Billion In Losses Research Articles

Evaluating the Reparative Potential of Secretome from Patient-Derived Induced Pluripotent Stem Cells during Ischemia–Reperfusion Injury in Human Cardiomyocytes

During a heart attack, ischemia causes losses of billions of cells; this is especially concerning given the minimal regenerative capability of cardiomyocytes (CMs). Heart remuscularization utilizing stem cells has improved cardiac outcomes despite little cell engraftment, thereby shifting focus to cell-free therapies. Consequently, we chose induced pluripotent stem cells (iPSCs) given their pluripotent nature, efficacy in previous studies, and easy obtainability from minimally invasive techniques. Nonetheless, using iPSC secretome-based therapies for treating injured CMs in a clinical setting is ill-understood. We hypothesized that the iPSC secretome, regardless of donor health, would improve cardiovascular outcomes in the CM model of ischemia–reperfusion (IR) injury. Episomal-generated iPSCs from healthy and dilated cardiomyopathy (DCM) donors, passaged 6–10 times, underwent 24 h incubation in serum-free media. Protein content of the secretome was analyzed by mass spectroscopy and used to treat AC16 immortalized CMs during 5 h reperfusion following 24 h of hypoxia. IPSC-derived secretome content, independent of donor health status, had elevated expression of proteins involved in cell survival pathways. In IR conditions, iPSC-derived secretome increased cell survival as measured by metabolic activity (p < 0.05), cell viability (p < 0.001), and maladaptive cellular remodelling (p = 0.052). Healthy donor-derived secretome contained increased expression of proteins related to calcium contractility compared to DCM donors. Congruently, only healthy donor-derived secretomes improved CM intracellular calcium concentrations (p < 0.01). Heretofore, secretome studies mainly investigated differences relating to cell type rather than donor health. Our work suggests that healthy donors provide more efficacious iPSC-derived secretome compared to DCM donors in the context of IR injury in human CMs. These findings illustrate that the regenerative potential of the iPSC secretome varies due to donor-specific differences.

Rainfall analysis of the May 2021 southeastern Texas and southern Louisiana flood

ABSTRACT This paper analyzes the heavy rainfall on 16–20 May 2021 across southeastern Texas and southern Louisiana. In Louisiana, five fatalities were attributed to the event and > $1 billion in losses. The rainfall was caused by two slow-moving, mesoscale convective vortices that concentrated precipitation south of Port Arthur, TX, Lake Charles, LA, and Baton Rouge, LA, in 6 to 12-hour periods. Using calibrated radar data, the Storm Precipitation Analysis System estimated 6-h rainfall totals of 254–362 mm south of Lake Charles, surpassing 200–500-year average recurrence intervals (ARI) for those locations, with an isolated area receiving 367.5 mm, equating to a > 800-year ARI. In East Baton Rouge Parish, LA, 203–268 mm was estimated in 6-h (200–500-year ARI), with an isolated area receiving 274 mm, equivalent to a 750-year ARI. A comparison of the May 2021 and August 2016 events, which both affected southern Louisiana, is made using a depth-area-duration analysis. While the forcing mechanisms differed, 1–6 h rainfall depths were similar at 1 –25,900 km2. The May 2021 event exceeded rainfall depths produced by the August 2016 event for 2–6 h durations for area sizes of 1–388 km2, highlighting the precipitation intensity produced by the May 2021 event.

Characterization of the Dominant Stages at Which Gas Flaring Is Introduced: Impacts and Policy Options to Ameliorate Them

Information on associated gas flaring in the Niger Delta is analyzed to characterize the various stages in the lifecycle of an oil exploration and production facility through which gas flaring is introduced into an area, leading to ecological, health, and economic impacts, and possible policies and implementation strategies for the reduction of the flared gas are offered. Gas flaring is currently occurring at a rate of 5318 million m3 per year in the Niger Delta region. The impacts of gas flaring on the ecological, human, and economic systems include poor air quality, acid rain, and soil degradation. Other impacts include decreased crop yield, increased incidences of respiratory diseases amongst inhabitants in the area, and losses of billions of USD yearly in economic costs. Three dominant stages in the introduction of gas flaring are characterized in this study, using the concept of the management of the invasion of non-native species into an ecosystem and it is determined that the cheapest method of minimizing gas flaring is to incorporate mitigation measures during the design stage of the oil exploration/production facility, rather than after flaring has commenced, as is the common emphasis. During the early periods of oil production, gas flaring is largely localized, and the gas utilization measures can still be implemented to good effect. However, once the facility is established, gas flaring becomes pervasive, stretching kilometers beyond the flare points. Current policies like the imposition of flare-out years on the oil producers have helped, but they have been insufficient as the intensity of gas flaring in the region continues at double the average global levels. Working with relevant stakeholders, a realistic flare-out deadline should be agreed upon, clear and specific gas flaring legislation should be drawn up that includes measures to mitigate gas flaring for each proposed oil facility, and proper data collection and management services would enable the implementation of the policies to be based on robust evidence. These policies would shift the costs of gas flaring away from the inhabitants of the region back to the operators of the facilities that are the sources of the flares.

