Dahlia Mosaic Virus Analysis: Statistica Tutorial and Report

This assignment presents a practical tutorial on analyzing Dahlia Mosaic Virus using the TIBCO Statistica software. The tutorial guides the user through the process of opening a dataset, selecting data mining recipes, and configuring various steps, including the use of Random Forest and SVM models. The analysis involves examining 10 cultivars of dahlias to check for changes in virus amounts through DNA microarray analysis. The assignment includes screenshots illustrating key steps in the Statistica workflow, a detailed process description, and a lessons learned section highlighting the student's experience with the software and the challenges encountered. The report concludes with a summary of the findings, including model summaries, lift charts, and C&RT trees, and demonstrates the application of machine learning processes such as support vector machines to determine the relative amount of virus infection in each plant. The tutorial also demonstrates applying v fold cross validation to validate the results.

Running head: Tutorial_KK 1
Dahlia Mosaic Virus: A DNA Microarray Analysis on 10 Cultivars from a Single Source
[Author Name(s), First M. Last, Omit Titles and Degrees]
[Institutional Affiliation(s)]

Tutorial_KK 2
Table of Contents
Dahlia Mosaic Virus: A DNA Microarray Analysis on 10 Cultivars from a Single Source....4
Part 1: Tutorial Outline....................................................................................................................4
Part 2: Process Description..............................................................................................................4
Part 3: Lessons Learned.................................................................................................................22
Conclusion.....................................................................................................................................22
References.....................................................................................................................................23
List of Figures
Figure 1: Opening Dataset...............................................................................................................4
Figure 2: Selection of data mining recipes......................................................................................5
Figure 3: Data mining recipes wizard..............................................................................................5
Figure 4: Choosing Dataset.............................................................................................................6
Figure 5: Data source.......................................................................................................................6
Figure 6: Adding variables...............................................................................................................7
Figure 7: Selected variables.............................................................................................................7
Figure 8: Configure all steps............................................................................................................8
Figure 9: Random Forest and SVM.................................................................................................8
Figure 10: Unchecking Configure all steps.....................................................................................9
Figure 11: Validation of Model........................................................................................................9
Figure 12: Run option....................................................................................................................10
Figure 13: Processing of Run option.............................................................................................10

Tutorial_KK 3
Figure 14: Output of data mining recipe........................................................................................11
Figure 15: Model Summary...........................................................................................................11
Figure 16: Lift Chart......................................................................................................................12
Figure 17: C&RT Tree...................................................................................................................14
Figure 18: Machine learning operation..........................................................................................15
Figure 19: Support vector machines..............................................................................................15
Figure 20: Support vector machines wizard..................................................................................16
Figure 21: Selection of Variables...................................................................................................16
Figure 22: Variable selection.........................................................................................................17
Figure 23: Continue sample’s testing............................................................................................17
Figure 24: Default settings of SVM...............................................................................................18
Figure 25: Kernel’s default settings...............................................................................................18
Figure 26: Applying v fold cross validation..................................................................................19
Figure 27: Machine learning process.............................................................................................19
Figure 28: Output of support vector machines..............................................................................20
Figure 29: Support vector machine................................................................................................20
List of Tables
Table 1: Output of Model Summary..............................................................................................13
Table 2: SVM Model Summary.....................................................................................................14
Table 3: Model summary...............................................................................................................22

Tutorial_KK 4
Dahlia Mosaic Virus: A DNA Microarray Analysis on 10 Cultivars from a Single Source
Part 1: Tutorial Outline
Microarray analysis is performed in this tutorial on the given data of 10 cultivars of
dahlias. It intends to check whether there has been any change in the amount of virus they have
Part 2: Process Description
The data analysis is performed on the Tibco STATISTICA software, so open it. Next, the given
dataset must be opened by following the below steps ("Free TIBCO Spotfire Trial", 2020).
Figure 1: Opening Dataset
Take the provided dataset and begin data mining process where it is required to select data
mining recipes option.

