Wordstat&QDAMiner

NOU! A aparut WordStat 2023!

Recent s-a lansat WordStat 2023 care face un pas semnificativ înainte în aplicarea dezambiguizării sensului cuvântului la modelele de topică. Funcția noastră unică de îmbogățire a subiectelor a suferit îmbunătățiri majore și am introdus câteva capacități noi de modelare a subiectelor pentru a ajuta utilizatorii să obțină informații și mai bune din datele lor. În plus, am implementat numeroase optimizări ale vitezei care fac software-ul mai receptiv și mai ușor de utilizat. Versiunea 2023.1 a introdus, de asemenea, importul și procesarea dosarelor financiare, integrarea Power BI, Gephi și NetDraw și palete de culori personalizabile pentru diagrame.

1. Improved Topic Enrichment / Îmbunătățirea subiectului
WordStat adaugă acum fraze mai relevante la subiectele extrase, oferind în același timp sugestii îmbunătățite pentru fraze suplimentare. În plus, acum se mândrește cu o mai mare acuratețe în identificarea expresiilor fals-pozitive, sau a excepțiilor, care pot fi încorporate în modelul subiectului pentru a ajuta la dezambiguizarea cuvintelor asociate cu contexte care nu au legătură cu subiectele extrase.

2. Topic Modeling Word Cloud / Modelarea subiectului în Word Cloud
Panoul de comparație din partea dreaptă a tabelului modelului de subiecte prezintă acum un nor de cuvinte nou adăugat care descrie vizual importanța relativă a cuvintelor de top din subiectul selectat. Acest nor de cuvinte poate fi personalizat, copiat în clipboard sau salvat pe disc în formate grafice standard, cum ar fi BMP, PNG sau JPEG.

3. New integrated text retrieval feature / Nouă funcție integrată de recuperare a textului
Un nou panou Sample Text convenabil din partea dreaptă a grilei de subiecte poate fi activat pentru a afișa automat propoziții sau paragrafe care se potrivesc cu subiectul selectat. Aceste segmente de text sunt prezentate în ordinea descrescătoare a relevanței, cu cuvintele subiectului afișate cu caractere aldine, facilitând înțelegerea esenței fiecărui subiect și identificarea exemplelor cheie care pot fi folosite pentru a-l ilustra. Acest instrument puternic oferă utilizatorilor o înțelegere mai profundă a datelor lor și facilitează o comunicare mai eficientă a constatărilor lor.

4. Improved Top Enrichment Speed / Viteza de îmbogățire maximă îmbunătățită
Datorită eforturilor semnificative de optimizare, procesul de îmbogățire a subiectelor a fost accelerat dramatic, rezultând câștiguri de performanță de până la 10 până la 20 de ori mai rapide decât versiunile anterioare.

5. Instantanious Phrase Extraction / Extragerea instantanee a frazei
Folosind puterea procesării multicore, extragerea frazelor este acum integrată perfect cu procesarea principală a textului, permițând utilizatorilor să acceseze rezultate aproape instantaneu. De exemplu, pe un set de date de peste 50.000 de recenzii ale clienților, extragerea celor 5000 de fraze cele mai frecvente poate fi finalizată acum în doar 0,4 secunde, comparativ cu cele 14 secunde cerute de versiunea anterioară.

6. Importation of Financial Filings / Importul dosarelor financiare standard
O nouă rutină de import permite utilizatorilor să importe anumite secțiuni ale dosarelor financiare standard și să le stocheze separat sau să le îmbine în documente unice. Rutina de extragere recunoaște automat numele companiei, perioada de timp (trimestru și an) și le stochează ca variabile pentru o analiză ușoară.

7. Export Text Analysis Results to Power BI / Exportați rezultatele analizei text în Power BI
WordStat oferă acum o integrare perfectă cu Microsoft Power BI, permițând utilizatorilor să exporte rezultatele analizei de text și metadatele în Power BI Desktop pentru tablouri de bord și rapoarte interactive. Prin exportarea rezultatelor analizei de text și a metadatelor în Power BI Desktop, utilizatorii pot crea vizualizări convingătoare, pot obține informații mai profunde din datele lor și pot împărtăși cu ușurință descoperirile lor altora.

8. Push Co-occurrence Data to Gephi or NetDraw / Mutați datele concomitente către Gephi sau NetDraw
Cu noua opțiune disponibilă de pe pagina Dendrogram, utilizatorii pot acum exporta date de co-apariție, împreună cu informații suplimentare, cum ar fi frecvența și numărul clusterului, în software-ul de analiză a rețelelor sociale precum Gephi și NetDraw. Aceste instrumente oferă vizualizări puternice care ajută utilizatorii să identifice modele și relații în datele lor. Gephi oferă algoritmi de aspect și funcții interactive pentru explorarea în timp real, în timp ce NetDraw oferă opțiuni de vizualizare pentru graficele de rețea.

