Search

A pandemic disease, COVID-19, has caused trouble worldwide by infecting millions of people. The studies that apply artificial intelligence (AI) and machine learning (ML) methods for various purposes against the COVID-19 outbreak have increased because of their significant advantages. Although AI/ML applications provide satisfactory solutions to COVID-19 disease, these solutions can have a wide diversity. This increase in the number of AI/ML studies and diversity in solutions can confuse deciding which AI/ML technique is suitable for which COVID-19 purposes. Because there is no comprehensive review study, this study systematically analyzes and summarizes related studies. A research methodology has been proposed to conduct the systematic literature review for framing the research questions, searching criteria and relevant data extraction. Finally, 264 studies were taken into account after following inclusion and exclusion criteria. This research can be regarded as a key element for epidemic and transmission prediction, diagnosis and detection, and drug/vaccine development. Six research questions are explored with 50 AI/ML approaches in COVID-19, 8 AI/ML methods for patient outcome prediction, 14 AI/ML techniques in disease predictions, along with five AI/ML methods for risk assessment of COVID-19. It also covers AI/ML method in drug development, vaccines for COVID-19, models in COVID-19, datasets and their usage and dataset applications with AI/ML.

The way patients are treated in Hospital Emergency Departments changes during the year, depending on many factors. One key component is weather temperature. Some seasonal maladies are tightly related to temperature, such as flu in cold weather or sunburn in hot weather. In this study, data from a hospital in Valencia was used to explore how harsh weather changes affect the emergency department, obtaining information about probable impacts of global warming effects in healthcare systems. Process mining techniques helped in the discovery of changes in the Emergency Departments. Some illnesses, such as heat stroke, are more prevalent during heatwaves, but more interestingly, the time to attend patients is also higher. Rapid changes in temperature are also analyzed through Process Mining techniques.

Consumption rates of polystyrene microplastics (beads of 6, 12, and 26 μm diameter) and their effects on energy metabolism and motor activity of the copepodCalanus helgolandicusliving in the Marmara Sea were investigated. All sizes of microplastic particles were actively consumed and excreted via fecal pellets, however, copepods displayed a significant preference for beads sized 6 μm. In a mixture of algae and microplastics beads of 6 μm, microplastics consumption rates linearly (r2= 0.78,n= 154) increased 800 times from 50.8 ± 17.3 to 8,612 ± 5,972 beads ind–1day–1with an increase in bead concentration from 10 to 44,000 beads ml–1. The total and basal metabolic rates as well as time spent swimming forC. helgolandicus, decreased 1.7, 1.8 and about 3-fold, respectively after 7–8 days exposure to microplastic treatments, which was similar to the metabolism and activity of starving animals in filtered water. In copepods consuming microplastics, all vital parameters decreased on the first day of exposure, indicating either accelerated starvation, probably due to increased losses of energy and biological matter in the formation of fecal pellets and/or traumatic/toxic effects of the polystyrene beads on the copepods. Our data from laboratory experiments indicate that the presence of large concentrations of microplastics in water, even when mixed with algae, lowered energy metabolism levels ofC. helgolandicus.

Providing care to someone with a chronic disease requires being physically, emotionally, and mentally energetic. In particular, the presence of a patient with a disease that requires palliative care changes daily activities and routines, increases responsibilities of those who take the responsibility for patient’s care and changes the roles in the family. The objective of this study was to determine the care burden and social support levels among the caregivers providing care for patients hospitalized in palliative care clinics. Materials and Methods: The population of this descriptive and correlational study consisted of caregivers taking care of patients treated at the Palliative Care Clinic of Training and Research Hospital between May and October 2018; the sample consisted of 73 caregivers who volunteered to join the study and were open to communication. Data were collected by questioning patients and their caregivers about their sociodemographic characteristics, using the Caregiver Burden Scale and the Multidimensional Scale of Perceived Social Support. Data analysis included percentages, the Kruskal-Wallis test, the Mann-Whitney U test, and the Spearman’s Correlation Coefficient. Results: The average caregivers’ age was 45.44±13.76 years; 75.3% of caregivers were females, 30.1% of caregivers were literate or had primary school degrees. Caregiver’s gender and educational levels were found not to affect caregiving and social support levels; however, the economic status affected caregiving and social support levels. There was a reverse correlation between the caregiver burden and their social support levels. Conclusions: With increased caregiver burden, their social support level decreased. Nurses caring for patients in palliative care clinics will benefit from educating and supporting caregivers about clinic and home care; it will result in a positive level of social support for both caregivers and patients receiving care.

