for the prevention, control, and treatment of diseases. In this research, using the collaborative filtering approach as one of the types of recommender systems as well as the K-means clustering algorithm, a Drug recommendation system for patients with Covid-19 in the treatment stage of the disease is presented. The results of this research show that this recommender system has an acceptable performance based on the evaluation criteria of precision, recall, and F1-score compared to the opinions of experts in this field. © 2023 IEEE.

Tracking the Evolution of Clusters in Social Media Streams

Tracking the evolution of clusters in social media streams is becoming increasingly important for many applications, such as early detection and monitoring of natural disasters or pandemics. In contrast to clustering on a static set of data, streaming data clustering does not have a global view of the complete data. The local (or partial) view in a high-speed stream makes clustering a challenging task. In this paper, we propose a novel density peak based algorithm, TStream, for tracking the evolution of clusters and outliers in social media streams, via the evolutionary actions of cluster adjustment, emergence, disappearance, split, and merge. TStream is based on a temporal decay model and text stream summarisation. The decay model captures the decreasing importance of textual documents over time. The stream summarisation compactly represents them with the help of cells (aka micro-clusters) in the memory. We also propose a novel efficient index called shared dependency tree (aka SD-Tree) based on the ideas of density peak and shared dependency. It maintains the dynamic dependency relationships in TStream and thereby improves the overall efficiency. We conduct extensive experiments on five real datasets. TStream outperforms the existing state-of-the-art solutions based on MStream, MStreamF, EDMStream, OSGM, and EStream, in terms of cluster mapping measure (CMM) by up to 17.8%, 18.6%, 6.9%, 16.4%, and 20.1%, respectively. It is also significantly more efficient than MStream, MStreamF, OSGM, and EStream, in terms of response time and throughput.

PACMAN: A Framework for Pulse Oximeter Digit Detection and Reading in A Low-Resource Setting

In light of the COVID-19 pandemic, patients were required to manually input their daily oxygen saturation (SpO2) and pulse rate (PR) values into a health monitoring system—unfortunately, such a process trend to be an error in typing. Several studies attempted to detect the physiological value from the captured image using optical character recognition (OCR). However, the technology has limited availability with high cost. Thus, this study aimed to propose a novel framework called PACMAN (Pandemic Accelerated Human-Machine Collaboration) with a low-resource deep learning-based computer vision. We compared state-of-the-art object detection algorithms (scaled YOLOv4, YOLOv5, and YOLOR), including the commercial OCR tools for digit recognition on the captured images from the pulse oximeter display. All images were derived from crowdsourced data collection with varying quality and alignment. YOLOv5 was the best-performing model against the given model comparison across all datasets, notably the correctly orientated image dataset. We further improved the model performance with the digits auto-orientation algorithm and applied a clustering algorithm to extract SpO2 and PR values. The accuracy performance of YOLOv5 with the implementations was approximately 81.0-89.5%, which was enhanced compared to without any additional implementation. Accordingly, this study highlighted the completion of the PACMAN framework to detect and read digits in real-world datasets. The proposed framework has been currently integrated into the patient monitoring system utilized by hospitals nationwide. IEEE

Research and Application of Online Course Recommendation System Based on TF-IDF Algorithm

In recent years, with the rapid development of Internet technology, a large number of online learning resources have emerged. Especially affected by the COVID-19 epidemic, online learning has become a very effective learning means. However, a large number of learning platforms and massive online teaching resources have the following three problems: 1) The quality of these courses is uneven and the evaluation standards are different;2) There are so many similar courses that it is difficult for learners to distinguish them;3) These classes are lack of unity and integration, and it is hard to recommend any hierarchical, coherent and systematic course resources to learners. Therefore, a recommendation model based on TF-IDF algorithm is designed to extract personalized-featured courses, use the nearest neighbor similarity to cluster the similarity of similar courses, and conduct the featured portrait of learners to realize online courses recommendation. Combined with the model design, this paper presents a tag-based online course resource recommendation system, which can fully explore learners' explicit and implicit preferences according to course tags, and recommend satisfactory MOOC resources for them with good application value. © 2022 IEEE.