Load Shedding and Its Crippling Effect on The South African Economy: An Entrepreneurial Perspective

The country has experienced industry shutdowns, decreased productivity, educational setbacks, increased unemployment rates, and negative healthcare outcomes. Load-shedding in South Africa is primarily caused by inadequate maintenance, internal mismanagement, corruption, and a lack of skills. This issue has had a significant impact on small businesses, as they often lack the necessary resources to handle power surges, equipment damage, and the costs associated with replacing stock. This paper explores how load-shedding has impacted the South African economy from an entrepreneurial perspective. The research is qualitative and exploratory in nature, and it was found that load-shedding has had a significant impact on businesses' finances, resources, and performance. Load-shedding in South Africa has had a severe impact on the daily lives of people, causing disruptions in businesses, households, and the overall economy. Eskom, the primary supplier of energy in South Africa, has been facing significant challenges in meeting the growing demand for electrical energy, leading to job losses, decreased productivity, and a loss of billions of rands. Small businesses have been particularly affected, with an estimated 75% of them failing after being operational for less than three years due to load-shedding. To ensure small businesses have electricity to operate during load-shedding it is recommended that small businesses adopt alternative energy sources like solar power to increase the businesses’ independence and mitigate the effects of load-shedding. The South African government should also encourage and support small businesses to collaborate with successful renewable energy companies.

Detection dogs fighting transnational narcotraffic: performance and challenges under real customs scenario in Brazil.

Narcotic Detection Dogs (NDDs) are essential tools in the fight against drug trafficking, acting with high precision and improving efficiency at border posts. When trained efficiently, these dogs can detect a great variety of compounds, such as cocaine, marijuana and its derivatives, and synthetic drugs, among others. Most of the knowledge on canine detection processes and efficiency has been determined in experimentally controlled conditions, but narcotic seizures detected by dogs in realistic anti-drug operations have not yet been critically determined in a Country with continental dimensions such as Brazil. This study aimed to evaluate the data set concerning the performance, operations, efficiency, and success rate of NDDs used by the Brazilian Customs Authority (Aduana) in the fight against drug trafficking. Narcotic seizure rates increased in luggage and packages detected by NDDs working at border crossings from 2010 to 2020, with an estimated value of over US$ 2 billion in losses to the cocaine drug trafficking business. NDD units also increased most narcotic groups seized in the same period. The number of NDDs and anti-drug operations, and Customs Border Post (CBP) influenced the rates of drugs seized. NDDs provided an increase of 3,157 kg/animal of drugs seized for every new dog introduced into the inspection systems.

Epigenetic Modifications of White Blood Cell DNA Caused by Transient Fetal Infection with Bovine Viral Diarrhea Virus.

Bovine viral diarrhea virus (BVDV) infections cause USD 1.5-2 billion in losses annually. Maternal BVDV after 150 days of gestation causes transient fetal infection (TI) in which the fetal immune response clears the virus. The impact of fetal TI BVDV infections on postnatal growth and white blood cell (WBC) methylome as an index of epigenetic modifications was examined by inoculating pregnant heifers with noncytopathic type 2 BVDV or media (sham-inoculated controls) on Day 175 of gestation to generate TI (n = 11) and control heifer calves (n = 12). Fetal infection in TI calves was confirmed by virus-neutralizing antibody titers at birth and control calves were seronegative. Both control and TI calves were negative for BVDV RNA in WBCs by RT-PCR. The mean weight of the TI calves was less than that of the controls (p < 0.05). DNA methyl seq analysis of WBC DNA demonstrated 2349 differentially methylated cytosines (p ≤ 0.05) including 1277 hypomethylated cytosines, 1072 hypermethylated cytosines, 84 differentially methylated regions based on CpGs in promoters, and 89 DMRs in islands of TI WBC DNA compared to controls. Fetal BVDV infection during late gestation resulted in epigenomic modifications predicted to affect fetal development and immune pathways, suggesting potential consequences for postnatal growth and health of TI cattle.

