|
 |
NovoSpark Visualizer is an advanced, powerful and feature rich visualization program that enables qualitative analysis of multi-dimensional data on a graphical image. The product comes with a variety of image viewing options that helps you get the most representative view of your data.
With NovoSpark Visualizer you will be able to visually compare individual observations and entire datasets, find data anomalies, identify data clusters and do many other things you used to do in your day-to-day data analysis, but much quicker and more efficiently.
Visualizer uses unique data visualization method to represent each multidimensional record as a colored curve, so the entire dataset or multiple datasets of any dimensionality can be visualized on a single image, with the same-colored curves representing records from the same dataset. The little ticks on the white-colored axis mark location of individual curves, while red triangular markers immediately identify the curves that represent "abnormal" records and may require special attention of a researcher.
Visualizer renders multi-dimensional data as a graphical image allowing you to focus on its visual characteristics rather than the tables of raw data. Our visualization technology guarantees that the resulting image preserves all information about the original data, so you can safely rely on the results inferred from the visual analysis of your data samples. The application supports all standard operations with images including rotation, moving, stretching, as well as using various image viewing options like projection views and "noise" filters.
The more similar the curves are on the image, the closer are the data records that they represent. Visualizer allows switching between different image views to study shapes of the curves in more detail and visually identify data records that stand out of the group. Clicking on a curve immediately highlights its record in the data grid allowing to quickly sort out any "suspects" that may exist in the orignal dataset.
Visualizer is ideal for studying the trends and changes in behavior of a dynamic process. The top view of a dataset representing such a process clearly shows turbulences in process behavior, and clicking on suspect areas can highlight the actual observations that cause the anomalies. The "tip of the day" window provides useful tips on how to improve appearance of the image and get the most information from it about the loaded data.
Sorting by different columns or choosing geometrical (Euclidian) or statistical (Mahalanobis) order for selected data variables arranges the curves along the white axis according to the value of the selected sort order. This allows to quickly identify clusters of records and segregate them into separate datasets.
Visualizer comes with a set of helper windows that show the most relevant information about your image and refer to the application features that can be used to improve its look and feel.
To shorten the learning curve and help a researcher start using the product right away, Visualizer presents a floating window with context-sensitive help showing the most relevant information about the image and providing reference to advanced application features that can be used to improve its look and feel.
The application allows to import tabular data from an external text file or database. A variety of data formats is natively supported including delimiter-separated, CSV, fixed-width text files, Microsoft Access, SQL Server and ODBC databases. The application provides an intuitive interface for selecting a data source and previewing the results of import.
Intuitive and easy-to-use data import interface enables loading of existing data from CSV-, delimiter-separated or fixed-width text files. Along with the various data import options the application shows a file preview panel to ensure that the right file is selected for import.
The application supports data import from the most popular types of databases and ODBC sources. The database preview panel shows contents of the target table to ensure that the right data source is selected for import.
Visualizer tries to automatically figure out the structure of the original data source and presents it to the user for review after successful data import. Each selected column is highlighted in the preview panel to let the user review its contents and change column properties before loading the data into the project.
You can also input all your data manually if the external data source does not exist or not available.
Visualizer offers built-in support for automatic handling of missing values in imported data.
The application supports all kinds of data editing giving an opportunity to "play" with the data and see how the changes affect its visual representation. This way not only do you have the flexibility of building any data sample within the application, but the image of this sample is automatically redrawn in sync with your edits, showing how the data would look like in the various "what if..." scenarios.
The researcher can always change the order or characteristics of the original data columns to see how that would affect the image look. Pure re-structuring of data layout may highlight new aspects in its visual representation, or help reduce dimesionality of the orignal data.
Contents of data grids can be changed on the fly giving the possibility to experiment with the original data and look at how the changes affect its visual representation. The records can be freely moved between datasets in the current project or copy/pasted to or from another data editor like Microsoft Excel.
You can load and view data from several datasets. The application allows picking a different color for each dataset, or render them all with a predefined or custom color palette. By moving records from one dataset to another you can visually sort them and form data clusters.
The application provides basic statistical information about each dataset. The records that lie outside of the configured statistical bounds are highlighted in red prompting for a more detailed investigation of the corresponding observations.
The application provides basic statistical information about each loaded dataset to be used for in-depth quantitative analysis of data structure.
The floating window showing normalized or standardized data helps seeing transformed values for heterogeneous data variables or datasets with different distribution characteristics.
The application is very flexible about where its visual elements can reside on the screen. You are free to move them around, resize or hide away to get the layout of the application window that is most convenient for you.
The intuitive user-friendly interface makes it easy to work with the data and interpret visualization results. The explanatory messages and tips, as well as shortcuts to the main application functions will shorten the learning curve and help you start using the product right away.
Visualizer allows switching between the most commonly used objectives of data analysis to ensure optimal configuration of application settings and image views for each task. All data and image settings can be saved into a project file and loaded later or transferred to another researcher for follow-up.
| Copyright (C) 2010 DEK Software International® (800) DEK-5595 |