WordStat
WordStat is a content analysisand text mining add-on module of QDA Miner[1].It was first released in 1998 after being developed by Normand Peladeau from Provalis Research. The latest version 6 was released in 2009.
The software is mainly used for business intelligence and competitive analysis of web sites, sentiment analysis, content analysis of open-ended questions, theme extraction from social media data, etc.
Here a some features of WordStat 6[2]
- Categorization of content using user defined dictionaries.
- Classification of documents using Naïve-Bayes or knearest neighbor algorithms applied either or words or concepts.
- Automatic topic extraction using first order (word co-occurrences)or second order (co-occurrence profiles) hierarchical clustering and multidimensional scaling.
- Correspondence analysis in order to identify words or concepts (or content categories) associated with any categorical meta-data associated with documents. It may also
be applied on ordinal or numerical data and dates to identify temporal trends.
- Ability to relate unstructured text with any structured data such as dates, numbers or categorical data for identifying temporal trends or differences between subgroups or for assessing relationship with ratings or other kind of categorical or numerical data. Tools in this area includes crosstabulation, computation of correlation, chisquare, Anova, and several nonparametric measures of association, correspondence analysis, heatmap with dual clustering, etc.
- Visualization tools to visualize and interpret text analysis results:
- Dendrogram with optional bar chart - 2D and 3D Multidimensional scaling - Proxmity plot - Heatmap (with dual clustering) - Bubble chart - Bar chart, pie chart, line chart, word clouds - Correspondence plots (2D and 3D)