Detailed seismic structure beneath the earthquake zone of Yogyakarta 2006 (Mw ∼6.4), Indonesia, from local earthquake tomography

The earthquake, which occurred in Yogyakarta, Indonesia, on May 26, 2006, at 22:53:58 UTC with Mw ∼6.4, was one of the most destructive earthquakes in Indonesia. The earthquake caused thousands of fatalities, tens of thousands of injuries, and hundreds of thousands of house damages in the Yogyakarta area and its surroundings at a loss of billions of dollars. Previous studies from seismic tomography and satellite radar imaging hypothesized that the earthquake was caused by activating a so far unknown fault east of the Opak Fault. Although, in the beginning, the Opak fault was suspected to be the source of the Yogyakarta earthquake in 2006. This assumption was made because the damage was maximum in the Bantul area west of the Opak Fault. This study demonstrates that our seismic tomography achieved a higher resolution than the previous study and could resolve a failed complex fault system. We utilized more aftershocks (2170 events) and smaller grid sizes for seismic tomography inversion. Four focal mechanisms from aftershocks for Mw ≥ 4.5 were also conducted to support structure interpretation in the study area. Our results successfully delineate the Opak Fault and the second fault, namely the Ngalang Fault, parallel to the eastern part of the fault at a depth of 9 km. Two faults could be indicated by the velocity contrast of Vp, Vp/Vs ratio, and Vs from a horizontal section tomogram. Our focal mechanisms also support seismic tomography, revealing two fault planes in our study area. The results show that the two faults are connected by the Oyo Fault, which is ruptured in the opposite direction compared to the two faults.

Predicting automobile insurance fraud using classical and machine learning models

Insurance fraud claims have become a major problem in the insurance industry. Several investigations have been carried out to eliminate negative impacts on the insurance industry as this immoral act has caused the loss of billions of dollars. In this paper, a comparative study was carried out to assess the performance of various classification models, namely logistic regression, neural network (NN), support vector machine (SVM), tree augmented naïve Bayes (NB), decision tree (DT), random forest (RF) and AdaBoost with different model settings for predicting automobile insurance fraud claims. Results reveal that the tree augmented NB outperformed other models based on several performance metrics with accuracy (79.35%), sensitivity (44.70%), misclassification rate (20.65%), area under curve (0.81) and Gini (0.62). In addition, the result shows that the AdaBoost algorithm can improve the classification performance of the decision tree. These findings are useful for insurance professionals to identify potential insurance fraud claim cases.

Food Waste from Campus Dining Hall as a Potential Feedstock for 2,3-Butanediol Production via Non-Sterilized Fermentation.

Food waste is a major issue that is increasingly affecting our environment. More than one-third of food is wasted, resulting in over $400 billion in losses to the U.S. economy. While composting and other small recycling practices are encouraged from person-to-person, it is not enough to balance the net loss of 80 million tons per year. Currently, one of the most promising routes for reducing food waste is through microbial fermentation, which can convert the waste into valuable bioproducts. Among the compounds produced from fermentation, 2,3-butanediol (2,3-BDO) has gained interest recently due to its molecular structure as a building block for many other derivatives used in perfumes, synthetic rubber, fumigants, antifreeze agents, fuel additives, and pharmaceuticals. Waste feedstocks, such as food waste, are a potential source of renewable energy due to their lack of cost and availability. Food waste also possesses microbial requirements for growth such as carbohydrates, proteins, fats, and more. However, food waste is highly inconsistent and the variability in composition may hinder its ability to be a stable source for bioproducts such as 2,3-BDO. This current study focuses specifically on post-consumer food waste and how 2,3-BDO can be produced through a non-model organism, Bacillus licheniformis YNP5-TSU during non-sterile fermentation. From the dining hall at Tennessee State University, 13 food waste samples were collected over a 6-month period and the compositional analysis was performed. On average, these samples consisted of fat (19.7%), protein (18.7%), ash (4.8%), fiber (3.4%), starch (27.1%), and soluble sugars (20.9%) on a dry basis with an average moisture content of 34.7%. Food waste samples were also assessed for their potential production of 2,3-BDO during non-sterile thermophilic fermentation, resulting in a max titer of 12.12 g/L and a 33% g/g yield of 2,3-BDO/carbohydrates. These findings are promising and can lead to the better understanding of food waste as a defined feedstock for 2,3-BDO and other fermentation end-products.