Co-Evolutionary Location-Routing Model of Medical Isolation Areas for Major Public Health Emergencies Considering the Uncertainty of Future Traffic Environment

With the continuous development of the COVID-19 pandemic, the selection of locations for medical isolation areas has not always been optimal for the timely transportation of infected people, or those suspected of being infected. This has resulted in failure to control the rate of spread of infection cases in time. To address this problem, this paper proposes a co-evolutionary location-routing optimization (CELRO) model of medical isolation areas for use in major public health emergencies to develop a rapid location-routing scheme for epidemic isolation, including the selection of locations of medical isolation facilities per area and the optimal route per vehicle to each infected person. Specifically, this paper solves the following two sub-problems: (i) calculate the shortest transportation times and corresponding routes from any medical isolation area to any person infected or suspected of being infected, and (ii) calculate the location scheme for distribution of isolation areas. Different from previous studies, the vehicle operating characteristics and the interference of uncertainty of the traffic environment are considered in the proposed model. To find an appropriate scheme for location of medical isolation areas with the shortest travel times, a co-evolutionary clustering algorithm (CECA), which is a combination of some separated evolutionary programming operations, is proposed to solve the model. Various network sizes and uncertainty combinations are used to design some comparative tests, which aim to verify the effectiveness of the proposed model. In the experiment section, CELRO reduced travel time by at least 14% compared with other methods. This finding can provide an effective theoretical basis for optimizing the spatial layout of medical isolation areas or the location planning of new medical facilities. © National Academy of Sciences.

Research on Financial Systemic Risk Assessment Algorithm Based on Risk Data Fuzzy Cluster Analysis

The financial industry is a high-risk industry. Once the financial industry risk happen, it will affect the economic development. Ensuring the safe, efficient and steady operation of finance and preventing systemic financial risks are the urgent needs of China's opening up to the outside world and building a well-off society in an all-round way. Stable and efficient economic development is the basis of financial risk prevention and control, which is the inherent requirement of high-quality economic development. Strengthening macro-prudential management has become the core content of financial regulatory reform in major international organizations and economies after the international coronavirus outbreak and preventing systemic financial risks is the fundamental goal of macro-prudential management. This paper takes the assessment and monitoring of China's systemic financial risks as the research object, and proposes an assessment algorithm of systemic financial risks based on risk data fuzzy clustering analysis. The established financial systemic risk measurement method can identify risks to a certain extent, deeply understand the nature, root and key areas of systemic financial risks, and build a long-term mechanism to prevent and resolve systemic financial risks. © 2022 IEEE.

Learning the Protein Language Model of SARS-CoV-2 Spike Proteins

Ahstract-SARS-CoV-2 virus has long been evolving posing an increased risk in terms of infectivity and transmissibility which causes greater impact in communities worldwide. With the surge of collected SARS-CoV-2 sequences, studies found out that most of the emerging variants are linked to increased mutations in the spike (S) protein as observed in Alpha, Beta, Gamma, and Delta variants. Multiple approaches on genomic surveillance have been performed to monitor the mutational status and spread of the virus however most are heavily dependent on labels attributed to these sequences. Hence, this study features a system that has the capability to learn the protein language model of SARS-CoV-2 spike proteins, based on a bidirectional long-short term memory (BiLSTM) recurrent neural network, using sequence data alone. Upon obtaining the sequence embedding from the model, observed clusters are generated using the Leiden clustering algorithm and is visualized to monitor similarities between variants in terms of grammatical probability and semantic change. Additionally, the system measures the validity of a user-generated next-generation sequence capturing potential sequence mutations indicative of viral escape, particularly mutations by substitutions. Further studies on methods uncovering semantic rules that govern spike proteins are recommended to learn more about other viral characteristics conclusive of the future of the COVID-19 pandemic. © 2023 IEEE.

An Improved Accuracy in Anticipating the User Nature using a novel logistic regression algorithm throughout the pandemic across online social media based on Indian metrics over Support Vector Clustering Algorithm

This study's primary goal is to apply machine learning classifier techniques to raise the intensity percentage of user nature detection in order to detect the impact of coronavirus on Twitter users by comparing Novel Logistic Regression and Support Vector Clustering algorithms. Materials and Methods: The accuracy percentage with a confidence interval of 95% and G-power (value =0.8) was determined many times using the LR method with test size =10 and the SVC algorithm with test size =10. The likelihood that an item belongs to one category or another is predicted using a LR model. Support Vector Clustering algorithm generates a line or hyperplane that divides the data into categories. Results and Discussion: LR model has greater efficiency (91%) when compared to Support Vector Clustering (59%). Two groups are numerically unimportant, according to the data obtained with a coefficient of determination of p=0.121 (p>0.05). Conclusion: LR performs substantially better than the Support Vector Clustering. © 2022 IEEE.

