Hierarchical Cluster Analysis [Simply explained]

Рет қаралды 54,891

Күн бұрын

What is Hierarchical Cluster Analysis? And how is it calculated?
A hierarchical cluster analysis is a clustering method that creates a hierarchical tree of objects to be clustered (Dendrogram). The tree represents the relationships between the objects and shows how the objects are clustered at different levels.
► Load sample data set
datatab.net/statistics-calcul...
► Online Calculator Hierarchical Cluster Analysis
datatab.net/statistics-calcul...
► Hierarchical Cluster Analysis Tutorial
datatab.net/tutorial/hierarch...
► E-BOOK
datatab.net/statistics-book
00:00 What is Hierarchical Cluster Analysis?
00:31 Example of Hierarchical Cluster Analysis
00:50 Calculate hierarchical cluster analysis
06:32 Calculate hierarchical cluster analysis online

Пікірлер: 51

@4chanFootballMemes 3 ай бұрын

I loved learning about "Heyrakikal" clustering

@odiakaolika5715 2 ай бұрын

You just made my evening with your simple explanation.

@datatab 2 ай бұрын

Glad it was helpful and many thanks for your feedback! Regards Hannah

@ozgurogur1297 Жыл бұрын

I found it very understandable and simple. thanks a lot!

@alexfrancois 11 ай бұрын

Beautifully explained, thanks! 🙏 Incredibly clear.

@masteroftheworld001 9 ай бұрын

well explained thank you so much

@shawnkim6287 9 ай бұрын

thank you so much. you clarified a lot!!!! 😀

@manuelleitner3196 Жыл бұрын

Great video, thank you!!!

@datatab Жыл бұрын

My pleasure!

@nakirambau7632 7 ай бұрын

thank you so much, you have explained it so well

@datatab 7 ай бұрын

Glad it was helpful!

@saurabhjoshi3010 6 ай бұрын

nicely explained

@osmancetinkaya8930 Жыл бұрын

How might be the sqr of 17 (16+1) =equal to 3,162 ? it must be 4,123 is not?

@manuelruelas3496 9 ай бұрын

The error is that the x distance is 3 (from 1 to 4) not 4, so it’s the sq root of 10.

@rodidoesburg4061 4 ай бұрын

How do you name the clusters? Just from left to right, so cluster 1, cluster 2, cluster 3. Or are there more methods to name a cluster?

@matheusdelima1743 Жыл бұрын

Great content. I'm a fan :)

@datatab Жыл бұрын

Glad it was helpful and many thanks for your nice feedback! Regards Hannah

@iqraahmad130 Жыл бұрын

youre kinda cute

@ibrahimabubakarzango9803 2 ай бұрын

Pls endeavour to avoid making mistakes thanks for comment section i could have got it so difficult to comprehend. That aspect of sqrt of 17 is terrible. But u did well and this video is good too

@datatab 2 ай бұрын

Hi thanks for youre feedback! We try to avoid mistakes, sorry for that and for the resulting trouble! Regards, Hannah

@Oladayo1 2 ай бұрын

well, that's because it's the sqrt of 10 not sqrt of 17. The mistake was using 4 instead of 3

@nazhifmuh.kasyfan2148 Ай бұрын

I would like to ask, is Hierarchical Cluster Analysis always associated with the Euclidean Distance? Thank you

@datatab Ай бұрын

Hi many thanks for your question, Hierarchical Cluster Analysis (HCA) is not always associated with the Euclidean distance. While Euclidean distance is commonly used, HCA can work with various distance metrics depending on the nature of the data and the analysis goals. Here are some common distance metrics used in HCA: - Euclidean Distance: This is the straight-line distance between two points in a multi-dimensional space. It's one of the simplest and most widely used distance metrics. - Manhattan Distance (also known as City Block or L1 distance): This is the sum of absolute differences between coordinates. It can be suitable when diagonal movement isn't meaningful. - Cosine Similarity: This measures the cosine of the angle between two vectors, commonly used in text analysis and other contexts where vector magnitude might vary. - Mahalanobis Distance: It accounts for correlations in data by incorporating the covariance matrix, making it suitable for data with different scales and correlations among variables. - Minkowski Distance: A generalization of Euclidean and Manhattan distances, with a parameter 'p' to control the degree of the norm. - Correlation-based Distance: This distance uses the correlation between data points rather than absolute differences. It's common in gene expression analysis or other contexts where relationships between variables matter more than absolute values. I hope this was helpful : ) Regards Hannah

@ricardorivashernandez4023 8 ай бұрын

Real good!

@samuraixyz22 Жыл бұрын

I would like to KZfaq tutorials like this. Do you have recommendations on what softwares to use?

@datatab Жыл бұрын

DATAtab : )

@samuraixyz22 Жыл бұрын

@DATAtab where can you learn more about it?

@rileyharper7679 6 ай бұрын

The Euclidean distance horizontal component at 2:17 should be 3 not 4 since 4 - 1 = 3. Also, the manhattan distance should be 4 and the maximum distance should be 3 for the same reason.

@playbros332 5 ай бұрын

I agree they are wrong, but shouldn't it be square root of 17, which is 4.12?

@fabianr9394 5 ай бұрын

Because you go 3 steps to the right and 1 up; so sqrt(3^2 + 1^2)@@playbros332

@luisamar8214 26 күн бұрын

How you calculate the distances between Lisa, Joe with the others?? you have a group of positions not just one... how do you do that? thankss!

@datatab 26 күн бұрын

Hi, in this case you would first calcualte the center between Lisa and Joe and then the diestance from this center to one other Person. Regards Hannah