Silhouette Method Research Articles

Classification of tall tower meteorological variables and forecasting wind speeds in Columbia, Missouri

The wind speeds given in 10 min intervals is forecast using multiple methods inclusive of persistence, statistical methods of ARIMA as well as artificial intelligence methods of Artificial Neural Networks. Tall tower meteorological variables in Columbia, Missouri are clustered using Self-Organizing Maps after the optimal number of clusters was determined using the Elbow and Silhouette methods among others. The optimal number of clusters, k was given as 4 for all methods. The data were then grouped into three Intervals which consisted of approximately 50 percent and over of vectors or rows from the data frame. These intervals were then used as training and testing for the forecast models of Long Short-Term Memory Networks with pressure and wind speeds as inputs as well as lagged wind speeds as inputs. Other models using these intervals in our analyses include Moving Autoregressive Integrated Moving Average (ARIMA) and persistence. From the results obtained from the ARIMA, the metric of the root mean square error (RMSE) ranged from approximately 0.6 to 1.0 ms−1 for forecast horizon 2 to 12 in increments of 2. Interval2 had the upper and lower values and thus showed most variability in errors because it encompassed most of spring, all of summer and the beginning of fall. The moving ARIMA showed lower errors than the LSTM with pressure and wind speeds inputs for all the intervals. This may be attributed to the difficulty in representing the system’s non-linearity and high dimensionality by using just the wind speeds and pressure as inputs. The lagged co-ordinates of the wind speed was then examined and used as inputs for the LSTM. The metric used for the evaluation of prediction of the forecast horizons of 60, 120, 180, 240, 300 and 360 min or 1, 2, 3, 4, 5 and 6 h ahead is the Normalized Root Mean Square Error (NRMSE). These models were compared to the benchmark model of persistence. It was determined that all of the models beat persistence and the LSTM with the lag series outperforms the LSTM with pressure and wind speed as inputs. The Moving ARIMA is now beaten by the lagged series LSTM in all intervals for at least 2 time forecast horizons of 60 and 120 min or 1 and 2 h. It is thus shown that the Artificial Neural Network method with the lagged series inputs is the best performing model.

Penerapan K-Medoids Clustering Dan Silhouette Method Untuk Strategi Pemasaran Program Donasi Pada Lembaga Amil Zakat

Donation data management is a complex challenge for amil zakat institutions in designing an effective marketing strategy for fundraising programs. In this study, the k-medoids algorithm was used to cluster the donation data with the aim of identifying patterns and characteristics of donors. The k-medoids algorithm was chosen because of its ability to handle unusual data and non-numeric attributes. Through clustering analysis, this study classifies donors based on attributes such as the number of donations, the frequency of donations, and the time interval for donating. And in determining the number of clusters in this study using the silhouette method to measure the quality of the resulting clustering. And getting the most optimal number of clusters is k = 3 with a silhouette score of 0.598782. The results of the study found that groups of donors had similar characteristics, such as donors who made high donations with regular frequency and donors who focused on donations for specific purposes. These findings can be used by charitable organizations in developing more effective fund management, marketing and targeting strategies

Analysis of Elbow, Silhouette, Davies-Bouldin, Calinski-Harabasz, and Rand-Index Evaluation on K-Means Algorithm for Classifying Flood-Affected Areas in Jakarta

