LATIC is a free and open source desktop application that allows you to analyze and count text and item characteristics in English, French, German and Spanish texts or items. The Stanford CoreNLP 4.4.0 is used to tag parts of speech and allows LATIC to provide very accurate results. You can find our evaluation in the documentation.
- Analysis at the word level
- Parts of speech (e.g. adjectives, interjections, nouns, ...)
- Word length (characters, syllables)
- Analysis at the sentence level
- Number of sentences
- Sentence length (characters with or without spaces, syllables, words)
- Analysis at the text level
- Readability indices (e.g. Flesch, LIX)
- Lexical diversity/type-token ratio
- Word count
- Syllable count
A detailed description of the text and item characteristics and instructions on how to use LATIC are provided in the documentation.
The Standford CoreNLP supports many other languages (e.g. Arabic, Italian, Hungarian), and we'd love to add them to LATIC. However, this requires the support of people who speak one of these languages at a very good level. If you'd like to implement one of the remaining languages with us, please let us know!
Of course, we'd also be happy if you'd like to help us improve LATIC in other ways. Feel free to create issues for feature requests, bugs, or other improvements.
To compile and package the project use
mvn clean compile package and run the file LATIC-1.2.2.jar using java -jar shade/LATIC-1.2.2.jar
Note that you have to launch the class AppLauncher to stat the application in your IDE.