Automatic item generation for passage-based assessment

1. A computer implemented method of passage-based automated item generation in a digital examination, the method comprising:

obtaining a first string of text, the first string of text being organized based on a linguistic context;

selecting a first textual term from the first string of text;

identifying a first set of linguistic characteristics corresponding to the first textual term by:

tokenizing the first string of text into separate words;

identifying a part-of-speech for each separate word; and

determining, with a machine learning-based linguistic pattern analyzer, phrase types corresponding to groups of separate words based on matching the groups of separate words to part-of-speech patterns;

obtaining a second textual term exhibiting at least one linguistic characteristic, wherein:

the at least one linguistic characteristic is different from any linguistic characteristic of the first set of linguistic characteristics corresponding to the first textual term,

the at least one linguistic characteristic is chosen from a list including a type of punctuation, a logical connector, a noun, an adverb, a verb, a tense, a prepositional phrase, and an infinitive phrase, and

automatically replacing the first textual term with the second textual term creates a grammatically or linguistically incorrect term in the first string of text that is absent when the first textual term is used in the first string of text, wherein the automatic replacement of the first textual term with the second textual term is not performed by a human;

generating a second string of text by replacing the first textual term in the first string of text with the second textual term, wherein the second textual term includes the at least one linguistic characteristic that is different from any linguistic characteristic of the first set of linguistic characteristics; and

generating a set of examination answers comprising the first textual term, second textual term, and other alternate textual terms;

wherein determining phrase types corresponding to the groups of separate words comprises:

training the machine learning-based linguistic pattern analyzer by:

obtaining training passages;

tokenizing the training passages into corresponding training words;

determining, with the machine learning-based linguistic pattern analyzer, training part-of-speech patterns corresponding to the training passages, and identifying multiple recurring linguistic patterns in the training passages; and

applying the machine learning-based linguistic pattern analyzer to the first string of text.

2. The method of claim 1 , wherein the at least one linguistic characteristic of the second textual term is selected from a same linguistic characteristic category as the first textual term.

3. The method of claim 1 , further comprising selecting the alternate textual terms from a set of linguistic terms exhibiting linguistic characteristics selected from a same linguistic characteristic category as the first textual term.

4. The method of claim 1 , further comprising generating a first linguistic pattern for the first string of text, the generating the first linguistic pattern comprising identifying a linguistic characteristic tag for each respective textual term in the first string of text and storing the linguistic characteristic tag in a data store.

5. The method of claim 4 , wherein the linguistic characteristic tag comprises an order index relative to a location of a respective textual term in the first string of text and a linguistic characteristic category corresponding to the respective textual term.

6. The method of claim 5 , wherein the first linguistic pattern matches a second linguistic pattern corresponding to the second string of text.

7. The method of claim 1 , wherein the second string of text is linguistically incorrect.

8. The method of claim 1 , further comprising displaying the second string of text on a graphical user interface.

9. The method of claim 1 , further comprising displaying the set of examination answers on a graphical user interface and obtaining an input from the graphical user interface, the input indicating one of the set of examination answers.

10. A system for generating items in a digital examination, the system comprising:

a data store, a graphical user interface, a machine learning-based linguistic pattern analyzer, and an item generation logical circuit, wherein the item generation logical circuit comprises a processor and a non-transitory memory with computer executable instructions embedded thereon, the computer executable instructions configured to cause the processor to:

obtain training passages;

tokenize the training passages into corresponding training words;

determine, with the machine learning-based linguistic pattern analyzer, training part-of-speech patterns corresponding to the training passages, and identify multiple recurring linguistic patterns in the training passages;

train the machine learning-based linguistic pattern analyzer based on input from the graphical user interface identifying training phrase types corresponding to groups of training words, training part-of-speech patterns, and indications that the machine learning-based linguistic pattern analyzer correctly determined the training part-of speech patterns;

obtain a first string of text, the first string of text being organized based on a linguistic context;

select a first textual term from the first string of text;

tokenize the first string of text into separate words;

identify a part-of-speech for each separate word;

determine, with the machine learning-based linguistic pattern analyzer, phrase types corresponding to groups of separate words based on matching the groups of separate words to part-of-speech patterns;

obtain a second textual term exhibiting at least one linguistic characteristic, wherein:

the at least one linguistic characteristic is different from any linguistic characteristic of a first set of linguistic characteristics corresponding to the first textual term,

the at least one linguistic characteristic is chosen from a list including a type of punctuation, a logical connector, a noun, an adverb, a verb, a tense, a prepositional phrase, and an infinitive phrase, and

automatically replacing the first textual term with the second textual term creates a grammatically or linguistically incorrect term in the first string of text that is absent when the first textual term is used in the first string of text, wherein the automatic replacement of the first textual term with the second textual term is not performed by a human;

generate a second string of text by replacing the first textual term in the first string of text with the second textual term, wherein the second textual term includes the at least one linguistic characteristic that is different from any linguistic characteristic of the first set of linguistic characteristics; and

generate a set of examination answers comprising the first textual term, second textual term, and other alternate textual terms.

11. The system of claim 10 , wherein the at least one linguistic characteristic of the second textual term is selected from a same linguistic characteristic category as the first textual term.

12. The system of claim 10 , wherein the computer executable instructions are further configured to cause the processor to select the alternate textual terms from a set of linguistic terms exhibiting linguistic characteristics selected from a same linguistic characteristic category as the first textual term.

13. The system of claim 10 , wherein the computer executable instructions are further configured to cause the processor to generate a first linguistic pattern for the first string of text by identifying a linguistic characteristic tag for each respective textual term in the first string of text, and store the linguistic characteristic tag in the data store.

14. The system of claim 13 , wherein the linguistic characteristic tag comprises an order index relative to a location of a respective textual term in the first string of text and a linguistic characteristic category corresponding to the respective textual term.

15. The system of claim 14 , wherein the first linguistic pattern matches a second linguistic pattern corresponding to the second string of text.

16. The system of claim 10 , wherein the second string of text is linguistically incorrect.

17. The system of claim 10 , wherein the computer executable instructions are further configured to cause the processor to display the second string of text on the graphical user interface.

18. The system of claim 10 , wherein the computer executable instructions are further configured to cause the processor to display the set of examination answers on the graphical user interface and obtain a second input from the graphical user interface, the second input indicating one of the set of examination answers.

19. The method of claim 1 , further comprising:

generating the digital examination as a linguistic examination with multiple questions and a set of answers, wherein the set of answers includes at least the first textual term and the second textual term; and

enabling to present the digital examination to an examinee on a second graphical user interface.

20. The method of claim 1 , wherein the incorrect term increases difficulty of the digital examination.