Jakarta is the capital city of Indonesia, which has a high population density, and is an area that is frequently hit by floods. This study aims to determine the classification of flood-affected areas in Jakarta between severe, moderate, and low. Design/method/approach: The study was conducted using the elbow, Silhouette, Davidson-Bouldin, and Calinski-Harabasz methods on the K-means algorithm, as well as the Rand method. index for evaluation. Grouping with 3 and 6 groups is the best grouping value based on Calinski-Harabasz. By using the davies bouldin index from the observations, the K value with a value of 6 has the smallest Davies-Bouldin value with a value of 0.2737. By using sillhoute, the experimental results obtained the best values sequentially, namely K=2, K=3, and K=6 with silhouette values of 0.866, 0.854, and 0.803. In this experiment, based on the elbow method, it was found that the best K value was K=3. This was obtained because it was based on observations on the appearance of the SSE data compared to the value of K. In the graph above, it can be seen that the largest decrease in data occurred at K=3 and after this decrease, the decline began to slope. The rand index is a method used to compare several cluster methods. If the value is >= 90 it is a very good result, if the value is in the range 80 to 90 it identifies a good index, whereas if it is below 80 it indicates a bad index. The results show that cluster three is verified as the best cluster with a value of 1, followed by a second alternative with cluster 2 of 0.9182. From several validation and evaluation methods it can be concluded that the best grouping can be done using 3 clusters. The results of the study yielded a value of 75.4% in low areas, 21.1% in moderate areas, and 3.5% in severe areas.

Which multimorbidity clusters in hypertensive patients are associated with longer hospital stays?

Abstract Funding Acknowledgements Type of funding sources: None. Introduction The number of people with hypertension will increase by 15-20% by 2025. Research shows that up to two thirds of hypertensive patients have other comorbidities. Multimorbidity defined as the coexistence of two or more diseases in the same person, is now a common phenomenon. Patients with multimorbidity have poorer health outcomes and are more likely to use unscheduled hospital healthcare. By identifying hypertensive clusters, their need for hospital care and the cost of care can be determined. Patient profiling has the potential to improve treatment outcomes and increase the efficiency of care. Purpose Identification and characterization of multimorbidity clusters in hypertensive patients and assessment of which of them were responsible for longer hospitalization. Material and Methods Data from 489 patients admitted to the cardiology department with hypertension as the main diagnosis over three years was used for the analysis. 163 patients (81 women, 82 men, mean age 63.93 ± 11.49) had multiple conditions (≥ 2 conditions). Patient profiles for each group were described by age, gender, hypertension grade and length of hospital stay. Partitioning Around Medoids method was used to divide patients into clusters. Dissimilarity between patients was measured with Gower distance. Number of clusters was determined using Silhouette method. Significance level for all statistical tests was set to 0.05. Results Hypertensive patients reported comorbidities such as heart failure (14%), diabetes (28%), chronic kidney disease (14%), stroke (12%), acute coronary syndrome (8%), obesity (37%). There were 12 clusters for the study group: (1) without comorbidities, (2) obesity only, (3) obesity and diabetes mellitus (DM),(4) heart failure (HF) and chronic kidney disease (CKD),(5) DM,(6) acute coronary syndrome (ACS),(7) obesity, stroke, and DM,(8) CKD and other diseases without HF,(9) obesity and ACS, (10) stroke, (11) HF and obesity,(12) DM, HF and often obesity. It was noted that ACS patients significantly older than patients without comorbidities and from patients in clusters 2, 3, 7, 8 and 10. In addition, patients with DM only and patients with DM, HF and obesity were significantly older than patients who were only obese. Hospitalization time was significantly longer in patients with HF than in patients from clusters 1, 2, 5, 7, 10. Significantly longer in patients with CKD but without HF than in clusters 1, 5, 7. Significantly longer in patients with HF and obesity than in clusters 1 and 7. Significantly longer in patients with obesity and DM and in patients with DM, HF and often obesity than in patients without comorbidities. Conclusions Comorbidities affect the length of hospitalization of patients with hypertension. Comorbidities such as HF, CKD, DM and obesity will prolong the hospitalization of hypertensive patients. Clinical profiles of patients allow forecasting the need for longer and more expensive medical care.

