Hierarchical Cluster Analysis

Top PDF Hierarchical Cluster Analysis:

Applied Hierarchical Cluster Analysis with Average Linkage Algoritm

Applied Hierarchical Cluster Analysis with Average Linkage Algoritm

This research was conducted in Sidoarjo District where source of data used from secondary data contained in the book "Kabupaten Sidoarjo Dalam Angka 2016" .In this research the authors chose 12 variables that can represent sub-district characteristics in Sidoarjo. The variable that represents the characteristics of the sub- district consists of four sectors namely geography, education, agriculture and industry. To determine the equitable geographical conditions, education, agriculture and industry each district, it would require an analysis to classify sub-districts based on the sub-district characteristics. Hierarchical cluster analysis is the analytical techniques used to classify or categorize the object of each case into a relatively homogeneous group expressed as a cluster. The results are expected to provide information about dominant sub-district characteristics and non-dominant sub-district characteristics in four sectors based on the results of the cluster is formed.
Baca lebih lanjut

7 Baca lebih lajut

Hierarchical Cluster Analysis Terhadap Pelanggan Pasar Beringharjo Yogyakarta

Hierarchical Cluster Analysis Terhadap Pelanggan Pasar Beringharjo Yogyakarta

Pelanggan adalah elemen terpenting dalam menjalankan suatu usaha. Tingginya jumlah pelanggan akan memberikan dampak positif terhadap peningkatan jumlah keuntungan suatu usaha. Berbagai penelitian mengungkapkan bahwa pelanggan sebagai konsumen pasar mampu memberikan stimulus terhadap karier usaha di masa mendatang. Pasar Beringharjo yang terletak di Yogyakarta sebagai salah satu pasar yang selalu ramai pengunjung merupakan cerminan bahwa pelanggan sebagai faktor utama dalam kesuksesan usaha. Oleh sebab itu, peneliti melakukan sebuah kajian tentang pengelompokkan pelanggan-pelanggan yang melakukan transaksi di Pasar Beringharjo untuk dapat memberikan inovasi baru terhadap kemajuan perekonomian melalui sektor pasar, serta arah kebijakan pasar. Bentuk pengelompokan didasarkan atas kemiripan pelanggan yang melakukan transaksi di Pasar Beringharjo. Analisis cluster adalah suatu analisis statistik multivariate yang dapat mengelompokkan obyek berdasarkan karakteristik yang dimilikinya, dalam hal ini peneliti menggunakan Hierarchical Cluster Analysis. Hasilnya yaitu 98% dari seluruh pelanggan yang di surve puas melakukan transaksi jual beli di Pasar Beringharjo. Hasil penelitian ini dapat digunakan sebagai acuan untuk melakukan analisis terkait kepuasan pelanggan di sebuah usaha. Pemerintah juga dapat menjadikan sebagai landasan kebijakan-kebijakan yang akan dijalankan di pasar, khususnya Pasar Beringharjo Yogyakarta.
Baca lebih lanjut

9 Baca lebih lajut

Image segmentation by histogram thresholding using hierarchical cluster analysis

Image segmentation by histogram thresholding using hierarchical cluster analysis

The hierarchical tree of this unification process is best viewed graphically as a dendrogram. Let us consider an example gray scale image, which contains 11 · 11 pixels and consists of 43 non-empty gray levels ranging from 98 to 198 with the histogram shown in Fig. 1(a). Our method generates the dendrogram shown in Fig. 1(b). The numbers along the horizontal axis of dendrogram represent the indi- ces of the gray levels numbered from 1 to 43. The height of the links between clusters represents the order of merging operations. The estimated thresholds for the t-level thres- holding are obtained by separating the dendrogram into t groups by cutting the branch. It indicates that the multi- level thresholding is achieved quite straightforwardly by our method. For the usual two-level thresholding, i.e., the usual binarization, the threshold is obtained by cutting the dendrogram at the highest branch.
Baca lebih lanjut