Tutorial_KK 5
Figure 2: Selection of data mining recipes
The following figure shows the opening of the data mining recipes wizard.
Figure 3: Data mining recipes wizard

Tutorial_KK 6
Here, it requires data file to be opened thus select the option open data file and choose the given
dataset.
Figure 4: Choosing Dataset
The following wizard shows the data source (Paul, 2017).
Figure 5: Data source

Tutorial_KK 7
Include the variables to the data mining process by selecting the suitable variables and press Ok.
Figure 6: Adding variables
The following wizard shows the information of the variables selected.
Figure 7: Selected variables

Tutorial_KK 8
Then, as shown in the below figure configure all steps.
Figure 8: Configure all steps
The above step helps to create target variables under it by clicking on the model building tab.
Then, press on Random Forest and SVM option as shown below.
Figure 9: Random Forest and SVM

Tutorial_KK 9
Now, it is required to uncheck Configure all steps option and return to data mining process.
Figure 10: Unchecking Configure all steps
Validate the models created with the help of Next button, and select Run option.
Figure 11: Validation of Model

Tutorial_KK 10
Figure 12: Run option
Figure 13: Processing of Run option
The following figure shows the output of data mining recipe.

Tutorial_KK 11
Figure 14: Output of data mining recipe
Model Summary
Figure 15: Model Summary

Tutorial_KK 12
Table 1: Output of Model Summary
1 2 3 4 5 6
Model
summary
Model ID Name Training error
(%)
Select for
evaluation
Advanced
settings
Annotatio
ns
5 Neural network 68.75 TRUE
4 SVM 68.75 TRUE
3 Boosted trees 81.25 TRUE
2 Random forest 89.58 TRUE
1 C&RT FALSE
Table Step options
Date and
time
2/2/2020 12:12:37
PM
Lift Chart

Tutorial_KK 13
Figure 16: Lift Chart
SVM Model Summary

Tutorial_KK 14
Table 2: SVM Model Summary
C&RT Tree

Tutorial_KK 15
Figure 17: C&RT Tree
Further, Machine Learning operation is conducted with the selection of machine learning option.
Figure 18: Machine learning operation

Tutorial_KK 16
The following figure shows the wizard contains the details of machine learning. Next, select the
option of support vector machines. Further, press OK option.
Figure 19: Support vector machines
As shown in the following figure, where the support vector machines wizard is displayed, opt the
variables.

Tutorial_KK 17
Figure 20: Support vector machines wizard
Figure 21: Selection of Variables
The variable selection is represented in the following figure. Move to support vector machines
with a click on OK button.

Tutorial_KK 18
Figure 22: Variable selection
Later, press on sampling tab and divide data into train. Continue sample’s testing as shown
below.
Figure 23: Continue sample’s testing
Default settings of SVM must be checked as shown below.

Tutorial_KK 19
Figure 24: Default settings of SVM
Kernel’s default settings must be checked as shown below.
Figure 25: Kernel’s default settings

Tutorial_KK 20
Afterwards, see the cross validation tab and view Apply v fold cross validation. Press OK for
moving to the SVM.
Figure 26: Applying v fold cross validation
The following figure represents the machine learning process.
Figure 27: Machine learning process

Tutorial_KK 21
The following figure represents the outcomes of support vector machines.
Figure 28: Output of support vector machines
Figure 29: Support vector machine

Tutorial_KK 22
Table 3: Model summary
Part 3: Lessons Learned
From this tutorial I have effectively had the experience of learning and working on Tibco
software. The aim of this tutorial is completed where the DNA probe analysis is utilized for
determining whether some named cultivars are highly resistant when compared to the rest of the
cultivars with various viruses. However, the other aim is also accomplished which clearly
determines the relative amount of virus infection in every tested plant. This has helped to gain
knowledge of the data mining analysis and necessary processes. I have work on the iteration and
modelling process. I have also learnt to retrieve the outcomes of support vector machines. This

Tutorial_KK 23
process has taught me to use the machine learning process whose results are displayed in this
report.
Conclusion
This report has completed the execution of the necessary results are determined. The first step
performed in this tutorial includes cleaning the dataset of 10 cultivars.
The DNA probe analysis is completed and the relative amount of virus infection in every tested
plant are also determined. The v fold cross validation is applied and the results of support vector
machines are represented.

Tutorial_KK 24
References
Free TIBCO Spotfire Trial. (2020). Retrieved 4 February 2020, from
https://www.tibco.com/spotfire-trial
Paul, J. (2017). Tibco Tutorials for beginners. Retrieved 4 February 2020, from
https://javarevisited.blogspot.com/2011/05/tibco-tutorials-for-beginners.html