9. Custom Chart Palettes / Palete de diagrame personalizate
WordStat 2023 introduce o nouă caracteristică care permite utilizatorilor să creeze palete de culori personalizate. Această caracteristică oferă un control mai mare asupra culorilor utilizate pentru diagrame, nori de cuvinte, grupare și alte vizualizări, permițând utilizatorilor să-și personalizeze rezultatele pentru a se potrivi nevoilor lor specifice.

Va stam la dispozitie cu cotatii de pret prompte si avatajoase!

WORDSTAT – SOFTWARE PENTRU ANALIZA TEXT

WordStat is a flexible and easy-to-use text analysis software – whether you need text mining tools for fast extraction of themes and trends, or careful and precise measurement with state-of-the-art quantitative content analysis tools. WordStat‘s seamless integration with SimStat – our statistical data analysis tool – QDA Miner – our qualitative data analysis software – and Stata – the comprensive statistical software from StataCorp, gives you unprecedented flexibility for analyzing text and relating its content to structured information, including numerical and categorical data.

What it is used for?

WordStat can be used by anyone who needs to quickly extract and analyze information from large amounts of documents. Our content analysis and text mining software is used for:

• Content analysis of open-ended responses, interview or focus group transcripts
• Business intelligence and competitive web sites analysis
• Information extraction and knowledge discovery from incident reports, customer complaints
• Content analysis of news coverage or scientific literature
• Automatic tagging and classification of documents
• Fraud detection, authorship attribution, patent analysis
• Taxonomy development and validation

WORDSTAT FOR STATA – ANALIZA DE CONTINUT SI TEXT MINING PENTRU STATA

Stata is a complete, integrated statistical software package created by StataCorp LP (www.stata.com). It provides a wide range of statistical analysis, data management, and graphics. Released in June 2013, version 13 added many new features, including a long string data type allowing one to store along with numerical and categorical data, documents up to 2 billion characters. One could thus create a statistical database with journal abstracts, news transcripts, patents, incident reports, customer feedbacks, interviews and so on.

WordStat for Stata was created to allow Stata users running under Windows, to apply text analytics techniques on any string variables stored in a Stata data file. WordStat combines natural language processing, content analysis and statistical techniques to quickly extract topics, patterns and relationships in large amount of text. It can process millions of words in seconds and compare extracted themes across any other numerical, categorical or date variables in the Stata file.

WordStat for Stata este un produs al Provalis research dedicat utilizatorilor de Stata. Toate informatiile legate de WordStat for Stata le gasiti aici.

QDA MINER – ANALIZA CALITATIVA A DATELOR

NEW in QDA Miner:

1. New Grid view mode for coding short responses

While appropriate for coding long documents, the standard document/case centric view of QDA Miner was less suited for coding short text responses such as response to open-ended questions or short comments. Now, QDA Miner 6 introduces a new grid view mode that provides a convenient and very efficient way to code this kind of text data. It is useful for everyone coding any type of open-ended comments, including surveys, employee comments, customer comments and allows one to quickly identify trends in a survey or major support issues that need to be addressed. It includes features such as:

• Drag-and-drop coding and annotation.
• Filtering of responses using text search expressions with Boolean operators
• Filtering of responses based on the number of codes (uncoded, coded, more than n codes, etc.) as well as on the presence or absence of specific codes.
• Sorting of rows either alphabetically, on text length, number of codes, or case number.
• Displays the number and percentage of coded responses,
• Computation of word clouds and word frequency analysis on text currently displayed in the grid.

2. Quotation Matrices

The quotation matrix allows you to create a large grid containing all coded text segments and/or comments where each cell represents the intersection of a specific code with either a specific case or a value of a categorical or numerical variable (age group, gender, source, etc.). Such a joint display provides a compact view of coded material ideal for reviewing work done by coders. More easily identify patterns, creating dense summaries of results, etc. This matrix may be created from the new RETRIEVAL | QUOTATION MATRIX command to obtain a codes x cases quotation matrix or from the ANALYSIS | CODING BY VARIABLES command for displaying coded materials by all values of a variable. It supports the following features:

• Displays either all comment types or specific ones based on subject, speaker etc.
• Text in each cell can be edited in with a rich text editor (font style, size and color, paragraph formatting, etc.)
• Multiple memos can be attached to individual cells.
• Rows and columns may be transposed.
• The matrix can be exported to disk in various formats, including Excel, CSV, TSV and a new PGRD format allowing one to review and edit the table outside of QDA Miner using a free grid viewer/editor.

