Data Mining Assignment: Tibco Statistica 13.3 Software Application

This document presents a comprehensive solution to a data mining assignment using Tibco Statistica 13.3. It begins with screenshots and descriptions of key steps, including dataset loading, variable selection, data preparation, and the creation of lift charts and summary spreadsheets. The document details the use of the "Bank Breakdown Data set" within the software, outlining data mining activities such as data preparation, data analytics, and data redundancy. Algorithms like boosted trees, neural networks, and CNRT are applied for statistical operations. The solution highlights the process of data cleansing, validation, and the final presentation of insights through plots and graphs, culminating in a model evaluation report to select the most effective model. The assignment involved the use of the software's features for data mining, statistical analysis, and machine learning techniques, as well as the challenges encountered during the process.

Data Mining Using Tibco Statistica 13.3

1
Table of Contents
Part 1: Screenshots and Descriptions..........................................................................................2
Part 2: Summary.......................................................................................................................... 7
References.................................................................................................................................. 8

2
Part 1: Screenshots and Descriptions
Screenshot 1- Dataset loading in Tibco Statistica 13.3
Loading the dataset into the software is the first stage in the data mining process (Bodhe
& Mankar, 2014). Here we need to select the dataset using data source selection window.
Screenshot 2 – Variable selection
In this activity, we selected Product, Gender, Age, Procedures as a target or categorical
variable. And the date is the input variable.

3
Screenshot 3- Data preparation

4
Data preparation is the refining process. It includes activities like data cleansing,
removing redundant data etc.
Screenshot 4 – Lift Chart
Lift chart for the model is shown in the below-given figure. It shows the effectiveness of
the models ("Call for Papers for Special Issue on Intelligent Data Preparation", 2004). It
performs calculations find the difference in the results with and without the presence of models.

5
Screenshot 5 –Summary Spreadsheet
The above screenshot shows the data summary. At the last stage of the data
preparation process, we can see this. It shows information about the data preparation process.
It shows the various details like variable name present in the dataset, type of the dataset, the
role of the dataset. (Ramkumar, Hariharan & Selvamuthukumaran, 2012) It also shows the
statistical information of the dataset like Mean, Standard deviation, Skewness and Kurtosis etc.
Screenshot 6 – Model Building Report

6
It shows the information about the different models used in the analysis. From the
screenshot, we can know that there are three different models are used in this model. And they
are C&RT, Neural Network, and Boosted Trees ("Tibco Support Portal", 2020).
Screenshot 7 – Model Evaluation report
The above figure shows the results summary of different models. It can be used for the
selection of the most effective model. Because it shows the error rate of the model.

7
Part 2: Summary
Tibco Statistica 13.3 is one of the commonly used statistical analysis systems. This
software system is very flexible than other similar analytics systems. By using this software user
can perform a different set of statistical analytics tasks. This software allows the users to
develop analytics workflows, developing great visualizations of the findings, predictive data
mining and machine learning features. It also provides forecasting, text analytic features etc.
In this project the different data mining activities like data preparation, data analytics,
and data redundancy etc. using Tibco Statistica 13.3 software tool. For that, the given TutorialA
document has been used. The given TutorialA document contains the detailed procedures of
the different data mining operations using Tibco Statistica 13.3 software. In part 1 of the
document the different data mining operations are described and appropriate results
screenshots are attached (Tuchkova & Kondrasheva, 2019). For that analysis, we used the
“Bank Breakdown Data set”. It is an internal sample dataset provided in the Tibco Statistica 13.3
software. It contains information about bank loan processing activities. The various data mining
processes are carried out using this dataset.
Data preparation is the first stage of the data mining process. This process allows the
users to make the corrections on the dataset required for data analysis. This activity includes
the data set downloading, data cleansing, data validation etc. After the successful completion of
the data preparation process, the data analysis task has been initiated. It contains algorithms
like “boosted tree, neural network, and CNRT”. Here these algorithms have been used for
performing the statistical operations in the machine learning process. These algorithms may use

8
for training and testing purposes. And finally, the insights of the dataset are shown as different
plots and graphs.

9
References
Bodhe, V., & Mankar, P. (2014). Preparation of Datasets for Data Mining Analysis Using
Horizontal Aggregation. International Journal Of Engineering Research, 3(7), 446-
448. doi: 10.17950/ijer/v3s7/708
Call for Papers for Special Issue on Intelligent Data Preparation. (2004). IEEE
Transactions On Knowledge And Data Engineering, 16(11), 1456-1456. doi:
10.1109/tkde.2004.67
Ramkumar, T., Hariharan, S., & Selvamuthukumaran, S. (2012). A survey on mining
multiple data sources. Wiley Interdisciplinary Reviews: Data Mining And Knowledge
Discovery, 3(1), 1-11. doi: 10.1002/widm.1077
Tibco Support Portal. (2020). Retrieved 19 January 2020, from
https://support.tibco.com/s/article/How-to-do-stratified-sampling-through-Data-
Miner-Recipes-DMR
Tuchkova, A., & Kondrasheva, P. (2019). The term "data mining". Tasks solved by data
mining methods. SCIENTIFIC DEVELOPMENT TRENDS AND EDUCATION. doi:
10.18411/lj-10-2019-26