Contagion Patterns Classification in Stock Indices: A Functional Clustering Analysis Using Decision Trees

This paper aims to identify the main determinants of the countries that present contagion during the period 2000–2021, based on the determination of the behavior patterns of 18 stock market indices of 15 of the main economies. To do that, first, the B-spline method and Bezier curves are used to smooth observations by minimizing the noise. Subsequently, the Functional Principal Component Analysis (FPCA) methodology is applied. Then, the K-means clustering algorithm is used to determine the main groups using the silhouette method and cross-validation, considering the sum of squares of the distances as the function to minimize. Finally, classification trees and macroeconomic and financial analyses are used to determine the rules of variables that give a direct explanation of the contagion (clustering) between the stock indices. The main empirical results obtained suggest that the most significant macroeconomic variables are the Gross Domestic Product, the Consumer Price Index, and Foreign Direct Investment, while in the financial aspect and the most representative are Domestic Credit and number of companies listed on the stock market. It is worth noticing that government spending does not have a significant effect at any time as a determinant of contagion. Finally, it is important to mention, and surprising, that Mexico’s IPC was not clustered in the same group of US stock market indices anytime, despite the strong commercial relationship and the geographical closeness.

Towards Detecting Influential Members and Critical Topics from Dark Web Forums

Conventionally, the Internet consists of three parts: Surface, Deep, and Dark Webs. Inthe last two decades, a massive increase in illicit activities took place on the differentplatforms of the Dark Web. Moreover, social networks on Dark Web implicateextremism dissemination on a wide scale. In this paper, we propose an approach togenerate textual patterns from discussions on Dark Web terrorist forums employingData Mining techniques. The discovered patterns help identify the influential membersand extract critical topics. We describe our system modules that perform datapreprocessing, text preprocessing with TF-IDF weighting, outlier detection, clusteringevaluation, clustering, and clustering validation, implemented with the RapidMinertool. We apply K-Means as the Clustering method with different distance metrics,evaluate the clustering process using Elbow and Silhouette methods, and validate itusing Davies-Bouldin Index. Furthermore, we investigate the effects of altering thedistance metrics for outlier detection on the Clustering results.

A practical method for generating whispers from singing voices: Application of improved phantom silhouette method

A practical method for generating whispers from singing voices: Application of improved phantom silhouette method

A combinatorial model for natural gas industrial customer value portrait based on value assessment and clustering algorithm

Frequent geopolitical events have reduced the stability of natural gas supply and caused drastic price fluctuations, which poses a new challenge to the natural gas consumer market. To improve the anti-risk ability of the natural gas industrial market, this study constructs a new customer value portrait framework to discern the industrial customer value based on different types of behavioral features and the emerging trends of the natural gas market. Specifically, we rediscover the value composition of natural gas industrial customers and establish a set of indicators to reflect the customer value in different dimensions with mixed data types. Then, a visualizable customer value classification model has been established by combining Gower’s dissimilarity coefficient with the PAM clustering algorithm. To ensure the accuracy of the clustering results, the optimal number of clusters is determined by gap statistics and elbow point, and the average silhouette method is used to detect the clustering effect as well as used in misclassified sample identification. To verify the applicability of the model, we used a certain amount of natural gas industrial customer data from a large state-owned oil and gas enterprise for application analysis and effectively divided customer value into three groups, demand-serving, demand-potential, and demand-incentive, according to their value characteristics and behavioral features. The results indicate that the framework proposed in this study can reasonably reflect and better characterize natural gas industrial customers’ value with different types of behavioral feature data, which can provide technical support for big data smart natural gas consumer marketing.

