Provalis Research Text Analytics Software
Provalis Research is a leader in Text Analysis software with qualitative, quantitative and mixed methods programs in their solution portfolio.
Text Analytics Software
• PROSUITE: Analysis of Numerical Data, Analysis of Textual Data
• QDA MINER 5: Text Management, Coding Analysis, Multilingual Interface
• WORDSTAT 7: Text Processing, Univariate Keyword Frequency Analysis
• Feature Extraction, Norm Creation and Comparison
• Keyword Retrieval Function
•Contact for Pricing
Small (<50 employees), Medium (50 to 1000 Enterprise (>1000 employees)
•Analysis of Numerical Data
•Analysis of Textual Data
• QDA MINER 5
• Coding Analysis
•Brand Image Analysis
•Voice of the Customer
•Media Framing Analysis
Provalis Research is a leader in Text Analysis software with qualitative, quantitative and mixed methods programs in their solution portfolio. The company is headquarted in Montreal, Canada, and was founded in 1989 and has been developing text analysis programs for more than 20 years. Provalis Research software products are used by more than 2,000 institutions including universities, governments, NGO’s and businesses.Provalis Research text analytics software is a collection of tools that allow users to explore, analyze and relate both structured and unstructured data. The main tools are QDA Miner, WordStat and SimStat.
Provalis Research Text Analytics Software
Provalis Research text analytics main tools are QDA Miner, WordStat and SimStat.
1. QDA Miner
QDA Miner is a computer assisted qualitative data analysis software (CAQDAS) package for manually coding, annotating, retrieving and analyzing small and large collections of documents and images.
QDA Miner provides on screen coding and annotation of texts and images with features offering greater flexibility and ease of use such as code splitting, merging, easy resizing of coded segments, interactive code searching and replacement or virtual grouping.
The tool also provides advanced computer assistance for coding with more than seven text search tools including keyword search, section retrieval, a powerful query-by-example search tool that learns from the user, and a unique cluster extraction and coding tool.
QDA Miner integrates statistical and visualization tools such as clustering, multidimensional scaling, heatmaps, correspondence analysis and sequence analysis. Users don’t need to purchase separate statistical software or transfer data to another application.
QDA Miner also provides advanced Geotagging and Time-tagging tools, which allows one to associate geographic and time coordinates to text segment or graphic area.
WordStat is a text analysis module for QDA Miner. WordStat combines content analysis and text mining methods. WordStat can apply existing categorization dictionaries to a new text corpus. It also may be used in the development and validation of new categorization dictionaries. It may be used for automatic document classification and categorization.
WordStat can analyze a large amount of unstructured data in few seconds. WordStat builds a content analysis dictionary in order to categorize automatically text data and quickly retrieve text segments related to a specific category. To save time, users can reapply in few seconds the same dictionary for a similar project, customize or use an existing dictionary.
WordStat also provides statistical and visualization tools that are easy to interpret such as word frequencies, clustering, correspondence analysis or heatmap.
Text can be transformed to statistical tables and graphics and at anytime, users can drill down to the source documents in order to see what is behind the numbers. WordStat provides a total control over the content analysis process and enough closeness to the data to achieve the perfect balance between text analysis efficiency and precision in the results.
WordStat provides tools such as such as bar charts, pie charts, bubble charts, dendrogram, concept maps, correspondence analysis and more to easily create presentations and write a professional report that includes statistical tables and graphics.
The text data can be analyzed in almost any languages, because the software relies on language independent techniques.There is no need to identify and fix manually spelling mistakes. WordStat can automatically correct them in the documents, and the user will save a lot of time. In order to standardize the writing of similar words and phrases and for obtaining more accurate statistical results, WordStat allows users to substitute any specific word or phrase with another one of your own choice.
SimStat is an advanced statistical analysis software offering a wide range of statistics such as multiple regression analysis, reliability analysis, factor analysis etc.
The interoperability of these tools: QDA Miner, WordStat and SimStat, allows researchers to integrate numerical and textual data into a single project and to seamlessly move back and forth between quantitative and qualitative data analysis.
Provalis Research Text Analytics tools may be used to analyze small or large amount of text data such as: interview or focus group transcripts, legal documents, social media data, customer emails, journal articles, online reviews, open-ended responses.
