Solo QuickStart

Table of Contents | Previous | Next

Solo Quick Start

Welcome to the Solo Quick Start page. This page is designed to help you get started fast with Solo by explaining, at a high level, the basic steps for carrying out a Principal Component Analysis (PCA), which is one of the most commonly carried out analyses in Solo. In addition, the basic steps for a PCA also touch on the basic steps for almost all of the other analysis methods that are available in Solo. Each step contains one or more links to the appropriate pages on this Wiki that provide detailed information about the step. Some important points to note about this Quick Start page are the following:

The steps detailed here are designed to complement the steps that are defined in "PCA in Wine Data," which is an online tutorial that walks you through the step by step basics of carrying out a PCA in Solo. The tutorial is located at [http://www.eigenvector.com/eigenguide.php]. You can use the procedure outlined on this page to follow along with the tutorial. At any time, you can pause the online tutorial, and click on a link next to a step to go to the indicated chapter to learn more about the step.

The tutorial and the quick start steps are based on using the wine DataSet, which is demo data that is loaded during the installation of Solo. You can repeat the steps listed here using this DataSet, or you can use another smaller DataSet, for example, arch.

Note: If you running Solo on a Windows OS, the demo data is loaded in C:\Program Files\EVRI\Solo\Demo_Data. Contact Eigenvector for assistance in locating the demo data for other OSs and/or for selecting a different DataSet.

Note: Before you carry out the quick start steps, it might be helpful to have a detailed overview of the windows and the modeling and analysis phases in Solo. See "Solo Windows" and "Analysis Phases".

Quick Start steps for an analysis in Solo

1.	Launch Solo. See "Launching Solo."

After you launch Solo, the Workspace Browser opens automatically. The Workspace Browser is your starting interface for Solo. The interface provides quick access to all of the data analysis tools. For information about the Workspace Browser its layout and its options-see:

"Workspace Browser."

"Workspace Browser Preferences."

2.	Import data into the Workspace Browser.

The selected data files are loaded into the Workspace Browser. After you import the data, different icons are displayed in the Workspace Browser for the different data types. You can save these data items to a workspace, and you can manipulate this data in the browser before you analyze it. See:

"Importing Data into the Workspace Browser."

"Icons in the Workspace Browser."

3.	Load the imported data into the PCA tool for analysis by dragging the data icon onto the Decompose (PCA) shortcut icon.

The Drag and Drop method is only one of the variety of methods that are available for opening an Analysis window and loading data. All of the available methods are discussed in detail in the in the appropriate sections this Wiki. For a detailed discussion of the Analysis window, see "Analysis Window."

4.	Optionally, to view the raw data in a spreadsheet layout prior to analysis, and if necessary, edit the data prior to analysis, open the data in the DataSet Editor window. See "DataSet Editor Window."

Information that you glean in this view can help you understand the patterns that you will see later when generating plots and other visual aids of sample relationships and variable relationships.

5.	Optionally, plot the raw data for review prior to analysis. See "Plot Controls Window."

6.	Select the appropriate preprocessing methods.

Data preprocessing describes any type of processing procedures that are performed on raw data to prepare it for another processing procedure and ultimately, analysis. Preprocessing linearizes the relationships among the variables in your DataSet and removes extraneous sources of variation that are of no interest to the analysis. A variety of preprocessing methods are available in Solo. See "Preprocessing Methods."

7.	Generate the calibration (initial) model.

The Calibration phase consists of model building and exploratory analysis. In this phase, which affects only the Calibration side of the Status pane, you identify any patterns or trends in the data, and any other information that you consider relevant, for example, any relationships that might exist between the x data and the y data, and use this information to build a model. See "Building the Model in the Calibration Phase."

8.	Create plots and other visual aids that assist you in examining and refining the model by excluding certain samples and/or variables to enhance the model performance.

To generate plots and other visual aids that show the relationship among the samples in your data, see "Plotting Scores and Statistical Values for a Calibration Model."

To generate plots and other visual aids that show the relationship among the variables in your data, see "Plotting Loads and Variable Statistics for a Calibration Model."

9.	Apply the model to new data to verify that the model will provide acceptable results for the analysis of validation data, which is data with known physical and/or chemical characteristics. See "Applying the Model in the Test and Validation Phase."

10.	Save the model to the Workspace Browser or to a file and use it at a later date ((File > Save Model on the Analysis window main menu), or export the model to a file or a predictor. (File > Export Model on the Analysis window main menu.)

Note: See Exporting_Models