# Data analysis

Hi,

What nodes are used to analyse numerical data in KNIME (Data analysis). Also how can you calculate the minimum, maximum, mode and standard deviation in knime

Thank you.

Hi,

there are so many KNIME nodes that can be used for analysing numerical data taht it would be really hard to list them all here. Which one can be used largely depends on your application and on what you are trying to achieve with it. If you can be more specific, I may be able to point you in the right direction.

Regarding your second question, look at the Statistics node, it can calculate all the basic statistics for multiple columns in a data table.

Cheers,
Marco.

Hi,

If you could sugest nodes that can apply descriptive statistic techniques, something that also provides contexual programming and usage of graphical representation for example sorting data by class, line or bar plot each of the features individually.

Thank you

Hi,

the Statistics node, already mentioned, provides the most common descriptive statistics and also produces histograms for each data column.

To sort your data you can use the Sorter node. To include/exclude specific data lines you can use the Row Filter node.

If you want to interactively bin and group your data, you can use the Histogram (interactive) node. Other nodes in the same group can produce box plots, line plots and many more graphical analysis and representations. If you want to do a more flexible binning, you can use the Auto-Binner node.

Last but not least, the very flexible GroupBy node allows to "pivot" data in multiple ways, using a combination of grouping and aggregation.

I hope this is enough to get you started.

Cheers,
Marco.

Hi,

Thank you for the help, I wanted to ask what does the groupby node do, does it create tables and can it be used to apply associtation rule mining

Thank you

Uzma Bibi

Hi Uzma,

this is the description of the GroupBy node taken from the KNIME documentation:

Groups the rows of a table by the unique values in the selected columns. A row is created for each unique value group of the selected column(s). The remaining rows are aggregated by the defined method. The output table therefore contains one row for each existing value combination of the selected group column(s).

I think it pretty much explains what the node does. The full description is within KNIME and online here: https://www.knime.org/files/nodedetails/_manipulation_row_row_transform_GroupBy.html

The best way to learn how this node (and any other node) works is to try it out and play with its configuration parameters. You can use the Table Creator node to quickly input some test data or the CSV Reader node to load in one of the many freely available sample data sets (for example here https://archive.ics.uci.edu/ml/datasets.html), connect a GroupBy node, configure it, look at its output. Change the configuration, look at the output again and so on.

Very soon you will get a very good understanding of what a specific node does and how it works.

Cheers,
Marco.