The product is deployed across a range of users including market researchers, pollsters and business intelligence professionals; crime analysts, fraud detection experts, lawyers, and paralegal professionals; HR managers; political scientists, sociologists and ethnographers; researchers in social sciences, medicine, psychology and communication studies; journalists, historians and research assistants.
1. Data capture / extraction capabilities
The Information extraction tool capabilities includes document, image and database importation in following formats.
Document importation: ANSI and Unicode text files, MS Word, WordPerfect, RTF and HTML, PDF.
Image file importation: JPG, PNG, BMP, WMF.
Data base importation: Excel, Access, Paradox, dBase, SPSS, NVivo, N6, Atlas.ti, Transana, Transcriber, Reference Information System, XML databases, as well as any other database with ODBC drivers
Others: A document conversion wizard can extract records and data from any document files, including XML, Blog entries, Email logs, Lexis-Nexis output, etc.
The Entity extraction capabilities let users apply existing content analysis dictionaries or create their own dictionaries. Clustering of words allows one to extract topics. One can click on a cluster and retrieve all segments associated with this cluster. Other extraction tools include a phrase finder and an “unknown words” extraction tool to easily identify misspelling, proper names and technical terms.
2.Text categorization / clustering capabilities
The tool allows many Text categorization / clustering capabilities. Users can apply existing content analysis dictionaries or create their own dictionaries composed of words, word patterns, phrases as well as proximity rules (such as AND, OR, NOT, NEAR, AFTER, BEFORE) for achieving precise measurement of concepts. Classification of documents is also possible by using Naïve-Bayes or k-nearest neighbor algorithms applied either on words or concepts.
Users can also extract automatically topics by using first order (word co-occurrences) or second order (co-occurrence profiles) hierarchical clustering and multidimensional scaling. This may also be combined with a categorization dictionary in order to cluster concepts. Same tools (hierarchical clustering and multidimensional scaling) may also be used to cluster documents based on words, phrases or content categories.
3.Document summarization capabilities
The frequency page shows the most basic information one can get (frequency of words or content categories) on documents. Clustering of words allows one to identify topics. One can click on a cluster and retrieve all segments associated with this cluster. Users can classify automatically content into categories by using an existing categorization dictionary or creating a new dictionary. Correspondence analysis can help users find information they were not looking for. It help users 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. Same tools (hierarchical clustering and multidimensional scaling) may be used to cluster documents based on words, phrases or content categories.
4. Sentiment analysis capability
Provalis Research provides the tools that would allow users to create sentiment analysis using either a dictionary or a machine learning approach. It also provides a sentiment analysis dictionary with instructions on the best way to customize it.
5. Visualization Features
The visualization features includes 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, and Correspondence plots (2D and 3D).
All charts are customizable. Ability to select and drill down to the text is available for all the above graphics except 2D and 3D MDS plots and 3D correspondence plots.
Filtering of documents may be performed globally using a separate dialog. Several of those graphics allow manual removal and rearrangement (or recomputation) of items.
Pricing Structure /licensing Cost/ Operating Model
Provalis Research charge by concurrent user. There are three categories of prices: Commercial prices, Governmental/ NGO’s prices and Academic prices.
Single user perpetual licenses:
|QDA Miner + WordStat||$4,995||$2,950||$960|
|QDA Miner + WordStat + SimStat||$5,675||$3,350||$1,165|
Single user annual licenses:
|QDA Miner + WordStat||$1,816||$1,072||$349|
|QDA Miner + WordStat + SimStat||$2,063||$1,218||$423|
Provalis Research offer quantity discounts from 5% to 50% when more than one user license are ordered.
Consulting / Services
Provalis Research can provide skilled consultants either on site or through web meetings for a full range of services to help customer quickly become productive and achieve more accurate and effective data processing and analysis.
Provalis Research can provide programming services to fine-tune the analytical tools and develop custom solutions to directly address our customer needs.
Provalis Research also provides free unlimited technical support for all products by either online or by phone. They also offer introductory and advanced training courses on all Provalis Research products. Courses can be adapted to address both academic as well as large and small business needs.
With around 20 years of experience in developing Text Analytics Software, Provalis Research is recognized as a market leader in Text Analytics solutions. The Text Analytics solution portfolio: QDA Miner, WordStat and SimStat are widely acclaimed as easy to use tools for a general business user. Provalis Research also offers multi channel customer support.