7 Baca lebih lajut

PENGELOMPOKAN WILAYAH DI KOTA SURABAYA BERDASARKAN INDIKATOR KESEHATAN MASYARAKAT TAHUN 2012 DENGAN HIERARCHICAL CLUSTER ANALYSIS MENGGUNAKAN WARD`S METHOD - ITS Repository

PENGELOMPOKAN WILAYAH DI KOTA SURABAYA BERDASARKAN INDIKATOR KESEHATAN MASYARAKAT TAHUN 2012 DENGAN HIERARCHICAL CLUSTER ANALYSIS MENGGUNAKAN WARD`S METHOD - ITS Repository

Penelitian sebelumnya dilakukan oleh Sofya Laeli pada tahun 2014 mengenai analisis cluster dengan Average Linkage Method dan Ward’s Method untuk data responden nasabah asuransi jiwa unit link yang diperoleh hasil penelitian bahwa Average Linkage Method memiliki kinerja yang lebih baik dibandingkan dengan Ward’s Method dikarenakan nilai rasio simpangan baku dalam dan antar cluster lebih kecil (Laeli, 2014). Penelitian lainnya dilakukan juga oleh Azizah, Soehono, dan Solimun pada tahun 2014 mengenai analisis cluster komponen utama nonlinier dan analisis two step cluster untuk data berskala campuran yang diperoleh hasil penelitian bahwa analisis two step cluster memiliki kinerja lebih baik dalam mengelompokkan data berskala campuran dibandingkan analisis cluster hirarki dengan transformasi komponen utama non linier berdasarkan kriteria rasio Sum Square Within Cluster (SSW) dengan Sum Square Between Cluster (SSB) (Azizah, Soehono, dan Solimun, 2014). 1.2 Rumusan Masalah
Baca lebih lanjut

86 Baca lebih lajut

B1J010176 16.

B1J010176 16.

Hasil analisis cluster biomassa dan karbon tersimpan kategori pohon dan anakan pohon CLUSTER Hierarchical Cluster analysis Resemblance worksheet Name: Resem1 Data type: Similarity [r]

11 Baca lebih lajut

Low-Rank Doubly Stochastic Matrix Decomposition for Cluster Analysis

Low-Rank Doubly Stochastic Matrix Decomposition for Cluster Analysis

The resulting cluster purities and NMIs are shown in Tables 1 and 2, respectively. We can see that our method has much better performance than the other methods. DCD shows the optimal performance for 22 and 18 out of 43 data sets in purity and NMI, respectively, which is substantially more frequently than for any of the other methods. Even for some other data sets where DCD is not the winner, its cluster purities still remain close to the best method. Our method shows particularly superior performance when the number of samples grows. For the 19 data sets with N > 4500, DCD is the top performer in 17 and 11 cases in purity and NMI, respectively. Note that purity corresponds to classification accuracy up to a permutation between classes and clusters. In this sense, our method achieves accuracy very close to many modern supervised approaches for some large-scale data in a curved manifold such as MNIST 2 , though our method does not use any class labels. For text document data set 20NG, DCD achieves comparable accuracy to those with comprehensive feature engineering and supervised classification (e.g. Srivastava et al., 2013), even though our method only uses simple bag-of-words Tf-Idf features and no class labels at all.
Baca lebih lanjut

25 Baca lebih lajut

Conference paper

Conference paper

In order to verify the existence of clustering, the spatial- temporal permutation model proposed by Kulldorff et al. (2005) was utilized. Once that the database was composed only by “cases”, or the number of points with deforestation (variable) on the study area. The population at risk is the forest. As there is evidence that distances from roads contributed to increase deforestation areas (Brandão Jr et al., 2007), it was chosen the distance from the roads in this model as categorical co-variable. The distances to the nearest road were classified arbitrarily into 6 categories. Thus, every case on the database received an attribute from 1 to 6, indicating the distance from roads. For instance, cases were attributed value 1 if the warning happens between 0 – 10 km from roads; 2 if between 10 – 20 km; and so on; and finally, 6 if higher than 50 km from the roads. An accessibility map (a map of the distance from the nearest roads) can be shown in Figure 2, for the later comparison of retrospective analysis with and without the distance from roads.
Baca lebih lanjut