Today, some newspapers have started to service online only. One of the reasons of this situation is competitiveness and increasing costs in press market. So newspaper sale planning requires adaptive scheduling and understanding customer behavior well. Estimating the right number of delivery is crucial. It is succeeded with expert knowledge supported with proper analysis and techniques. This study focuses on understanding newspaper sales amount by revealing critical variables and their importance on sales. This is a critical shortcoming especially for newspaper companies, which always try to reduce the number of sell outs and increase the number of return in the delivered newspapers offer companies financial benefits. In this study, it was analyzed that whether different weather conditions (sunny, cloudy or rainy) and days (weekday or weekend) are effective in daily newspaper sales for Ankara, Izmir and Istanbul. One-way Anova, Two-way Anova, t-test, Levene’s, Kolmogorov-Smirnov, and The Kruskal–Wallis tests were used. Sales data was collected from 71 vendors of a national press firm in three cities between 12 September–31 December 2018.

Defining customer requirements in a huge amount of data of the digital era is crucial for companies in a competitive business environment. Customer segmentation has been attracted to a great deal of attention and has widely been performed in marketing studies. However, boundary data which are close to more than one segment may be assigned incorrect classes, which affects to make the right decisions and evaluations. Therefore, segmentation analysis is still needed to develop efficient models using advanced techniques such as soft computing methods. In this study, an intuitionistic fuzzy clustering algorithm were applied to customer data in a supermarket according to the amount spent in some product groups. The data represent 33-month customer shopping data in a supermarket for eight product groups. The results indicate the intuitionistic fuzzy c-means based customer segmentation approach produces more reliable and applicable marketing campaigns than conditional fuzzy c-means and k-means segmentation method.

Although renewable energy sources have more significance than traditional sources, limited energy sources such as natural gas have been consumed in daily life. Natural gas energy is an environmental-friendly and easy-to-use energy source. Furthermore, it is the cleanest burning of all fossil fuels. Since the amount of gas consumed by companies is much more, the companies’ consumption amount was investigated. The companies concerning the consumption amount were clustered by fuzzy c-means algorithm in order to enhance the understandability thanks to the collection of 1049 companies’ data. This study, at the same time, presents the effects of weather conditions on the consumption amount in Istanbul. Hence, a calendar view visualization is created to show the relationship between consumption and weather conditions.

[EN] There are some studies and methods in the literature to understand customer needs and behaviors from the path. However, path analysis has a complex structure because the many customers can follow many different paths. Therefore, clustering methods facilitate the analysis of the customer location data to evaluate customer behaviors. Therefore, we aim to understand customer behavior by clustering their paths. We use an intuitionistic fuzzy c-means clustering (IFCM) algorithm for two-dimensional indoor customer data; case durations and the number of visited locations. Customer location data was collected by Bluetooth-based technology devices from one of the major shopping malls in Istanbul. Firstly, we create customer paths from customer location data by using process mining that is a technique that can be used to increase the understandability of the IFCM results. Moreover, we show with this study that fuzzy methods and process mining technique can be used together to analyze customer paths and gives more understandable results. We also present behavioral changes of some customers who have a different visit by inspecting their clustered paths.