Contribution of Signaling Partner Association to Strigolactone Receptor Selectivity.

The parasitic plant witchweed, Striga hermonthica, results in agricultural losses of billions of dollars per year. It perceives its host via plant hormones called strigolactones, which act as germination stimulants for witchweed. Strigolactone signaling involves substrate binding to the strigolactone receptor, followed by substrate hydrolysis and a conformational change from an inactive, or open state, to an active, or closed state. In the active state, the receptor associates with a signaling partner, MAX2. Recently, it was shown that this MAX2 association process acts as a strong contributor to the uniquely high signaling activity observed in ShHTL7; however, it is unknown why ShHTL7 has enhanced MAX2 association affinity. Using an umbrella sampling molecular dynamics approach, we characterized the association processes of AtD14, ShHTL7, a mutant of ShHTL7, and ShHTL6 with MAX2 homologue OsD3. From these results, we show that ShHTL7 has an enhanced standard binding free energy of OsD3 compared to those of the other receptors. Additionally, our results suggest that the overall topology of the T2/T3 helix region is likely an important modulator of MAX2 binding. Thus, differences in MAX2 association, modulated by differences in the T2/T3 helix region, are a contributor to differences in signaling activity between different strigolactone receptors.

Collusion Indicators in the GameStop Short Squeeze

&#x0D; &#x0D; &#x0D; A new theory and model of the stock market that incorporates decision noise in the spirit of behavioral economics, collaboration in the spirit of experimental economics, and financial data and features in the spirit of econophysics is used to support a theory from 2005 that stock-market collusion should lower the level of stock-market decision noise - i.e., a quantitative way of seeing how traders influence each other in a stock market. This idea is used to find two rare indicators of the GameStop short squeeze – one of which occurred six days before the rapid price increase, which could have potentially reduced the over $19 billion in losses of short sellers by a significant amount. The relationship predicted by the model between temperature and aggregate volume has been validated by data from the GameStop short squeeze.&#x0D; &#x0D; &#x0D;

An Analysis of Malicious URL Detection Using Deep Learning

Considerable progress has been achieved in the digital domain, particularly in the online realm where a multitude of activities are being conducted. Cyberattacks, particularly malicious URLs, have emerged as a serious security risk, deceiving users into compromising their systems and resulting in annual losses of billions of dollars. Website security is essential. It is critical to quickly identify dangerous or bad URLs. Blacklists and shallow learning are two techniques that are being investigated in response to the threat posed by malicious URLs and phishing efforts. Historically, blacklists have been used to accomplish this. Techniques based on blacklists have limitations because they can't detect malicious URLs that have newly generated. In order to overcome these challenges, recent research has focused on applying machine learning and deep learning techniques. By automatically discovering complex patterns and representations from unstructured data, deep learning has become a potent tool for recognizing and reducing these risks. The goal of this paper is to present a thorough analysis and structural comprehension of Deep Learning based malware detection systems. The literature review that covers different facets of this subject, like feature representation and algorithm design, is found and examined. Moreover, a precise explanation of the role of deep learning in detecting dangerous URLs is provided.

Malicious URL Detector using Machine Learning

In recent years, the digital world has advanced significantly, particularly on the Internet, which is critical given that many of our activities are now conducted online. the risk of a cyberattack is rising rapidly. One of the most critical attacks is the malicious URL intended to extract unsolicited information by mainly tricking inexperienced end users, resulting in compromising the user’s system and causing losses of billions of dollars each year. In this paper, we provide an extensive literature review highlighting the main techniques used to detect malicious URLs that are based on machine learning models, taking into consideration the limitations in the literature, detection technologies, feature types, and the datasets used. In this research paper we have implemented the application that it can check the URL given by the user is benign, defacement, malware, phishing. After checking then it will display the type of URL in user interface. The increasing sophistication of cyber threats, particularly in the form of malicious URLs, poses a significant challenge to internet security. Traditional signature-based approaches and heuristic methods have proven insufficient in detecting rapidly evolving malicious URLs.This research paper presents a comprehensive study on the development and evaluation of a malicious URL detector based on machine learning known and novel malicious URLs. Keyword: Phishing, URL, machine learning, cyber security, random forest, malicious.