Dynamic spillovers between clean energy and non-ferrous metals markets in China: A network-based analysis during the COVID-19 pandemic

Due to the close production link between clean energy and non-ferrous metals, their price and market dynamics can easily affect one another through production costs. Furthermore, with the increased financialization of clean energy and non-ferrous metals markets, investment risk can easily spread between them. Therefore, this paper intends to explore the risk contagion between the two markets through the spillover index model and the minimum spanning tree (MST) method. Employing the data collected in China, this paper quantifies the magnitude of risk transfer by the volatility spillovers of eight clean energy stock markets as identified in The Energy Conservation and Environmental Protection Clean Industry Statistical Classification 2021 and the eight corresponding non-ferrous metals futures markets, while fully considering the heterogeneity between sub-markets. First, we find that risk is mainly transmitted from clean energy to non-ferrous metals. Second, this paper identifies not only the most influential market but also the shortest path of risk contagion based on the MST topology analysis. Last, the empirical results show that the COVID-19 has increased the scale of risk transmission between the two markets and their connectivity. During the COVID-19 period, the shortest path between the two markets shifted from "hydropower–gold” to "smart grid–zinc”, and the systematically influential markets correspondingly become smart grid and zinc. The results obtained in this paper might have practical implications for policymakers seeking to achieve effective risk management, which could also facilitate investors for diversification benefits. © 2023 Elsevier Ltd

Real-time Control of the Air Volume in Ventilation Facilities by Limiting Concentration with Cluster Algorithms

Ventilation improves indoor air quality and reduces airborne infections. It is particularly important at present because of the COVID-19 pandemic. Commercially available ventilation facilities can only be instantly turned on/off or at a set time with adjustable air volumes (high, middle, and low). However, maintaining the indoor carbon dioxide concentration while reducing the energy consumption of these facilities is challenging. Hence, this study developed clustering algorithms to determine the carbon dioxide concentration limit thus enabling real-time air volume adjustment. These limit values were set using the existing energy recovery ventilation (ERV) controller. In the experiment, dual estimation was adopted, and the constructing building energy models from data were sampled at a low rate to compare that the ventilation facilities are only turned on/off. In addition, switching control is closely related to fuzzy control;that is, fuzzy control can be considered a smooth version of switching control. The experimental results indicated that the limits of 600 and 700 ppm were suitable to effectively control the real-time air volume based on the ERV operation. An ERV-based carbon dioxide concentration limit reduced the energy consumption of ventilation facilities by 11%implications of this study. IEEE

Feature Selection For High Dimensional Data Using Supervised Machine Learning Techniques

In recent years, feature selection has become an increasingly active field of data science and machine learning research. Most of the datasets that are being used nowadays for various machine learning tasks consist of thousands of features (columns), which make them extremely complex and difficult to work with. In this paper, we propose a feature selection methodological pipeline that can be used to reduce the complexity of high dimensional datasets through the elimination of redundant and/or non-informative features as well as to improve the performance of machine learning models which are trained on high dimensional datasets. The proposed method has been applied to high-dimensional biomedical data and compared against a classic filter-based feature selection algorithm. Specifically, the method was applied to gene expression profiles of a single-cell RNA-seq dataset from healthy and infected by covid-19 human samples. © 2022 IEEE.

Entropy Based Clustering of Viral Sequences

Clustering viral sequences allows us to characterize the composition and structure of intrahost and interhost viral populations, which play a crucial role in disease progression and epidemic spread. In this paper we propose and validate a new entropy based method for clustering aligned viral sequences considered as categorical data. The method finds a homogeneous clustering by minimizing information entropy rather than distance between sequences in the same cluster. We have applied our entropy based clustering method to SARS-CoV-2 viral sequencing data. We report the information content extracted from the sequences by entropy based clustering. Our method converges to similar minimum-entropy clusterings across different runs and limited permutations of data. We also show that a parallelized version of our tool is scalable to very large SARS-CoV-2 datasets. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

USED CAR CLUSTERING in BANDUNG CITY USING K-PROTOTYPES ALGORITHMS

Since the case of the 2019 Coronavirus Disease pandemic or commonly referred to as Covid-19, the use of public transportation has slowly begun to become an option as transportation to reduce the spread of the corona virus cluster, therefore some people prefer to buy private vehicles. However, due to the increasing price of cars, some people prefer to buy used cars. On the used car buying and selling platform, OLX Autos Indonesia, the demand for used cars increased by 15% to 20%. Therefore, this study was conducted to determine the characteristics of the cluster formed from the used car sales dataset taken from AtapData (atapdata.ai). AtapData is an open data site in Indonesia that can be used for research related to Data Science. This cluster model was created using the K-Prototypes algorithm, Silhouette Score and Davies Bouldin Index to evaluate the resulting cluster results. This clustering model will produce three clusters. The results of the three clusters will have one thing in common, namely brands that dominate sales, including Toyota, Honda, Daihatsu, Nissan, and Mitsubishi. Clustering evaluation using the Silhouette Score method produces a value of 0.7744140503593034. And for the evaluation of the Davies-Bouldin Index it produces a value of 0.4999221950856398. © 2022 IEEE.