Process mining (PM) supports organizations by improving their processes using event log data collected from information technology systems. Its primary purposes are discovering actual process models, monitoring and comparing actual and desired workflows, and enhancing processes by considering the discovered model and desired flow. Because process mining gains attraction day by day, various technology companies developed process mining tools to support organizations managing their business processes with data science. Technology selection is a complicated multi-criteria decision-making (MDCM) problem under several criteria and experts’ evaluation, including uncertainty and subjectivity. Spherical fuzzy set is a powerful concept to cope with uncertainty by presenting a wider decision-making area and identifying hesitancy. A fuzzy MDCM approach based on spherical fuzzy AHP is offered in this study to manage the problem of selecting process mining technology under uncertain and ambiguous conditions. Then, one-at-a-time sensitivity analysis is applied to reduce the decision-makers’ subjectivity. This study results in that Price, Process Discovery, Process Analysis&Analytics are the most relevant criteria to decide PM technology. It is interesting that even although Operational Support is one of the less important criteria, it may change the decision on selecting the best PM technology.

Indoor customers may have different purposes to visit a shopping mall. Understanding the visiting aims results in better customer relationship management. One of the ways to explain the customer purpose is to discover customer paths. Customers mainly visit stores related to their purposes. The main problem is to discover customer paths from paths. Since customers have changeable mood and there are many stores in a shopping mall, customer paths are generally too complex to evaluate. To overcome that problem, we use process mining technique. Process mining is a technique that has some algorithms to discover business processes from event logs in the databases. In this study, we consider the visited stores as an activity in a process. PALIA Suite, a process mining tool that includes several clustering methods for processes, is used to discover and cluster indoor customer paths to evaluate differences among the visits.

Customer oriented systems provides advantages to companies in competitive environment. Understanding customers is a fundamental problem to present individualized offers. Gender information, which is one of the demographic information of customers, mainly cannot be obtained by data collection technologies. Therefore, various techniques are developed to predict unknown genders of customers. In this study, customer genders are predicted from their paths in a shopping mall using fuzzy set theory. A fuzzy classification method based on Levenshtein distance is developed for string data that refer to the indoor customer paths. Although there are several ways to predict the gender, no study has focused on path-based gender classification. The originality of the study is to classify customer data into the gender classes using indoor paths.

The amount of health expenditure at the household level is one of the most basic indicators of development in countries. In many countries, health expenditure increases relative to national income. If out-of-pocket health spending is higher than the income or too high, this indicates an economical alarm that causes a lower life standard, called catastrophic health expenditure. Catastrophic expenditure may be affected by many factors such as household type, property status, smoking and drinking alcohol habits, being active in sports, and having private health insurance. The study aims to investigate households with respect to catastrophic health expenditure by the clustering method. Clustering enables one to see the main similarity and difference between the groups. The results show that there are significant and interesting differences between the five groups. C4 households earn more but spend less money on health problems by the rate of 3.10% because people who do physical exercises regularly have fewer health problems. A household with a family with one adult, landlord and three people in total (mother or father and two children) in the cluster C5 earns much money and spends large amounts for health expenses than other clusters. C1 households with elementary families with three children, and who do not pay rent although they are not landlords have the highest catastrophic health expenditure. Households in C3 have a rate of 3.83% health expenditure rate on average, which is higher than other clusters. Households in the cluster C2 make the most catastrophic health expenditure.

[EN] Understanding human behavior can assist in the adoption of satisfactory health interventions and improved care. One of the main problems relies on the definition of human behaviors, as human activities depend on multiple variables and are of dynamic nature. Although smart homes have advanced in the latest years and contributed to unobtrusive human behavior tracking, artificial intelligence has not coped yet with the problem of variability and dynamism of these behaviors. Process mining is an emerging discipline capable of adapting to the nature of high-variate data and extract knowledge to define behavior patterns. In this study, we analyze data from 25 in-house residents acquired with indoor location sensors by means of process mining clustering techniques, which allows obtaining workflows of the human behavior inside the house. Data are clustered by adjusting two variables: the similarity index and the Euclidean distance between workflows. Thereafter, two main models are created: (1) a workflow view to analyze the characteristics of the discovered clusters and the information they reveal about human behavior and (2) a calendar view, in which common behaviors are rendered in the way of a calendar allowing to detect relevant patterns depending on the day of the week and the season of the year. Three representative patients who performed three different behaviors: stable, unstable, and complex behaviors according to the proposed approach are investigated. This approach provides human behavior details in the manner of a workflow model, discovering user paths, frequent transitions between rooms, and the time the user was in each room, in addition to showing the results into the calendar view increases readability and visual attraction of human behaviors, allowing to us detect patterns happening on special days., This research was funded by ITACA SABIEN and partially supported by CONICYT REDI 170136.

