Robustness of classical fuzzy C-means (FCM)

Nasution, Bahrul Ilmi; Kurniawan, Robert

doi:10.1109/icoiact.2018.8350729

Cited by 7 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the context of mapping stunting areas, clustering becomes a highly useful tool for revealing hidden structures and relationships among various regions. This process enables researchers and policymakers to better understand patterns that may be associated with stunting levels in different areas [66], [67], [68], [69], [70]. Several clustering techniques applicable to mapping stunting areas include K-Means Clustering, which groups areas based on average attribute values such as stunting rates.…”

Section: Generalized Linear Latent Variable Model (Gllvm)mentioning

confidence: 99%

“…DBSCAN (Density-Based Spatial Clustering of Applications with Noise) identifies areas with high density as clusters, aiding in discovering patterns of high stunting density in specific regions [71]. Fuzzy C-Means allows data to belong to more than one cluster with different membership levels, useful when there is uncertainty or overlap in stunting data in a particular area [68], [72], [73].…”

Section: Generalized Linear Latent Variable Model (Gllvm)mentioning

confidence: 99%

See 1 more Smart Citation

Understanding Pediatric Health Trends in Papua: Insights From SUSENAS, RISKESDAS, Remote Sensing, and Its Relevance to Prabowo and Gibran’s Free Lunch and Milk Program

Caraka,

Supardi,

Kaban

et al. 2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

This research endeavors to conduct a thorough investigation into pediatric health in the Papua and West Papua regions of Indonesia, employing a multifaceted approach that integrates data from various sources including RISKESDAS, SUSENAS, and remote sensing indicators such as NDVI (Normalized Difference Vegetation Index) and PDSI (Palmer Drought Severity Index). By amalgamating socio-economic data, health metrics, and environmental variables, the study seeks to achieve a comprehensive understanding of the factors influencing child health. Advanced analytical tools, notably the Generalized Linear Latent Variable Model (GLLVM), are utilized to conduct a dual-analysis of the intricate interplay between health indicators and environmental variables. The incorporation of NDVI and PDSI, derived from remote sensing data, introduces an ecological dimension to the investigation, facilitating a nuanced exploration of the impact of vegetation health and drought severity on pediatric health outcomes. This interdisciplinary research endeavors to unveil correlations, patterns, and potential causal relationships between environmental factors and the prevalence of various health issues among children, including pneumonia, acute respiratory infections, diarrhea, underweight, and stunting. The anticipated results aim to not only enrich our understanding of pediatric health in Papua and West Papua but also to contribute to broader discussions concerning the complex connections between health, socio-economic conditions, and environmental variables. By integrating remote sensing data with health and socio-economic datasets, this study provides a comprehensive perspective, enabling a more robust assessment of the determinants of child well-being. The insights gleaned from this research are poised to have significant implications for policymakers, environmental scientists, and public health professionals, informing the development of strategies aimed at enhancing child health outcomes in these regions. Furthermore, this study underscores the importance of initiatives like the Free Lunch and Milk Program proposed by Prabowo and Gibran in the 2024 Indonesian

show abstract

Section: Generalized Linear Latent Variable Model (Gllvm)mentioning

confidence: 99%

Section: Generalized Linear Latent Variable Model (Gllvm)mentioning

confidence: 99%

Understanding Pediatric Health Trends in Papua: Insights From SUSENAS, RISKESDAS, Remote Sensing, and Its Relevance to Prabowo and Gibran’s Free Lunch and Milk Program

Caraka,

Supardi,

Kaban

et al. 2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…To assess the hitting of the clusters obtained from data set, one of the validity indexes Partition Coefficient (PC) was used (Equation 4) [43].…”

Section: Algorithm 1 -Pseudo Code Of Fcm Clustering Algorithmmentioning

confidence: 99%

Detection of types cyber-bullying using fuzzy c-means clustering and xgboost ensemble algorithm