Forecasting onion armyworm using tree-based machine learning models

In the Philippines, the province of Nueva Ecija produces fifty-four percent of its annual onion production. However, the level of onion growth production was reduced; since the outbreak of 2016, armyworms destroyed thousands of hectares of farms resulting in a loss of billions of pesos, which lead to the decline of the onion harvest. In this study, we develop machine learning models to forecast an outbreak of armyworms to help evade or reduce the damage caused by an armyworm outbreak. Climatic data; particularly Maximum temperature, Minimum Temperature, Ultraviolet Index, Humidity, Cloudiness, Wind Speed, Sun Hours, Rainfall, and Pressure from the Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA) and armyworm outbreak occurrences data from the Provincial Agriculture Office (PAO) of Nueva Ecija was used as the dataset for this study Using Tree-based machine learning models Decision Tree and Random Forest. Binary classifiers were developed and evaluated to forecast the occurrence or non-occurrence of the armyworm outbreak and the use of feature importance to distinguish the most critical climatic features that significantly contribute to forecasting an armyworm outbreak in the province of Nueva Ecija. These tree-based models produced satisfactory results, with the Random Forest model exhibiting a better forecasting capability than the Decision Tree model.

Machine Learning for Personalized Risk Assessment and Early Detection of Heart Disease

Abstract: Today in our busy life it very difficult for us to look after our health. Cardio vascular diseases are very common now days which cause loss of billions of lives world-wide. Our lifestyles have a major impact on our health causing various chronic diseases. Machine learning (ML) can revolutionize the field of cardiovascular disease prediction by providing more accurate, nuanced, and personalized risk assessment, leading to improved patient health. A properly trained machine learning model can easily detect or predict the heart disease. Within the context of using artificial intelligence for forecasting and assessing the occurrence of heart ailments, here are various factors that contribute to the heart disease to be considered, such as lifestyle choices and medical conditions to predict whether the person is at high risk or not. ML technology can also analyze large datasets of patient data to identify patterns and risk factors that cannot be easily distinguished, predicted, or detected using traditional methods. Estimators like k-NN classifier, Decision tree classifier, Gradient boosting classifier and Gaussian Naive bayes (NB) classifier, are provided with different characteristics or feature extracted from the dataset, demonstrated consistent and reliable performance in predicting heart disease. The utilization of these attributes can potentially aid the medical field for on time detection and diagnosis of the heart disease.

Prediction of Financial Crime Using Machine Learning

The purpose of data analytics is to uncover previously unknown patterns and make use of such patterns to help in making educated decisions across a wide range of contexts. Because of advances in modern technology and the fact that credit cards have become an easy target for fraudulent activity, the incidence of credit card fraud has considerably increased in recent years. Credit card fraud is a significant issue in the industry of financial services, and it results in annual losses of billions of dollars. The development of a fraud detection algorithm is a difficult endeavor due to the paucity of real-world transaction datasets that are available due to confidentiality concerns and the very unbalanced nature of the datasets that are publicly available. Use a dataset from the real world in conjunction with a variety of supervised machine learning algorithms to identify potentially fraudulent credit card transactions. In addition, make use of these techniques to create a super classifier through the use of ensemble learning methods. Determine which variables are the most significant and could perhaps lead to a higher level of accuracy in the identification of fraudulent credit card transactions. In addition, we evaluate and discuss the performance of a number of other supervised machine learning algorithms that are currently available in the literature in contrast to the super classifier that can be implemented.

Vehicle target detection algorithm based on yolov5

With the rapid advancement of automobile manufacturing technology and the intensification of competition among automobile brands, the development of autonomous driving has been pushed to the forefront of automobile development, which causes approximately $650 billion in losses due to traffic accidents worldwide every year. For in the complex vehicle target scene, because of its many vehicle targets, dense targets often exist between the occlusion, overlap, a variety of different weather reasons, rain, fog, sunny days, etc. resulting in obstruction of the field of view, and the vehicle camera often intercepted images mostly exist blurred, ghosting and other problems, making the vehicle target detection for detailed feature extraction requirements are high, the detection accuracy is often difficult to meet the requirements, proposed based on YOLOv5s vehicle target detection algorithm is proposed. ACmix attention mechanism is introduced to make the model achieve the purpose of expanding the target perception field, adaptively focusing on different target regions, and capturing more information features, etc. The test shows that the accuracy of the model increases by 1.1% on a subset of BDD100K data set after improvement. The accuracy of the improved model increased by 1.1% on the subset of BDD100K dataset and by 1.2% on the PASCAL VOC2007 general dataset.

