Implementasi Fuzzy C-Means Clustering dalam Pengelompokan UKM Di Kabupaten Rokan Hulu

Rouza, Erni; Fimawahib, Luth

doi:10.33633/tc.v19i4.4101

Cited by 9 publications

(8 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Metode fuzzy c-means merupakan metode pengembangan dari k-means dimana semua anggota atau objek dapat menjadi klaster. Ada beberapa kelebihan metode fuzzy c-means yakni iterasi yang digunakan cenderung lebih sedikit, partisi data yang diperoleh tepat dan akurat [3], cukup stabil terhadap outlier dan akurat dalam menempatkan titik centroid [7], serta memberikan akurasi yang tinggi [8]. Metode ward merupakan salah satu metode klaster hirarki dimana pengelompokkan dilakukan dengan meminimumkan nilai varians antar klaster.…”

Section: Pendahuluanunclassified

Perbandingan Metode Fuzzy C-Means dan Ward Pada Pengelompokkan Desa Berdasarkan Indikator Potensi Desa

Akolo,

Pratama,

Nadjamuddin

2023

Euler J. Ilm. Mat. Sains dan Teknol.

View full text Add to dashboard Cite

Bone Bolango is one of the districts that has experienced many village and sub-district expansion processes. This expansion process changes the village's potential data. Village potential is the carrying capacity for developing villages in order to improve community welfare. In order to accelerate village development, it is necessary to group villages according to their characteristics so that development is more focused and on target. The aim of this research is to group villages based on indicators of village potential so that groups of villages that have the same characteristics can be obtained, as well as to find out the best method for grouping villages in Bone Bolango Regency. The research results show that the optimum cluster for grouping villages in Bone Bolango Regency based on village potential indicators is the cluster using the ward method because it provides the smallest Xie-Beni index value compared to the fuzzy c-means method. The optimum number of clusters is three clusters. Cluster 1 has high average characteristics consisting of 57 villages, cluster 2 has low average characteristics (except livestock production) consisting of 94 villages and cluster 3 has characteristics of large area and high food production consisting of 9 villages.

show abstract

Section: Pendahuluanunclassified

Perbandingan Metode Fuzzy C-Means dan Ward Pada Pengelompokkan Desa Berdasarkan Indikator Potensi Desa

Akolo,

Pratama,

Nadjamuddin

2023

Euler J. Ilm. Mat. Sains dan Teknol.

View full text Add to dashboard Cite

show abstract

“…Research by Widiyanto (2019) In research by Rouza & Fimawahib (2020) based on small and medium business data for data clustering using Fuzzy C-Means. The best cluster is obtained, namely 3 clusters with UAT (User Acceptance Test), namely the average validation value is almost close to 1, this shows that Fuzzy C-Means Clustering has a high accuracy rate of 80-90%.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Industrial Business Grouping Using Fuzzy C-Means and Fuzzy Possibilistic C-Means Methods

Lestari,

Kartini,

Budiman

et al. 2023

telematika.

View full text Add to dashboard Cite

The industrial business sector plays a role in the development of the economic sector in developing countries such as Indonesia. In this case, many industrial businesses are growing, but the data has not been processed or analyzed to produce important information that can be processed into knowledge using data mining. One of the data mining techniques used in this research is data grouping, or clustering. This research was conducted to determine the comparison results of the Cluster Validity Index on Fuzzy C-Means and Fuzzy Possibilistic C-Means methods for clustering industrial businesses in Tanah Bumbu Regency. In each process, 5 trials were conducted with the number of clusters, namely 3, 4, 5, 6, and 7, and for the attributes used: Male Labor, Female Labor, Investment Value, Production Value, and BW/BP Value. Furthermore, this study will evaluate the Cluster Validity Index, namely the Partition Entropy Index, Partition Coefficient index, and Modified Partition Coefficient Index. This research provides the best performance results in the Fuzzy C-Means method with the results of the Cluster Validity Index on the Partition Entropy Index of 0.21566, Partition Coefficient Index of 0.88078, and Modified Partition Coefficient Index of 0.82117, and the best number of clusters is 3 with the labels of low competitive industry clusters, medium competitive industry clusters, and highly competitive industry clusters.

show abstract

“…The research conducted by Erni Raouza and Luth Fimawahib, using cross-sectional data with variables in the form of turnover, assets, and total manpower, showed that the Fuzzy C-Means Clustering method had average validation value of close to 1, indicating that the method had a high accuracy rate of 90%. The implementation and testing had been done, it is concluded that the method is able to classify the types of SMEs in accordance with the Law Number 20 of 2008 [7].…”

Section: Introductionmentioning

confidence: 99%

MSME Sales Clustering Based on Business Aid Distribution Priority Using K-Affinity Propagation

Tarisya Qurrota A'yuni,

Baiq Nina Febriati,

Lazuardy Ilham Effendie

et al. 2023

ENTHUSIASTIC

View full text Add to dashboard Cite

In rural areas of Indonesia, micro, small, and medium enterprises (MSMEs) are often isolated; however, they have been proven to play an important role as the economic backbone of millions of communities. In fact, the sluggish development of MSMEs in Indonesia become a severe problem for the community welfare. The government continues to strive for the welfare of the local communities, one of which is by supporting the existing MSMEs. However, the provision of government assistance may not be optimal for the incorrect target of the MSMEs. This study informs the government and other related parties regarding subdistrict groups whose MSMEs are considered to be their target. The k-affinity propagation method was used to find a set of representative examples, called exemplars, that best summarize the data. The result shows that sub-districts clusters based on general welfare in five commodities. K-affinity propagation algorithm clusters vary by commodity. Data fluctuation from each commodity’s three factors causes this. From this research, it can be determined which subdistricts have the most or least prosperous MSMEs in each of the five commodities analyzed.

show abstract

Implementasi Fuzzy C-Means Clustering dalam Pengelompokan UKM Di Kabupaten Rokan Hulu

Cited by 9 publications

References 1 publication

Perbandingan Metode Fuzzy C-Means dan Ward Pada Pengelompokkan Desa Berdasarkan Indikator Potensi Desa

Perbandingan Metode Fuzzy C-Means dan Ward Pada Pengelompokkan Desa Berdasarkan Indikator Potensi Desa

Comparison of Industrial Business Grouping Using Fuzzy C-Means and Fuzzy Possibilistic C-Means Methods

MSME Sales Clustering Based on Business Aid Distribution Priority Using K-Affinity Propagation

Contact Info

Product

Resources

About