Department of Language Modelling
NCN grant

Readability – Measuring the degree of readability of nonliterary Polish texts

The project determined lexical and morphosyntactic features influencing the readability of Polish texts and built a tool for assessing it.

Lead Investigator: Włodzimierz Gruszczyński

Grant Number
2011/03/B/HS2/05799
Start Date
16 August 2012
End Date
15 March 2015

Project factsheet

English name: Measuring the degree of readability of nonliterary Polish texts
Polish name: Mierzenie stopnia zrozumiałości polskich tekstów użytkowych (pozaliterackich)
Project type: A National Science Centre grant 2011/03/B/HS2/05799
Duration: 16 August 2012 ‒ 15 March 2015
Project Web page: http://wiki.nlp.ipipan.waw.pl/readability (in Polish, authorization required)
Principal investigator: Włodzimierz Gruszczyński

Project summary

The aim of the project is to determine the features of a text in Polish that influence its readability. We will be occupied mainly with surface features, such as lexical and morphosyntactic features. We assume that many such features, quite easy to determine using automatic text analysis, influence text readability.

The tangible effect of the project will be a monograph, presenting the results of the research, and a computer tool for assessing the degree of readability and other stylistic-statistic indicators, such as: vocabulary richness, text subjectivity, egotism and maybe others. The computer tool will allow to determine the degree of readability of a given text and to determine those text features which are responsible for its incomprehensibility (e.g. it will point out difficult words or too complicated sentences). The indicators computed by the program will be put on a scale elaborated in the project. The scale will be developed using psycholinguistic methods.