5 Baca lebih lajut

ANALISIS CLUSTER UNTUK DATA CAMPURAN KATEGORIK DAN NUMERIK (Cluster Analysis for Mixed Cagtegorical and Numeric Data Types)

ANALISIS CLUSTER UNTUK DATA CAMPURAN KATEGORIK DAN NUMERIK (Cluster Analysis for Mixed Cagtegorical and Numeric Data Types)

Abstract: Various clustering algorithms have been developed to group data into clusters. This paper describes clustering objects with mixed categorical and numeric data types. The methods used are two step clustering and transforms mixed data using nonlinear principal component analysis then groups the output resulted using hierarchical aglomerative clustering. The results show that the number of optimal cluster using both methods have the same optimal number of cluster but the rank of ratios of distance measure and distribution of cluster membership are different.
Baca lebih lanjut

1 Baca lebih lajut

Directory UMM :Data Elmu:jurnal:A:Agricultural Systems:Vol67.Issue3.2001:

Directory UMM :Data Elmu:jurnal:A:Agricultural Systems:Vol67.Issue3.2001:

The hypothesis that a certain socioeconomic pro®le predisposes to urban sheep keeping can be accepted. The pro®le of a likely sheep keeper in Bobo-Dioulasso can be described as follows: being in an unstable job situation (ACTCHANG) with limited chances to ®nd an ocial job because of restricted education (EDUCAT), the decision-maker of the household can make use of its high manpower potential (SUMRED), use the space and take the liberty of decision on the compound as there are no other families (NBHH) to keep sheep and often cattle (CATTLE). While according to logistic regression the parents' activity has to be added: the sheep keepers' parents often also kept livestock (ACTPAR); cluster analysis suggested to include the age (AGE) and the time spent in the town or the fact to be born in Bobo- Dioulasso (SINCE/BORNIN) with a trend to older decision makers, who spent a long time in town. The more frequent this combination of characteristics will occur, the more sheep keepers will be found in the respective town.
Baca lebih lanjut

14 Baca lebih lajut

ANALISIS DAN IMPLEMENTASI CLUSTER-SMOOTHED PADA COLLABORATIVE FILTERING ANALYSIS AND IMPLEMENTATION OF CLUSTER-SMOOTHED FOR COLLABORATIVE FILTERING

ANALISIS DAN IMPLEMENTASI CLUSTER-SMOOTHED PADA COLLABORATIVE FILTERING ANALYSIS AND IMPLEMENTATION OF CLUSTER-SMOOTHED FOR COLLABORATIVE FILTERING

Dari data awal yang terdiri dari user id, item id dan nilai rating diolah menjadi bentuk matriks m x n sehingga siap diolah pada proses clustering. Setelah terbentuk cluster, data memasuki proses smoothing. Proses smoothing ini yang akan mengubah data matriks menjadi data training. Data training inilah yang akan digunakan dalam membantu proses prediksi dari data testing sampai menghasilkan MAE sistem.

9 Baca lebih lajut

Cluster Analysis for SME Risk Analysis Documents Based on Pillar K-Means

Cluster Analysis for SME Risk Analysis Documents Based on Pillar K-Means

As a case study, we conducted it in one of national private banks in Indonesia where the SME financing is one of their core businesses. We collected about 519 risk analysis documents from the Risk Management division. All of the documents are in Microsoft Words (*.doc an *.docx format), consisting of narrative opinions in Bahasa Indonesia. There are seven opinion points delivered in the documents; those are 1) Credit Scoring 2) Financial Performance 3) Proposed Loan Facility 4) Business Performance 5) Repayment Ability and Cash Flow 6) Legal Analysis, and 7) Foreign Exchange (optional). All of the parts were analyzed based on 5Cs Credit Criteria (Character, Capacity, Capital, Condition, and Collateral).
Baca lebih lanjut