Features of the clusterization method application in marketing research of the pharmaceutical market of Ukraine

The aim of the article. The purpose of this study is to group companies operating on the pharmaceutical market of Ukraine, using clustering methods based on sets of statistical indicators of their activities for the period 2008-2022, with the aim of the isolated groups further evaluating in terms of perspectives for the development of their marketing activities in this market. Analysis results. The article is devoted to the features of the cluster analysis using with the aim to distinguish groups of pharmaceutical enterprises operating on the pharmaceutical market of Ukraine based on the use of a unified system of performance indicators. Key features of the clusterization methods application are given, which will allow to form groups from individual elements of the studied population based on scientifically proven approaches and to develop productive managerial solutions in the field of their marketing activities in accordance with the obtained results. The current state and key trends in the development of the pharmaceutical market of Ukraine are analyzed. Cluster analysis was implemented using the Sklearn library in the jupyter notebook environment. The Silhouette method was used to determine the number of clusters. At the final stage of clusterization it is necessary to profile the selected clusters by describing their key parameters and characteristics. As a result, three clusters were obtained, which are conventionally named as "Leaders", "Average" and "Outsiders" in accordance with the revealed characteristic features and indicators of market activities performance of the studied enterprises.In this way, the method cluster analysis makes it possible to consolidate a large number of competing enterprises fairly quickly by a set of indicators important for researchers, which have different natures, measurements, sources of their generation, as well as different frequencies of relevant calculations. In turn, the visualization of the results of this method application allows to visualize the promising directions of a more in-depth marketing analysis of the features of the market activity of competing pharmaceutical enterprises. Conclusions and directions for further research. Clustering of enterprises operating on the pharmaceutical market of Ukraine allows to visualize their current market positions on the basis of the consolidation of both the observation units themselves and the key indicators of their activities. It also allows to identify the main features and characteristics of their marketing activities, which is a valuable resource for further competitive benchmarking by those pharmaceutical companies that claim to strengthen their own market positions based on the development and practical application of effective marketing strategies and tactics of their activities in conditions of increased competition on the market and a number of negative trends currently taking place on it.

Toward a smart health: big data analytics and IoT for real-time miscarriage prediction

BackgroundWe are living in an age where data is everywhere and grows up in a very speedy way. Thanks to sensors, mobile phones and social networks, we can gather a hug amount of information to understand human behavior as well as his individual life. In healthcare system, big data analytics and machine learning algorithms prove their effectiveness and efficiency in saving lives and predicting new diseases. This triggered the idea of taking advantages of those tools and algorithms to create systems that involve both doctors and patients in the treatment of disease, predict outcomes and use real-time risk factors from sensors and mobile phones.MethodsWe distinguish three types of data: data from sensors, data from mobile phones and data registered or updated by the patient in a mobile app we created. We take advantages from IoT systems such as Raspberry Pi to collect and process data coming from sensors. All data collected is sent to a NoSql Server to be then analyzed and processed in Databricks Spark. K-means centroid clustering algorithms is used to build the predictive model, create partitions and make predictions. To validate results in term of efficiency and effectiveness, we used clustering validations techniques: Random K, Silhouette and Elbow methods.ResultsThe main contribution of our work is the implementation of a new system that has the capability to be applied in several prediction disease researches using Big Data Analytics and IoT. Also, comparing to other studies in literature that use only medical or maternal risk factors from echography; our work had the advantage to use real-time risk factors (maternal and medical) gathered from sensors, react in advance and track diseases. As a case study, we create an e-monitoring real-time miscarriage prediction system to save baby’s lives and help pregnant women. In fact, doctors receive the results of clustering and track theirs patient through our mobile app to react in term of miscarriage to avoid non-suitable outcomes. While pregnant women receive only advices based on their behaviors. The system uses 15 real-time risk factors and our dataset contains more than 1,000,000 JSON files. Elbow method affirm three as the optimal number of clusters and we reach 0.99 as a value of Silhouette method, which is a good sign that clusters are well separated and matched.