In recent years, enormous amounts of digital data have been generated. In parallel, data collection, storage, and analysis technologies have developed. Recently, there has been an increasing trend of people moving towards urban areas. By 2030 more than 60% of the world's population will live in an urban environment. Urban areas are big data resource because they include millions of citizens, technological devices, and vehicles which generate data continuously. Besides, rapid urbanization brings many challenges, such as environmental pollution, traffic congestion, health problems, energy management, etc. Some policies for countries are required to cope with urbanization problems. One of these policies is to build smart cities. Smart cities integrate information and communication technology and various physical devices connected to the network (the internet of things or IoT) to both improve the quality of government services and citizen welfare. This chapter presents a literature review of big data, smart cities, IoT, green-IoT concepts, using technology and methods, and applications worldwide.

In recent years, the use of various digital devices that continuously generate massive amounts of heterogeneous, structured or unstructured data has increased. In parallel to generation, data collection, storage, and analysis technologies have developed. Big data sources have a variety of data quality. Preparing and clearing data is one of the first step of mining big data. It is often important to address the full data set found in different data sources to achieve the right result. Various techniques have been used to increase the accuracy of the data comparison. Deterministic and probabilistic linkage algorithms are the two main techniques used in literature. They have different steps to reach qualified and integrated results. To easily interpret the results of the linkage algorithm, a confusion matrix can be used. Measurements such as sensitivity, specificity, positive predictive value, negative predictive value, false positive rate, and false negative rate, are considered to evaluate output quality.

Energy Demand Side Management (EDSM) is defined as the modification of consumer demand for energy. EDSM can be used for different purposes such as reducing peak demand, filling low demand time, strategic demand savings, strategic growth of demand, load-shifting, and flexible load shaping. EDSM can be managed by either Direct Load Control, in which local and central controllers directly affect the consumption, or Indirect Load Control which uses incentives, tariffs, and additional payments that affect consumer behavior. In recent years, with the advances in information and communication technologies, remote monitoring and control of electric devices have become possible. This technology has also enabled gamification approach, which aims to shape human behavior through the intrinsically motivating qualities used in well-designed digital games, to be used for EDSM domain. However, the current literature shows that gamification of EDSM can involve various gamification elements such as avatars, challenges, feedback, leaderboard, rewards, and rankings. Thus, development of an efficient gamification approach requires selection of appropriate gamification elements that will be used in the system. In this chapter, the design of gamification approach is formulated as a multi-criteria decision-making problem, and Hesitant Fuzzy Analytic Hierarchy Process (AHP) method which integrates decision makers’ hesitancy into the decision problem is used for the solution. To this end, an extended literature review is conducted to present current gamification studies in the literature, then gamification alternatives and decision model are proposed. Finally, a real-world numerical example is conducted.

Recent technological advances enabled companies to collect, store and process a large amount of data. Automated collection of human behavior is one of the recent developments in data collection field. Companies can analyze the behaviors of their customers and get insight into their needs by using automated collection technology. In this study, we analyze location-based services data collected from a major shopping mall in Istanbul. The data is composed of 293 locations and 12070 unique visitors. The results show the most frequent routes that users follow during different periods.