10 Baca lebih lajut

DISTINCTNESS ASSESSMENT ON YARDLONG BEAN (Vigna sesquipedalis (L.) Fruhw.) VARIETIES (CASE STUDY FOR FIVE YARDLONG BEAN VARIETIES IN PVP RIGHT APPLICATION)

DISTINCTNESS ASSESSMENT ON YARDLONG BEAN (Vigna sesquipedalis (L.) Fruhw.) VARIETIES (CASE STUDY FOR FIVE YARDLONG BEAN VARIETIES IN PVP RIGHT APPLICATION)

The result of the distinctness shows great contradiction between result of Duncan analysis and range scoring notation (Table 1) such as number of days to 50% flowering (flowering age) character. All of the candidates were located in notation 5 (medium) as well as the control varieties, excluding Putih Super and Pangeran which were located in notation 4 (early maturing). Subsequently, one difference of notation mark means the differences are not clear enough (PPVT, 2006). However, the field visual obser-vations obviously found that the flowering age of Putih Super and Pangeran had matured earlier than other varieties.
Baca lebih lanjut

9 Baca lebih lajut

Phylogenetic Analysis Of Mangosteen (Garcinia Mangostana L.) And Its Relatives Based On Morphological And Inter Simple Sequence Repeat (ISSR) Markers

Phylogenetic Analysis Of Mangosteen (Garcinia Mangostana L.) And Its Relatives Based On Morphological And Inter Simple Sequence Repeat (ISSR) Markers

489 0.78. Between G. mangostana and G. celebica similarity was 0.63. Based on this analysis, it can be assumed that G. mangostana is closely related with G. celebica and G. malaccensis. Our results strengthened the findings of Sobir and Poerwanto (2007) and Sinaga (2008) that G. celebica is similar to G. mangostana (based on AFLP markers). Matra (2010) used SSR alleles through IGMB001 (Ibaraki/IPB Garcinia mangostana Bogor 001) indicated that G. mangostana has an allele of equal size to G. malaccensis (233 bp) and G. Mangostana has the same allele at 252 base pairs as G. celebica. Yapwattanaphun (2004) stated that G. hombroniana did not group with G. malaccensis and G.mangostana based on internal transcribed spacer. Tirtawinata (2003) reported in grafting compatibility between G. mangostana and G. celebica on vegetative propagation.
Baca lebih lanjut

13 Baca lebih lajut

ANALISA KELAYAKAN PRODUK GENTENG UNGGULAN (Studi kasus Perusahaan Genteng Mendit-Malang)

ANALISA KELAYAKAN PRODUK GENTENG UNGGULAN (Studi kasus Perusahaan Genteng Mendit-Malang)

Metode yang digunakan adalah analisa cluster untuk membentuk segmen, cluster yang terbentuk cluster 1 dan cluster 2, cluster yang dominan ditunjukkan oleh jarak yang lebih panjang pada dendogram oleh sebab itu yang terpilih adalah genteng karang pilang., pemilihan genteng karang pilang karena hasil Anova karang pilang memiliki nilai F hitung lebih dari F tabel 5% sebesar 4,965. Analisa kelayakan genteng genteng karang pilang adalah dengan NPV sebesar Rp. 74478883, PI sebesar 1.90, IRR sebesar 25,44 % dan BEP atau titik impas dicapai apabila perusahaan menjual dengan harga Rp. 2032,11 per unit.
Baca lebih lanjut