Ali Süzen,

Duman

2023

CRJ

View full text Add to dashboard Cite

In this study a neural network model (XGB_CTD) that will prediction which type of bullying the users may expose to, through dataset gained by the cyberbullying scale applied to the young internet users is formulated. Extreme Gradient Boosting (XGboost) algorithm, one of the ensemble learning methods is used in this method. There while this model contains 13 input parameters taken from the scale, there exist one output parameter classified one of the 9 outputs. The reliability of the data set obtained through survey is confirmed by statistical methods. Data set has been fragmented with Fuzzy C-Means (FCM) which is one of fuzzy clustering algorithms. Hyper-parameters for the maximum efficiency of the model training have been defined as model, learning and boosting method. Independent variables in data set have been scaled through standard normalization. As a result, the model has yielded % 91,75 accuracy rate in prediction of the classification as 9 different cyberbullying types. The same data set has been trained by different machine learning algorithms. It is seen that the proposed model has reached the highest accuracy when compared to the conventional machine learning algorithms. This study aims at prediction cyberbullying through the proposed model including different questions without claim by the young users as they were bullied. Similarly, type of the cyberbullying will also be able to be estimated by the help of internet using habits of the young users. Therefore, it is thought that the young can be prevented from experiencing psychological pressure or digital life fear.

show abstract

“…FCM clustering merupakan metode pengelompokan yang mempertimbangkan tingkat keanggotaan himpunan fuzzy sebagai dasar pembobotan yang memungkinkan objek untuk bergabung ke dalam suatu kelompok yang ada (Syoer & Wahyudin, 2021). FCM merupakan metode clustering yang robust ketika optimal (Nasution & Kurniawan, 2018). Metode FCM ini juga memiliki keunggulan yakni unsupervised dan dapat mencapai pusat cluster yang konsisten dan konvergen.…”

Section: Fuzzy C Means Clusteringunclassified

Pengelompokan Provinsi di Indonesia Menurut Indikator Pasar Tenaga Kerja

Ayuningtyas¹,

Suryanata²

2022

semnasoffstat

View full text Add to dashboard Cite

Daya saing tenaga kerja menjadi isu penting dalam agenda pembangunan global maupun nasional. Posisi daya saing tenaga kerja Indonesia masih berada di bawah negara lain. Capaian daya saing tenaga kerja di Indonesia dapat dilihat dari berbagai indikator pasar tenaga kerja. Tujuan penulisan ini yakni mengelompokan seluruh provinsi di Indonesia berdasarkan indikator pasar tenaga kerja tahun 2019 menggunakan metode Fuzzy C-Means (FCM) clustering. Seluruh provinsi berhasil dikelompokkan menjadi 3 cluster. Cluster pertama beranggotakan 2 provinsi dengan karakteristik daya saing cukup baik. Cluster kedua beranggotakan 4 provinsi dengan karakteristik memiliki produktivitas tenaga kerja dan rata-rata upah tenaga kerja yang tinggi, namum memiliki permasalahan dalam pencapaian indikator pasar tenaga kerja lainnya. Sedangkan cluster ketiga beranggotakan 28 provinsi dengan karakteristik capaian indikator yang moderat. Indeks ketepatan, seperti Dunn's Fuzziness Coefficients dan Fuzzy Silhouette Index menunjukkan nilai mendekati 1 yang menunjukkan bahwa cluster yang terbentuk telah memiliki struktur yang kuat. Kebijakan yang tepat sesuai kelompok sasaran diharapkan dapat mendorong peningkatan daya saing tenaga kerja Indonesia agar dapat berkontribusi dalam tercapainya tujuan pembangunan.

show abstract

Robustness of classical fuzzy C-means (FCM)

Cited by 7 publications

References 16 publications

Understanding Pediatric Health Trends in Papua: Insights From SUSENAS, RISKESDAS, Remote Sensing, and Its Relevance to Prabowo and Gibran’s Free Lunch and Milk Program

Understanding Pediatric Health Trends in Papua: Insights From SUSENAS, RISKESDAS, Remote Sensing, and Its Relevance to Prabowo and Gibran’s Free Lunch and Milk Program

Detection of types cyber-bullying using fuzzy c-means clustering and xgboost ensemble algorithm

Pengelompokan Provinsi di Indonesia Menurut Indikator Pasar Tenaga Kerja

Contact Info

Product

Resources

About