PREVENTION STRATEGY OF MAIN NON-COMMUNICABLE DISEASES USING ARTIFICIAL NEURAL NETWORKS

ABSTRACT: The aim of this work was to build four models for the prediction of Noncommunicable Diseases (NCD) to support early diagnosis by applying deep Artificial Neural Networks (ANN) and clustering; through comparative analysis, the best paradigm was chosen. Following the methodology for data science, the objectives, project requirements and the quality of the data were analyzed, cleaning and normalization was carried out, the identification of the optimal number of groups by means of the silhouette method. Each model was trained with 70% of records and validated with the remaining 30% using the confusion matrix and F1. The comparative analysis showed the best performance of the multilayer perceptron with deep learning (over a network with clustering radial basis functions), its accuracy, precision, sensitivity, specificity and F1 were 99%, 97%, 100%, 97% and 98% respectively for lung cancer, 99%, 98%, 100%, 100% and 98% for breast cancer, Alzheimer with 98.6%, 100%, 96%, 100% and 97%, and depression with 91% , 87%, 93%, 88% and 88%, these models constitute an automated solution to strengthen medical diagnosis, they are considered a strategic support in the prevention of NCD, will contribute to improving the prognosis and quality of life of vulnerable people. Keywords: Prediction, artificial neural networks, data science, physical disorders, mental disorders

Performance analysis of Kmeans with modified initial centroid selection algorithms and developed Kmeans9+ model

Partition the sample space to the user requirement is an easy and efficient cluster method in many applications. Kmeans is one of the simple and efficient partition algorithms used in many cluster solutions. We suggest Multivariant Silhouette method to predict the cluster count value for Kmeans algorithm. The work proposes three new algorithms for initial seed selection. The Kmeans algorithm is modified using statistical measure Mean, Median, Partition centre for the cluster centroid calculation. In conventional Kmeans algorithm, the samples are compared with all the partition centroids to decide the inclusion of a sample in a cluster. In Kmeans9+ algorithm, the samples are compared only with the centroids of current and eight nearest neighbouring cluster partitions. It reduces the needless comparisons of samples with cluster centroids and improve the efficiency of the algorithm. The performance of created algorithms is evaluated using data from UCI and KBPE SSLC. Cluster efficiency analysis is done using the cluster evaluation indices such as Silhouette and Dunn Index. Nine nearest neighbour uniform partition cluster model Kmeans9+ improve the performance of the K means algorithm and obtain the natural cluster results with minimum iterations.

K-평균 알고리즘을 이용한 스마트팜 농작업별 소요시간에 따른 군집분석

This study attempts to establish a standardized model of smartfarm working process for paprika. First, we classify main works by each production stage and then we evaluate difficulties and lengths of major works by cross-check. Finally we investigate real working hours of each main work in twenty paprika smartfarms. Based on the real data of working hours, we utilize a cluster analysis using the K-means algorithm and Silhouette method in order to categorize paprika smartfarms. The results from the K-means cluster analysis show that sixteen farms were clustered into ten groups. We conclude that each farm was not clearly categorized.

K-Means Clustering Algorithm Approach in Clustering Data on Cocoa Production Results in the Sumatra Region

Cocoa agricultural production in Indonesia is currently very low while demand continues to increase every year, so it is very important to build a model that can categorize cocoa farming data. The main objective of this research is to analyze agricultural data using data mining techniques that specifically use the K-Means Clustering algorithm, and Gaussian Mixture Models. In this research, we used quantitative research because it measure number-based data. The results of cocoa production so far still depend on land area, then the number of cocoa trees has a significant effect on the amount of production so it is very important for the government and researchers to develop technologies that can increase cocoa production yields where the demand for cocoa is currently very high in demand worldwide because it can classify the cocoa quality from good quality to poor quality. Based on testing the K-Means Clustering and Gaussian Mixture Model algorithms on data on cocoa production in four provinces, namely North Sumatra, West Sumatra, Lampung and Aceh which were optimized by the Silhouette method, it produced cluster values ​​of 2, 3 and 4. second with a value of 59.8%.