1 Baca lebih lajut

Cluster analysis of typhoid cases in Kota Bharu, Kelantan, Malaysia

Cluster analysis of typhoid cases in Kota Bharu, Kelantan, Malaysia

cluster analysis of the disease provides new insight into the pattern of typhoid outbreaks. As shown in this study, the signiicant clustering occurs in year 2001 to 2005 except in year 2004. The inding was not signiicant in 2004 most likely due to the limitation of the analysis used in this study. For each of the analysis method there will be some limitation that will depend on the method of calculation used in each analysis. For the nearest neighbourhood index, the form and boundaries of the area analysed could affect the result. If the studied area is long and narrow, the expected distance would be smaller and this could make the result more likely to become a cluster. The nearest neighbourhood index is also inluenced by cases located in the same place and very close to each other. If many cases were like that, the mean distance would become smaller and could affect the result. This limitation was removed under K function analysis because it counted using all the distance relevant in the study area, thus limiting the inluence of cases that occur in the same place.
Baca lebih lanjut

8 Baca lebih lajut

Model Based Analysis Cluster Study in Normal Distribution Multivariate Data

Model Based Analysis Cluster Study in Normal Distribution Multivariate Data

Cluster analysis is data method classify objects into groups based on similarity or dissimilarity. One of approach is model based clustering. The assumptions used is the data derived from a mixture of two or more distribution probability with certain proportions. The final cluster is determined by BIC. The object of each cluster were obtained by EM algorithm. This study aims to assess the effectiveness of the model based clustering on the data are from multivariate normal distribution. Effectiveness would include the percentage of classification errors produced at a several distance, comparing with the k-means, and their application. If the distance between the center of a large and diverse cluster each different variables, then averaging the resulting classification error rate small. generally model based to cluster is more effective than the method of k-means. The MAP was better than the MLE since it can overcome the singularity problem, the rest same as MLE.
Baca lebih lanjut

145 Baca lebih lajut

Dealing with hierarchical

Dealing with hierarchical

Depth of a Sub-Tree SELECT node.name, COUNTparent.name - sub_tree.depth + 1 AS depth FROM nested_category AS node, nested_category AS parent, nested_category AS sub_parent, SELECT n[r]

34 Baca lebih lajut

ANALISIS CLUSTER UNTUK DATA CAMPURAN KATEGORIK DAN NUMERIK 80 (Cluster Analysis for Mixed Cagtegorical and Numeric Data Types)

ANALISIS CLUSTER UNTUK DATA CAMPURAN KATEGORIK DAN NUMERIK 80 (Cluster Analysis for Mixed Cagtegorical and Numeric Data Types)

Banyak metode yang dapat digunakan untuk melakukan pengclusteran, yang dapat dikategorikan sebagai metode berhirarki (Hierarchical Methods) dan metode tidak berhirarki (Nonhierarchical Methods). Metode berhirarki terbagi menjadi dua, yaitu metode agglomerative (penggabungan) dan metode divisive (pemecahan). Pada metode berhirarki penggabungan objek ke dalam kelompok-kelompok dilakukan dengan menggunakan tiga metode, yaitu metode pautan tunggal (Single Linkage Method), metode pautan lengkap (Complete Linkage Method) dan metode rata-rata kelompok (Average Linkage Method).
Baca lebih lanjut

5 Baca lebih lajut

Analisis cluster dan aplikasinya - USD Repository

Analisis cluster dan aplikasinya - USD Repository

Analisis cluster adalah salah satu teknik analisis statistik yang digunakan untuk meringkas data dengan cara mengelompokkan obyek-obyek berdasarkan kesamaan karakteristik tertentu yang dimiliki masing-masing obyek. Kesamaan karakteristik tersebut dinyatakan dalam ukuran jarak antar obyek. Pembentukan kelompok-kelompok berdasarkan jarak, obyek yang mirip seharusnya berada da- lam kelompok yang sama dan mempunyai jarak yang lebih kecil. Sebaliknya ob- yek yang berbeda berada dalam kelompok yang berbeda dan mempunyai jarak yang lebih besar.

128 Baca lebih lajut

Show all 10000 documents...