Systems and methods for calculating text difficulty

Abstract

Disclosed are systems, methods, and products for language learning that automatically extracts keywords from resources using various natural-language processing product features, which can be combined with custom-designed learning activities to offer a needs-based, adaptive learning methodology. The system may receive resources having text and then determine a text difficulty score that predicts how difficult the resource is for language learners based on any number of factors, including any number of semantic and syntactic features of the text. Training resources labeled with metadata may be used to train a statistical model for determining difficulty scores of newly received text. Resources may be grouped based on difficulty score, and groups of resources language learners' proficiency levels.

What is claimed is: 
     
         1 . A server-implemented method comprising:
 extracting, by a server, text from a plurality of resources of a corpus stored in a database;   identifying, by the server, one or more lexical features in the text of each of the plurality of resources according to a lexical feature bank computer file, wherein the lexical feature bank computer file comprises a listing of one or more features related to language difficulty of the text with respect to reading the text;   assigning, by the server, a value to each lexical feature of the one or more lexical features according to a relative language difficulty that each lexical feature contribute to a complete difficulty associated with the text;   determining, by the server, a text difficulty score for the text of each of the plurality of resources using each value associated with each of the one or more lexical features; and   clustering, by the server, one or more resources of the plurality of resources having a similar difficulty based on the text difficulty score for the text of each of the plurality of resources to generate a lesson.   
     
     
         2 . The server-implemented method of  claim 1 , further comprising:
 outputting, by the server, the text difficulty score for the text of each of the plurality of resources to a user interface of a computing device of a content curator.   
     
     
         3 . The server-implemented method of  claim 1 , wherein the lexical feature bank computer file further comprises the listing of one or more features related to the language difficulty of the text with respect to understanding the text. 
     
     
         4 . The server-implemented method of  claim 3 , wherein the language difficulty of the text with respect to reading or understanding the text is determined in view of a non-native learner of a current language of the text. 
     
     
         5 . The server-implemented method of  claim 1 , wherein the one or more features comprises a list of one or more semantic features. 
     
     
         6 . The server-implemented method of  claim 1 , wherein the one or more features comprises a list of one or more syntactic features. 
     
     
         7 . The server-implemented method of  claim 1 , further comprising:
 assigning, by the server, a weight to each lexical feature of the one or more lexical features in the text of each of the plurality of resources.   
     
     
         8 . The server-implemented method of  claim 7 , further comprising:
 executing, by the server, a logistic regression algorithm to assign the weight to each lexical feature of the one or more lexical features in the text of each of the plurality of resources.   
     
     
         9 . The server-implemented method of  claim 8 , wherein the weight of each lexical feature corresponds to how much difficulty each lexical feature contributes to linguistic difficulty of the text. 
     
     
         10 . The server-implemented method of  claim 9 , further comprising:
 executing, by the server, a statistical model to determine the text difficulty score for the text of each of the plurality of resources using each weight assigned to each lexical feature of the one or more lexical features in the text of each of the plurality of resources.   
     
     
         11 . A system comprising:
 a server comprising a processor and non-transitory machine-readable storage containing instructions that when executed by the processor:
 extract text from a plurality of resources of a corpus stored in a database; 
 identify one or more lexical features in the text of each of the plurality of resources according to a lexical feature bank computer file, wherein the lexical feature bank computer file comprises a listing of one or more features related to language difficulty of the text with respect to reading the text; 
 assign a value to each lexical feature of the one or more lexical features according to a relative language difficulty that each lexical feature contribute to a complete difficulty of the text; 
 determine a text difficulty score for the text of each of the plurality of resources using each value associated with each of the one or more lexical features; and 
 cluster one or more resources of the plurality of resources having a similar difficulty based on the text difficulty score for the text of each of the plurality of resources to generate a lesson. 
   
     
     
         12 . The system of  claim 11 , wherein the server is further configured to:
 output the text difficulty score for the text of each of the plurality of resources to a user interface of a computing device of a content curator.   
     
     
         13 . The system of  claim 11 , wherein the lexical feature bank computer file further comprises the listing of one or more features related to the language difficulty of the text with respect to understanding the text. 
     
     
         14 . The system of  claim 13 , wherein the language difficulty of the text with respect to reading or understanding the text is determined in view of non-native speakers of a current language of the text. 
     
     
         15 . The system of  claim 11 , wherein the one or more features comprises a list of one or more semantic features. 
     
     
         16 . The system of  claim 11 , wherein the one or more features comprises a list of one or more syntactic features. 
     
     
         17 . The system of  claim 11 , wherein the server is further configured to assign a weight to each lexical feature of the one or more lexical features in the text of each of the plurality of resources. 
     
     
         18 . The system of  claim 17 , wherein the server is further configured to:
 execute a logistic regression algorithm to assign the weight to each lexical feature of the one or more lexical features in the text of each of the plurality of resources.   
     
     
         19 . The system of  claim 18 , wherein the weight of each lexical feature corresponds to how much difficulty each lexical feature contributes to linguistic difficulty of the text. 
     
     
         20 . The system of  claim 19 , wherein the server is further configured to:
 execute a statistical model to determine the text difficulty score for the text of each of the plurality of resources using each weight assigned to each lexical feature of the one or more lexical features in the text of each of the plurality of resources.