SEGMENTASI PELANGGAN MENGGUNAKAN METODE K-MEANS CLUSTERING BERDASARKAN MODEL RFM (RECENCY, FREQUENCY, MONETARY)

Companies or entrepreneurs must better understanding customers data in all aspects, including detecting similarities and differences among customers, predicting their behavior, and offering better options and opportunities to customers. Customer segmentation is carried out to obtain this information, which is part of CRM (Customer Relationship Management). One of the general models in the application of customer segmentation is the RFM (Recency, Frequency, and Monetary) model. This research method uses a combination of the RFM model and clustering. RFM is used as a description of customer behavior in conducting transactions. Clustering is a process that is widely used and is designed to categorize data. Clustering uses the K-Means Algorithm to determine the number of clusters using the Elbow and Silhouette methods. The application of RFM analysis and the K-Means resulted in two customer segments, namely potential customers and non-potential customers. Potential customers have the characteristics of frequent transactions and also large expenses. Non-potential customers have the characteristics of infrequent transactions and also standard expensesKeywords: Customer Segmentation, RFM Model, K-Means Clustering

ANALISIS CLUSTER FAKTOR PENUNJANG PENDIDIKAN MENGGUNAKAN ALGORITMA K-MEANS (STUDI KASUS: KABUPATEN KARAWANG)

Education is one of the means to create and improve the quality of better human resources. This is expected to improve human welfare. Based on data from the Ministry of Education and Culture, there are 4 sub-districts in Karawang Regency that do not have state high schools. This can result in difficulties for students who have financial deficiencies which can ultimately lead to dropping out of school. Then apart from that distance can also be an obstacle. The purpose of this study is to apply the clustering method for the distribution of educational supporting factors in Karawang District so that later this research can improve the quality of education evenly in Karawang District, not only concentrated in certain areas by paying attention to educational supporting factors. The algorithm used in this research is K-Means. The elbow method used in determining the optimal cluster supported by the silhouette method resulted in the best number of clusters being 2 clusters. The results of the clustering evaluation resulted in Davies-Bouldin Index (DBI) value of 0.408 and Silhouette Coefficient value of 0.747 (strong structure). Cluster 1 consists of 7 sub-districts and Cluster 2 consists of 23 sub-districts. Based on the results of clustering analysis, Cluster 1 has the average number of attributes of schools, teachers, classes, laboratories, libraries at all levels of education (SD, SMP, SMA, SMK, and SLB) with state status and the population shows higher results if compared with the average number of each attribute in Cluster 2. So it can be concluded that Cluster 1 is a high category and Cluster 2 is a low category.

EPCO-18. UNSUPERVISED MACHINE LEARNING ON TUMOR IMMUNE TRANSCRIPTOMIC DATA REVEALS DISTINCT IMMUNOLOGIC SUBTYPES OF GLIOBLASTOMA

Abstract Glioblastoma (GBM) has demonstrated a limited therapeutic response to single-agent immunotherapies, highlighting the need for a better understanding of GBM tumor immune phenotypes to inform targeted immunologic treatments. To elucidate GBM immunologic phenotypes we performed unsupervised machine learning on microarray transcriptomic data from 815 newly diagnosed GBM patients. We utilized immune genes only and performed k-means clustering for k=2-10 with within-cluster sum of squares and silhouette methods used to identify the optimal number of clusters. Identified subtypes were further characterized using xCell to estimate the tumor infiltration of 64 immune and stromal cell types. Two primary immunologic subtypes were identified: subtype 1 (423 patients) and subtype 2 (392 patients). There was no difference in patient age (mean age 58 years) or gender between subtypes. Subtype 1 was primarily classical subtype GBMs (50.5%, p< 0.001), while subtype 2 was composed of primarily mesenchymal subtype GBMs (74.2%). Relative to subtype 1, subtype 2 was characterized by high expression of immune checkpoint and co-stimulatory proteins, including CD86, OX40L, IDO1, and LAG3. Subtype 2 also had increased activation of immune genes associated with cytolytic activity and antigen processing. In addition, subtype 2 had increased infiltration of M2 macrophages, monocytes, dendritic cells, and neutrophils while subtype 1 had higher infiltration of CD8 naïve T cells. Multivariate logistic regression confirmed that subtype 2 was characterized by monocytes, M1 and M2 macrophages, neutrophils, and dendritic cells while subtype 1 was characterized by CD8 naïve T cells, regulatory T cells, and naive B cells. Subtype 2 had reduced progression-free survival, but there was no difference in overall survival between the two subtypes. In summary, unsupervised machine learning of GBM immune transcriptomic data identified two immune phenotypes, one of which was more immunologically active than the other, potentially informing future personalized immunotherapy treatments.