Diabetes is one of the emerging threats to public health all over the world. According to projections by the World Health Organization, diabetes will be the seventh foremost cause of death in 2030 (WHO, Diabetes, 2020. https://www.afro.who.int/health-topics/diabetes). Diabetic retinopathy (DR) results from long-lasting diabetes and is the fifth leading cause of visual impairment, worldwide. Early diagnosis and treatment processes are critical to overcoming this disease. The diagnostic procedure is challenging, especially in low-resource settings, or time-consuming, depending on the ophthalmologist's experience. Recently, automated systems now address DR classification tasks. This study proposes an automated DR classification system based on preprocessing, feature extraction, and classification steps using deep convolutional neural network (CNN) and machine learning methods. Features are extracted from a pretrained model by the transfer learning approach. DR images are classified by several machine learning methods. XGBoost outperforms other methods. Dimensionality reduction algorithms are applied to obtain a lower-dimensional representation of extracted features. The proposed model is trained and evaluated on a publicly available dataset. Grid search and calibration are used in the analysis. This study provides researchers with performance comparisons of different machine learning methods. The proposed model offers a robust solution for detecting DR with a small number of images. We used a transfer learning approach, which differs from other studies in the literature, during the feature extraction. It provides a data-driven, cost-effective solution, which includes comprehensive preprocessing and fine-tuning processes.

Labor intensive and hazardous nature of the construction activities plays an important role on the increase of the amount of accidents and fatalities on sites. One of the most important sources of fatalities occurring on construction sites is falls-from-height (FFH). Despite the various efforts for the solution over decades, the yearly statistics still indicate high amount of fatalities and severe injuries due to FFH accidents on construction sites. Medical literature emphasize that the time passed after the accident is a critical factor for avoiding preventable deaths and permanent disabilities of trauma patients. The objective of this study is to timely detect FFH accidents on construction sites by using a wearable device and to provide a real-time notification to the emergency medical team (EMT) leveraging Internet-of-Things (IoT). This is expected to maintain the earliest possible medical intervention on site in order to help reducing fatal and severe consequences of FFH accidents for construction workers. Test results of the system evaluation show that the fall is detected correctly and the alert message is sent to the prescribed addresses with 100% sensitivity. The system has shown a good accuracy for true detection of the fall height with an overall error rate of 10.8%. Another metric which shows the detection of the disconnected network time of the system has been surveyed and the results are accurate with an overall error rate of 3.16%.

Clustering is an approach used in data mining to classify objects in parallel with similarities or separate according to dissimilarities. The aim of clustering is to decrease the amount of data by grouping similar data items together. There are different methods to cluster. One of the most popular techniques is K-means algorithm and widely used in literature to solve clustering problem is discussed. Although it is a simple and fast algorithm, there are two main drawbacks. One of them is that, in minimizing problems, solution may trap into local minimum point since objective function is not convex. Since the clustering is an NP-hard problem and to avoid converging to a local minimum point, several heuristic algorithms applied to clustering analysis. The heuristic approaches are a good way to reach solution in a short time. Five approaches are mentioned briefly in the chapter and given some directions for details. For an example, particle swarm optimization approach was used for clustering problem. In example, iris dataset including 3 clusters and 150 data was used.

Study design Retrospective cohort study. Introduction A new method of accurately assessing the compromised elbow's proprioception was developed for this post-surgical population using information from previous neurophysiologic proprioception studies of healthy elbows. Hypothesis This retrospective cohort study investigated the patterns and the degree of proprioceptive impairment and recovery following arthroscopic surgery for chronic lateral epicondylitis. Material and methods Participants had undergone arthroscopic elbow surgery two years prior to this study (n = 15). Healthy, non-injured volunteers with similar demographics (n = 15) served as controls. Both groups were evaluated using quantitative measures of joint position sense for proprioceptive functioning. In order to obtain the most accurate proprioceptive measurements, inter-individual interaction and visual input biases were eliminated. Retrospective chart reviews were performed to compare qualitative self-reported measures of proprioceptive function in arthroscopic surgery patients before surgery and two years post-surgery. Results Active and passive joint repositioning outcome measurements were similar between groups (p > 0.05). No significant differences were found among any angles except one: passive joint position sense at 120° of flexion (p Discussion Outcomes of arthroscopic treatment of chronic lateral epicondylitis with no decortication yielded outcomes measure similar to those of healthy controls. The sole significant difference was at 120° flexion passive joint repositioning, with a higher negative angular deviation from the target point. We propose that our study results and specific proprioception method may have implications for improving accuracy of future elbow arthroscopy and proprioceptive recovery in this population. Level of evidence II, low-powered prospective randomized trial.

