The new features of SPAD 7.3

Graphical visualisation of data

The interactive graphical visualisation lets you access many graphics after selecting data in one or more diagrams. It provides a wide range of graphics:

• Graphics on categorical variables (sectors, bar charts, bar charts for grouped variables),
• Graphics on continuous variables (standard bar charts or bar charts with unequal classes, box plots, various indicators),
• X-Y graphics (gallery and scatter plots),
• Specific graphics (outlines, chessboards, radars, concentration curves, Lift curves and Roc curves).

The graphics can be displayed in a global view (mosaic) or in separated views (slides). The user can easily define its own models, its graphic plans and obviously export the structured graphics into documents (Excel, PDF, RTF, HTML, PPT) or in the form of images (PNG, JPEG, GIF, SVG, WMF).

Automatic graphic generators

 All the graphics are available as execution results of a method. They are updated after any execution or data modification. Two methods are available: the "Automatic graphic generator" that lets you access all the types of graphics and the "Lift curve generator".

The "Lift curve generator" is particularly interesting to compare the relevance and quality of the predictive models.

Advanced cross tables (in the Data Management Expert module)

The "Advanced cross tables" method lets you create tables by defining a row structure, a column structure and if necessary calculation variables on which mathematical functions are applied.

The created tables can be analyzed by statistical methods as contingency tables, or can be heterogeneous and complex tables that give "Reporting" kind of results..

Interface with R

SPAD offers a ‘’Plug-in’’ that lets you use the R(*) language functionalities in the SPAD environment. This new function extends considerably the range of the SPAD methods. It is provided with all the R functions required to access the SPAD data and recover all the R results (graphics and new data), which constitutes a real gateway between both products.

(*) The R language is a GNU project similar to S. It is a programming language used for data processing and statistical analysis.

Data Management

Among the new functions: the multi-tables stacking, the horizontalization of data and new recoding functions (Standardize, rank creation from a continuous variable). Moreover, the methods that required a significant execution time such as the join have been sensibly optimized.

Lemmatization algorithms for Text Mining

The "lemmatization" tool of SPAD now contains two algorithms used to search for equivalent "lexical forms": the "Phonex" and the "Soundex". These new tools let you gain time when processing plain and not standardized text corpus.

The statistical tests

This method lets you compute standard statistical tests: parametric and non-parametric tests, normality tests, tests on contingency tables, calculation of confidence intervals on mean or variance.

The interface lets you ask for and calculate hundreds of tests with a few clicks. The conclusions are displayed in "plain text" to facilitate the reading for non-expert users

The scheduler (optional module)

This new option lets you define and launch executions in batch mode. This new tool is particularly useful for long and/or recurrent executions. It lets you execute successively several diagrams from different projects.

Haut de page ↑