Data Representation
Isalos enables you to manipulate your data using the well-known spreadsheet environment. In the spreadsheets data are represented as tables of instances: in the columns the values of the different variables-features are included, and each row contains an instance-example which is characterized by the values of the different features. The core concept in the operation of Isalos is that each tab acts like a node, where data are imported and exported in a tabular format, following processing and transformation using specific functions.
Table of contents
In practice, each tab consists of an input spreadsheet (left-hand) [1] from which data are imported to the function-node [2] and an output spreadsheet (right-hand) [3] where results are presented. The output spreadsheet of the tab can be imported to the next tab [4] and eventually it is possible to build predictive models through a series of spreadsheets that comprise a workflow [5] of well-defined steps1. The main Isalos functionalities are available through the top ribbon of the user interface: Data Transformation
[6], Analytics
[7], Statistics
[8], and Plot
[9].
Data Input
You can import your data to Isalos from an existing file, an existing spreadsheet within the current project, or by manually entering the data values.
Import from file
To import data into a new tab, right-click on the spreadsheet located on the left-hand side and select Import from file
[1]. Then, choose the directory containing your data file and select the file you wish to upload [2]. Accepted formats include CSV, XLSX, and XLS.
In the pop-up window select if the first row contains column name headers and if the first column contains row IDs by selecting the boxes User Header
and User Row ID
, respectively [3]. In case that an XLSX or a XLS file is imported, select the sheet from which data are imported from the Select Sheet dropdown
list [4]. From the Select Separator
dropdown list [5] select the character string delimiting columns between semicolon (;
) and comma (,
). Click on the Execute
button [6] to complete the data import.
Import from spreadsheet
To import data from an existing spreadsheet, right-click on the spreadsheet located on the left-hand side and select Import from SpreadSheet
[1]. Then, choose from the Select input tab
dropdown list the name of the tab from which data are going to be read and imported [2]. Please note that data cannot be imported from the spreadsheet of the first tab of the workflow.
Manual import
Alternatively, you can directly insert your data by manually typing the values on the left-hand side spreadsheet. It is also possible to copy and paste your data to the input spreadsheet from an external file.
Export spreadsheet data
You can export the processed data and the results at any stage of the analysis workflow by right-clicking on any of the two input and output spreadsheets and selecting Export SpreadSheet
data [1].
In the File Preferences
window select the File Extension
between the XLSX and CSV [2] and check the boxes of User Header
and User Row ID
[3] if you wish to save the column and row names respectively. In case that the XLSX extension is selected, select also if the input or output spreadsheet (depending on your initial selection of spreadsheet) is also going to be exported in the file, by clicking on the Include Input Sheet
(or Include Output Sheet
) [4]. In this case, the input and output spreadsheets are going to be saved as different sheets in your XLSX file [5]. In case that the CSV extension is selected, click on the Select Separator
dropdown menu [6], and select the the character string delimiting columns between semicolon (;
) and comma (,
).
Click on the Execute
button [7] and type the name of the file and select the folder where it is going to be saved [8] and click on the Save
button to complete the export process.
Clear spreadsheet
By right-clicking on any spreadsheet and by selecting Clear SpreadSheet
, its data is removed.
Workflow Development
In Isalos each tab acts like a node, where data are imported (through the left-hand spreadsheet), processed (using one of the available functions), and exported (through the right-hand spreadsheet). In each tab a specific function is applied on the data so, to develop a complete analysis workflow, you should add more tabs and allow data flow between input and output spreadsheets.
Insert new tabs
When you open an empty Isalos project, a single tab named Action
is displayed [1]. You can rename this tab by right clicking on the name of the tab and selecting Rename
[2]. In the presented configuration window, type the new name of the tab [3] and click on OK
button [4]. You can follow this process whenever you wish to rename an existing tab.
By clicking on the +
symbol next to the name of the last tab [5], you can insert a new tab-node to your analysis workflow. In the configuration window, type the name of the tab [6] and click on OK
button [7]. An empty tab is created, ready for data input and analysis.
Right-clicking on any tab and selecting Delete
will remove the tab [8]. This action is irreversible.
Workflow representation
Each tab functions as a node in the data analysis workflow. These nodes are represented in the workflow outline area at the top of the interface [1]. When a new tab is added, a corresponding node is automatically created in the workflow outline [2]. As the analysis progresses and data flows between tabs (see Import from spreadsheet), the nodes are automatically connected, forming a fully integrated workflow [3]. Clicking on a specific node switches the interface to the related tab.
Tips
- You can periodically save the intermediate results of your data analysis or modelling using the
Export SpreadSheet
data option, for further analysis or study of the effects of each workflow step. This is particularly useful for debugging purposes. - To have an overview of the analysis through the workflow outline, as the analysis progresses choose to import your data using the
Import from SpreadSheet
option rather than copying and pasting data between tabs.
References
- Varsou D-D, Tsoumanis A, Papadiamantis AG, Melagraki G, Afantitis A. Isalos Predictive Analytics Platform: Cheminformatics, Nanoinformatics, and Data Mining Applications. Springer International Publishing; 2023. doi.org/10.1007/978-3-031-20730-3_9.
Version History
Introduced in Isalos Analytics Platform v0.1.18
Instructions last updated on May 2024