Detection of COVID-19 in X-Ray Images Using Constrained Multi-view Spectral Clustering

Machine learning, and specifically classification algorithms, has been widely used for the diagnosis of COVID-19 cases. However, these methods require knowing the labels of the datasets, and use a single view of the dataset. Due to the widespread of the COVID-19 cases, and the presence of the huge amount of patient datasets without knowing their labels, we emphasize in this paper to study, for the first time, the diagnosis of COVID-19 cases in an unsupervised manner. Thus, we can benefit from the abundance of datasets with missing labels. Nowadays, multi-view clustering attracts many interests. Spectral clustering techniques have attracted more attention thanks to a well-developed and solid theoretical framework. One of the major drawbacks of spectral clustering approaches is that they only provide a nonlinear projection of the data, which requires an additional clustering step. Since this post-processing step depends on numerous factors such as the initialization procedure or outliers, this can affect the quality of the final clustering. This paper provides an improved version of a recent method called Multiview Spectral Clustering via integrating Nonnegative Embedding and Spectral Embedding. In addition to keeping the benefits of this method, our proposed model incorporates two types of constraints: (i) a consistent smoothness of the nonnegative embedding across all views, and (ii) an orthogonality constraint over the nonnegative embedding matrix columns. Its advantages are demonstrated using COVIDx datasets. Besides, we test it with other image datasets to prove the right choice of this method in this study. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

A Social Distance Monitoring Method Based on Improved YOLOv4 for Surveillance Videos

Social distance monitoring is of great significance for public health in the era of COVID-19 pandemic. However, existing monitoring methods cannot effectively detect social distance in terms of efficiency, accuracy, and robustness. In this paper, we proposed a social distance monitoring method based on an improved YOLOv4 algorithm. Specifically, our method constructs and pre-processes a dataset. Afterwards, our method screens the valid samples and improves the K-means clustering algorithm based on the IoU distance. Then, our method detects the target pedestrians using a trained improved YOLOv4 algorithm and gets the pedestrian target detection frame location information. Finally, our method defines the observation depth parameters, generates the 3D feature space, and clusters the offending aggregation groups based on the L2 parametric distance to finally realize the pedestrian social distance monitoring of 2D video. Experiments show that the proposed social distance monitoring method based on improved YOLOv4 can accurately detect pedestrian target locations in video images, where the pre-processing operation and improved K-means algorithm can improve the pedestrian target detection accuracy. Our method can cluster the offending groups without going through calibration mapping transformation to realize the pedestrian social distance monitoring of 2D videos. © 2023 World Scientific Publishing Company.

Prediction and Analysis of Infectious Disease Visit Data Based on DPC

Since the outbreak of the COVID-19 in early 2020, the prevention and control of infectious diseases has been raised to a higher level. However, tuberculosis still ranks in the forefront of the incidence rate of various infectious diseases in China. The tuberculosis epidemic has also brought great economic pressure and negative social impact to the society every year. Therefore, we have always been very concerned about how to effectively prevent and control the spread of tuberculosis. However, the diagnostic data of tuberculosis are often high-dimensional, huge, messy and difficult to be used effectively. How to extract knowledge from the data to help medical staff find the incidence trend of tuberculosis to assist decision-making has become a practical topic. In this paper, after clarifying and standardizing the original data, the density peak clustering (DPC) algorithm is used for deep mining. The knowledge is extracted through clustering analysis and visualization. Finally, analysis results can intuitively illustrate the effectiveness and practical research significance of this work. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

DNN and Model Combined Passive Localization and Social Distancing with Partial Inertial Aiding

