WordStat 5.1 New Features

For a more detailed description of all these new features, download the WordStat 5.1 Addendum.

DICTIONARIES PAGE

• New pre-processing module for 4-gram and 5-gram transformations.
• Ability to search the dictionary for specific strings.
• A Show Warning option allows one to obtain a warning dialog box showing compatibility problems of dictionaries (illegal characters, words in both the categorization and exclusion list, phrases starting with an excluded word, etc.).
• Improved dialog box for entering and editing words, phrases and categories (multiple-line edit box).

FREQUENCIES PAGE

• Bar charts and pie charts can now be used to display the frequency distribution of items.
• Ability to copy the frequency table or part of it to the clipboard.
• Search dialog box to quickly find items in large frequency tables.
• Ability to include a column (or variable) containing the total number of words in the document when exporting data.

CROSSTAB PAGE

• Ability to copy a crosstab table or part of it to the clipboard.
• Search dialog box to quickly find items in large frequency tables.

KEYWORD-IN-CONTEXT PAGE

• The KWIC list now supports the display of rules.

FEATURE EXTRACTION PAGE

• Ability to compare frequency or case occurrence between classes of categorical variables.
• Identification of overlaps between extracted phrases.
• Ability to filter table to display phrases containing specific strings or those characteristic of a class of categorical variable.
• The TFxIDF statistic has been added to the table of extracted phrases.
• Ability to select phrases meeting a minimum number of cases criterion.
• Speed optimization of phrase extraction routine (about 2x faster).
• Memory optimization allows phrase extraction to be applied on larger document collections.

CORRESPONDENCE ANALYSIS

• Random noises may be added to point locations in correspondence plots in order to differentiate items plotted on top of each other.

AUTOMATIC DOCUMENT CATEGORIZATION

• The feature selection in classification can now be performed on frequencies and percentages (not just case occurrences).
• It is now possible to classify documents stored in another Simstat or QDA Miner data file.

OTHERS

• The document conversion wizard may now be used to classify documents stored in another Simstat or QDA Miner data file.

New features in version 5.0 can be viewed here.