To Discriminate General Election system in Thailand by using K-Means Clustering

Thailand uses the ballot paper in the general election since 1933. In the present day, technology has been involved in daily basics. This study aims to explore the hypothesis of the election system in Thailand which can use the technology in the election process name as electronic voting (e-voting) or still need to use the traditional method. Before implementation, it should study in terms of the area that is ready to implement the new method and where still need to use the current one. This study takes the relevant factors to analyze with the data of each area collected from various sources. The clustering method used in this study is k-means. Then to find the acceptable k cluster the silhouette method is used. The result is 2 cluster is a perfect fit with the 11 factors that used in this study. The first cluster is Bangkok which is the capital city to be matched with the e-voting method and the second cluster is the remainder province in Thailand (76 provinces). This can be used for the next study in terms of supply chain design for e-voting and developing the ballot paper logistics.

An unsupervised cluster-based feature grouping model for early diabetes detection

Diabetes mellitus is often a hyperglycemic condition that poses a substantial threat to human health. Early diabetes detection decreases morbidity and mortality. Due to the scarcity of labeled data and the presence of oddities in diabetes datasets, it is exceedingly difficult to develop a trustworthy and accurate diabetes prognosis. The dataset and groupings of the features using the elbow and silhouette methods have been clustered using K-means. Various machine learning approaches have also been applied to the cluster-based dataset to predict diabetes. We propose an unsupervised cluster-based feature grouping model for early diabetes identification using an open-source dataset containing the data of 520 diabetic patients. On the cluster-based dataset and the complete dataset, the maximum Accuracy (ACC) is 99.57% and 99.03%, respectively. The best Precision, Recall, minimum mean squared error (MSE), maximum mean squared error (MSE), and F1-Score of 1.000 are obtained from multi-layer perceptron (MLP), random forest (RF), and k-Nearest Neighbors (KNN), 0.984 from random forest (RF) and support vector machine (SVM), 0.010 from RF, 0.067 from KNN, and 99.20% from RF, respectively. A comparison table displays the anticipated outcomes and highlights the aspects of this research that are most likely to occur as intended. The preprocessed data and codes are available on the GitHub repository to https://github.com/mhashiq/Early-stage-diabetes-risk-prediction.

Assessment of Accessibility and Activity Intensity to Identify Future Development Priority TODs in Hefei City

Hefei, the capital city of Anhui Province, China, has been experiencing rapid development due to fast population growth and continuous city expansion since 2010. The Hefei government began to operate the subway system in 2017 and introduced the concept of transit-oriented development (TOD) as a strategy to reduce traffic pressure and environmental pollution. By 2020, there were 77 TODs in operation. However, some TODs could not attract enough public transport passengers or increase the economic activities. This study analyzed the Hefei City TOD network and tried to identify TODs that need development priority among the existing TODs to guide the efficient allocation of resources for the development of the TOD network. First, this study measured the accessibility and activity intensity at each TOD by using the node–place model. Second, the 77 existing TODs were divided into 4 priority levels by applying the silhouette method. Level 1 and level 2 TODs have development priority and are referred to as “Unstable TODs” and “Unbalanced TODs,” respectively. Finally, this paper provides some strategies for developing these priority TODs.