At the outbreak of COVID-19, researchers worldwide are seeking approaches to containing this disease. It is necessary to monitor social distance in enclosed public areas, such as subways or shopping malls. Passive localization, such as surveillance cameras, is a natural candidate for this issue, which is meaningful for rapid response to finding the infected suspect. However, the latest surveillance camera system is rotatable, even movable. And it is impossible for professionals to regularly calibrate the extrinsic parameters in a large-scale application, like COVID-19 suspect monitoring. We propose an inertial-aided passive localization method using surveillance camera for social distance measurement without the necessity to obtain extrinsic parameters. Moreover, the hardware modification cost of the off-the-shelf commercial camera is low, which suits the immediate application. The method uses SGBM (Semi-Global Block Matching) for 3D reconstruction and combines YOLOv3 and Gaussian Mixture Model (GMM) clustering algorithm to extract pedestrian point clouds in real time. Combining the 2D DNN-based and model-based methods makes a better balance between the computational load and the detection accuracy than end-to-end 3D DNN-based method. The inertial sensor provides an extra observation for the coordinate transformation from the camera frame into the world ground frame. Results show we can get a decimeter-level social distancing accuracy under noisy background and foreground environments at a low cost, which is promising for urgent COVID-19 public area monitoring. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Clustering of COVID-19 Vaccination Recipients in DKI Jakarta Using the K-Medoids Algorithm

The COVID-19 pandemic has become a serious problem that has attacked various aspects of life such as social, economic, religious, and others. The government has held a COVID-19 vaccination program as an effort to deal with the COVID-19 problem since January 13, 2021. Many problems occurred due to difficulties in dividing the vaccination recipient areas. This is due to the large number of regions with different conditions for each region. One of the efforts to assist the process of processing large vaccination data is data mining techniques and using the clustering method with the K-medoids algorithm. In this study, data on COVID-19 vaccination was grouped in the East Jakarta area using the K-medoids algorithm clustering method. The calculation is carried out using the Euclidean Distance equation and the value of S > 0. The grouped area categories are at the kelurahan level which will then be divided into several clusters. The clustering process was carried out with RapidMiner on 267 kelurahan data on four main attributes, namely the number of targets, the number of vaccine doses 1, the number of vaccine doses 2, and the number of vaccine doses 3. The clustering process was carried out in 6 simulations with variations of k medoids as much as 2 to 7. The results of clustering show the best number of clusters obtained in the simulation is cluster 6 with the smallest Davies Bouldin Index (DBI) value of 0.209. The clusters obtained are clusters 0 to cluster 5. The cluster that is prioritized in giving vaccinations is cluster 2 with 67 items because its members are areas in DKI Jakarta and give a high score in cases of COVID-19 compared to other clusters. © 2022 IEEE.

Evaluating COVID-19 Sequence Data Using Nearest-Neighbors Based Network Model

The SARS-CoV-2 coronavirus is the cause of the COVID-19 disease in humans. Like many coronaviruses, it can adapt to different hosts and evolve into different lineages. It is well-known that the major SARS-CoV-2 lineages are characterized by mutations that happen predominantly in the spike protein. Understanding the spike protein structure and how it can be perturbed is vital for understanding and determining if a lineage is of concern. These are crucial to identifying and controlling current outbreaks and preventing future pandemics. Machine learning (ML) methods are a viable solution to this effort, given the volume of available sequencing data, much of which is unaligned or even unassembled. However, such ML methods require fixed-length numerical feature vectors in Euclidean space to be applicable. Similarly, euclidean space is not considered the best choice when working with the classification and clustering tasks for biological sequences. For this purpose, we design a method that converts the protein (spike) sequences into the sequence similarity network (SSN). We can then use SSN as an input for the classical algorithms from the graph mining domain for the typical tasks such as classification and clustering to understand the data. We show that the proposed alignment-free method is able to outperform the current SOTA method in terms of clustering results. Similarly, we are able to achieve higher classification accuracy using well-known Node2Vec-based embedding compared to other baseline embedding approaches. © 2022 IEEE.

Improving the Prediction of Passenger Numbers in Public Transit Networks by Combining Short-term Forecasts with Real-time Occupancy Data

Passengers of public transportation nowadays expect reliable and accurate travel information. The need for occupancy information is becoming more prevalent in intelligent public transport systems as people started avoiding overcrowded vehicles during the COVID-19 pandemic. Furthermore, public transportation companies require accurate occupancy forecasts to improve service quality. We present a novel approach to improve the prediction of passenger numbers that enhances a day-ahead prediction with real-time data. We first train a baseline predictor on historical automatic passenger counting data. Next, we train a realtime model on the deviations between baseline prediction and observed values, thus capturing events not addressed by the baseline. For the forecast, we attempt to detect emerging patterns in real time and adjust the baseline prediction with deviations from the patterns. Our experiments with data from Germany show that the proposed model improves the forecast of the baseline model and is only outperformed by artificial neural networks in some instances. If the training sets only cover a limited period of up to four months, our approach outperforms competing methods. For larger training sets, there are mixed results in the sense that for some test cases, certain types of neural networks yield slightly better results, but our method still performs well with less training effort, is explainable along the whole prediction process and can be applied to existing prediction methods. Author