3. Enhanced annotation feature

It is now possible to attach up to six types of comments to a single code mark. Annotations may serve different purposes such as formulating hypothesis, communicating concerns with team members, summarizing, etc. You are no longer restricted to a single comment type. The removal of such limitation and the introduction of the quotation matrix feature (see above) offers new possibilities for generating condense view of summaries, concerns, hypotheses, etc. It gives you much more flexibility on how to instruct, explain codes, pose, and answer questions.

4. Word Frequency Analysis and Word Cloud

Interactive word clouds and word frequency tables can now be obtained on any document variable or on results of retrieval operations (text, coding, section or keyword retrieval) as well as for a single document or for text displayed in the new grid view. One may tailor the word cloud (font, color, shape, etc.), customize stop words lists and perform text searches from it or from the associated word frequency table.

5. Importation of Nexis UNI and Factiva Files

It is now possible to import news transcripts from the LexisNexis and Factiva output files. After selecting one or multiple .DOCX or RTF files obtained from those services, QDA Miner will extract and store in separate variables the title and body of the news transcript, its source, the publication date, and other relevant information. Such a feature should prove useful for reputation management, brand management, crisis communication, media framing analysis, comparative media studies, etc.

6. Improved Importation of Excel, CSV and TSV files.

When importing files from Excel, CSV or TSV files a new wizard dialog box will allow you to select variables, rename them, import variable description, and perform batch data type conversions This gives you greater flexibility to set up your analysis, make it more precise and start it more quickly, saving time and resources.

7. Deviation Table

The CODING BY VARIABLES feature now offers the possibility to produce a deviation table that allows one to obtain a list of codes most or least characteristic of different values of an independent variable as compared to other classes of this variable.

8. Export Results to Tableau Software

One can now export results to Tableau Software allowing one to use its advanced interactive data visualization tools. This feature is available from the CODING FREQUENCY and the CODING BY VARIABLES dialog boxes.

9. Numerical Transformation

A new numerical transformation dialog box allows you to compute numerical variables from other variables with up to 50 transformation functions including trigonometric, statistical, random number functions. Conditional transformation can also be performed using an IF-THEN-ELSE logical structure.

10. Binning

A binning feature can now be used to transform continuous values into a smaller number of distinct categories. It may be used to reduce the effect of numerical outliers, abnormal distributions, or convert a continuous numerical variable into an ordinal one. It is especially useful for creating graphic displays of comparisons when the number of distinct values in the numerical variable is too high.

11. Support of Missing Values

You can now associate to numerical, categorical, and short string variables up to three values that will be treated as missing data.

12. Silhouette plot

A new silhouette plot feature has been added to the hierarchical cluster analysis, allowing one to assess the quality of the cluster solution and identify potential misclassified items.

13. Date transformation

Date and date and time variables can now be used to create other categorical or numerical variables such as months, days or weekdays, months, years, etc.

14. Improved code filtering feature.

The code filtering feature may now be used to filter cases based on the presence, the absence of specific codes or combinations of codes.

15. Donut, Radar, 100% Stacked Bar and Area Charts.

A donut chart can now be used to display relative codes or class frequencies (CODING FREQUENCY and VARIABLE STATISTICS dialog boxes). The charting feature of the CODING BY VARIABLES dialog box also adds the possibility to create a radar chart, a 100% stacked bar chart as well as two types of stacked area charts.

16. Ordering of series in comparison charts.

The relative position of a series of comparisons charts created from the CODING BY VARIABLES dialog box may now be manually adjusted, allowing you to achieve more appealing or revealing visualizations.

17. Color Coding of items in Correspondence Plot

Color gradients may now be used to represent the position of specific words or variable classes on the third (depth) dimension or 2D as well as 3D correspondence plot. Up to four colors may be chosen to create those gradients.

18. Improved Bubble Chart

It is now possible to transpose rows and columns of bubble charts and finely adjust the size of the bubbles.

19. Link Analysis Buffer

A link analysis buffer allows one to move back to previous link diagrams and then forward.

20. New Table Format and Table Editor

A new proprietary table format (*.pgrd) has been added to the exportation of tables to disk, allowing one to easily edit and annotate tables produced by QDA Miner. A free standalone table viewer may also be downloaded from our web site, allowing anyone to view, edit and annotate saved tables.

21. Numerous Additional Improvements

Several new options and interface improvements have been made to existing dialog boxes (code color selection, graphic options, etc.), management and analysis features.

Who uses QDA Miner?

QDA Miner qualitative data analysis software can be used by anyone who needs to code text or pictures, annotate, search, explore and extract information from small or large collections of documents and images, including:

• Researchers in social sciences, medicine, and psychology
• Sociologists, political scientists and ethnographers
• Business intelligence analysts, market researchers, pollsters, and CRM professionals
• Crime analysts, fraud detection experts, lawyers, and paralegal professionals
• Journalists, historians and research assistants
• Document management specialists and librarians