Teknolojinin ve bilimin son yüzyildaki hizli gelisimi ile kendini her platformda daha da belirgin bir bicimde hissettiren rekabet kavrami, siyasi arenada da daha belirgin bir bicimde bas göstermistir. Rekabet olgusuyla birlikte daha önceleri sezgisel hareketlerde bulunan karar mekanizmalari bilimsel perspektiften bagimsiz hareket edemez hale gelmistir. Bilim insanlari farkli karar verme teknikleri üreterek ve günden güne mevcut karar tekniklerini gelistirerek piyasadaki bu ihtiyaca cevap vermeye calismaktadirlar.Oyun Teorisi ise bu karar verme tekniklerinden en göze carpani ve etkilesimli karar verme tekniklerinin en önde gelenlerindendir. Literatür incelendiginde oldukca genis bir calisma sahasi olan oyun teorisinin, siyasi arenadaki karar problemlerini cözme konusunda fazlaca calisilmadigi görülmüstür. Bu calisma bu boslugu doldurmak, bundan sonraki calismalara önayak olmak ve genel olarak secimlerinde siyasi partilerin karar vericilerine ve secmenlere fikir katki saglamak amaci tasimaktadir. Calismada secmenler icin önem tasiyan nitelikler belirlenmis, parti stratejileri incelenmis, hangi stratejinin secmen üzerinde ne kadar etkili oldugu arastirilmis ve oyun matrisine yerlestirilerek cözüme ulastirilmaya calisilmis ve bu cözümlerin yorumlari yapilmistir

The results from an experimental blasting program that was performed at the special explosives training field of the General Directorate of Security in Ankara, where new residential blocks are planned to be constructed nearby, are presented. With the objective of estimating the blast-induced ground vibration effects on the proposed structures, various blasting parameters of nineteen surface and underground explosions were recorded in two directions at three measurement stations. Site-specific empirical relationships between peak particle velocity, the amount of explosive and the distance were developed. These relationships were used to construct a practical blasting chart, which gives the maximum amount of explosive to be used as a function of distance, for future underground and surface blasting operations in the training field. Since the use of peak particle velocity in the field of civil engineering has been limited so far, site-specific parameters were also estimated to predict the blast-induced horizontal peak ground acceleration. Then, an attempt was made to investigate the dynamic responses of four and six storey reinforced concrete structures that consist of frame and shear wall type structural systems under the measured accelerations using finite element analysis. The limitations of this approach were discussed within the context of damage estimation. (C) 2013 Elsevier Ltd. All rights reserved.

The impact and popularity of competition concept has been increasing in the last decades and this concept has escalated the importance of giving right decision for organizations. Decision makers have encountered the fact of using proper scientific methods instead of using intuitive and emotional choices in decision making process. In this context, many decision support models and relevant systems are still being developed in order to assist the strategic management mechanisms. There is also a critical need for automated approaches for effective and efficient utilization of massive amount of data to support corporate and individuals in strategic planning and decision-making. Data mining techniques have been used to uncover hidden patterns and relations, to summarize the data in novel ways that are both understandable and useful to the executives and also to predict future trends and behaviors in business. There has been a large body of research and practice focusing on different data mining techniques and methodologies. In this study, a large volume of record set extracted from an outpatient clinic’s medical database is used to apply data mining techniques. In the first phase of the study, the raw data in the record set are collected, preprocessed, cleaned up and eventually transformed into a suitable format for data mining. In the second phase, some of the association rule algorithms are applied to the data set in order to uncover rules for quantifying the relationship between some of the attributes in the medical records. The results are observed and comparative analysis of the observed results among different association algorithms is made. The results showed us that some critical and reasonable relations exist in the outpatient clinic operations of the hospital which could aid the hospital management to change and improve their managerial strategies regarding the quality of services given to outpatients.