Post-Flood Resilience Assessment of July 2021 Flood in Western Germany and Henan, China

In July 2021, devastating floods occurred in western Germany and Henan, China, resulting in extreme loss of life and property damage. Despite the differences in context, climate change contributed to these events. Flood resilience generally means the system’s ability to recover from floods. A post-flood resilience assessment seeks to determine the impact of the flood on the area, the duration it took to recover, the effectiveness of the measures taken to reduce the risk of flooding, and ways to enhance flood resilience. The post-flood review capacity method was used to assess the event and calculate the flood resilience index. Western Germany experienced a 500-year return period flood in connection with the low-pressure system, Bernd, while Zhengzhou in Henan experienced a 1000-year return period flood with the influence of Typhoon In-Fa and the Western Pacific subtropical high. More than 107,000 people were affected in Germany, with 205 deaths that account for USD 40 billion in economic losses, whereas in Henan, 14.786 million people were affected, and 398 people died, which accounts for USD 18.9 billion in losses. Germany was more impacted and took longer to restore essential services than Henan, China. The flood resilience index shows that the resilience level of both countries is low. The severe rainstorms in Zhengzhou and the Ahr River Valley exposed weaknesses in urban disaster management, particularly in urban areas, such as subway flooding and risk communication with the public. The events highlighted the need to better understand risks and their consequences, early warning systems, preparedness, and emergency response.

Application of Probabilistic Seismic Hazard Assessment to Understand the Earthquake Hazard in Attock City, Pakistan: A Step towards Linking Hazards and Sustainability

Within the last three decades, twelve major earthquakes (Mw &gt; 6.0) have jolted Pakistan and contributed to a heavy death toll and an economic loss of billions of dollars, which is immense for any underdeveloped country. Despite the generalized description of seismic hazards in various regions of Pakistan, densely populated cities still require a detailed and integrated vulnerability analysis to overcome the impact of a significant earthquake. This study aims to integrate seismic hazard assessment schemes to understand the vulnerability of Attock city against an earthquake. It initially evaluates the threat from an earthquake due to tectonic activity in the region, splits the region (about 200 km radius) into six seismic zones and uses area source parameters. The ground motion prediction equations compatible with the study area’s seismotectonic environment are also used in this study. Peak horizontal ground acceleration (PGA) and 5% damped spectral acceleration are critical features of ground motions. The site classification is carried within Attock city, indicating the presence of SB (foundation condition with Vs30 = 760 m/s), SC (foundation condition with Vs30 = 400 m/s) and SD (foundation condition with Vs30 = 300 m/s). The peak ground accelerations for a return period of 475 years at the SB, SC and SD sites are estimated as 0.23 g, 0.28 g and 0.30 g, respectively. Uniform hazard spectra are obtained for each site classification at three return periods (475, 975 and 2475 years). Another possible threat can be the local site conditions of the study area, as Attock city exists on the unlithified sediments of upper Pleistocene to Recent alluvial deposits. That is why microtremor recordings are conducted at 20 sites within Attock city to understand the fundamental frequency (f0), horizontal to vertical spectral amplitude (A0) and Kg parameter, a seismic vulnerability index. The values of f0 are found between 0.6 and 9 Hz and A0 is observed between 2.1 and 5 Hz, whereas the Kg is estimated between 0.24 and 20 Hz. Despite evidence of the seismic vulnerability of Attock city, the current building designs and infrastructure development are not synchronized with the uniform hazard response spectra and the soil amplification, thus enhancing the exposure of the study area to disaster during a major earthquake. This study will be instrumental in pre-disaster mitigation strategies for urban planners and policymakers.

Robust Framework for Malevolent URL Detection using Hybrid Supervised Learning

The In this Internet Era, most of the people are using world-wide-web (www) based websites extensively for accomplishing their daily activities including E-commerce. As per latest report, total of 1.5 billion URLs are accessed every day and it is increasing every second. From another perspective, malicious URLs are employed in URL phishing- which in turn steal customer credentials and thereby lead to loss of billions of dollars. Due to these aspects, malevolent URL detection is gaining huge attention in the literature in the past few years. However, the state-of-the art literature employ only popular ML techniques and small datasets with less diversity for detecting malicious URLs. Further, the existing malicious URL detection methods are less focusing on bottleneck issues such as Overfitting and hyper-parameter tuning of ML models, which play a significant role in deciding the prediction accuracy of the proposed detection framework. In order to tackle these issues, this research article proposes a robust Malevolent URL Detection framework, which utilizes a hybrid machine learning algorithm, Support Vector Regression (SVR) and hyper parameters tuning strategies to enhance the prediction accuracy. The experimental results conducted on training and testing datasets clearly demonstrate the performance of the proposed detection framework in terms of PR and accuracy metrics respectively.