Systems and methods for generating distractors in language learning
Disclosed are systems, methods, and products for language learning software that automatically extracts text from resources using various natural-language processing features, which can be combined with custom-designed learning activities to offer a needs-based, adaptive learning methodology. The system may receive a text, extract keywords pedagogically valuable to non-native language learning, assign a difficulty score to the text using various linguistic attributes of the text, generate a list of potential distractors for each keyword related to a resource to implement in learning activities. Distractors may be of various types, which are dynamically selected from a distractor store depending on a learning activity chosen to meet a learner's needs. Distractors may vary in difficulty, and may be dynamically selected based on a learner's overall proficiency or based on a learner's abilities in specific language skills.
1. A computer-implemented method comprising:
automatically selecting, by a computer, from a keyword store a set of one or more target words stored in the keyword store, each respective target word having a word difficulty score based upon one or more skill scores of a learner-record stored in a learner database,
wherein the keyword store is configured to store metadata associated with each respective target word, and
wherein the metadata associated with each respective target word in the keyword word store indicates the word difficulty score of the respective target word;
generating, by the computer, a set of one or more syntactic distractors comprising one or more words having a same root word as the target word and having a grammatical difference; and
generating, by the computer, at least one more distractor from a group of:
a set of one or more semantic distractors comprising one or more words having a definition that is related to a target word,
a set of one or more orthographic distractors comprising one or more words of the plurality of words having an edit distance of each respective word satisfying an edit distance amount setting, wherein the edit distance of the word is a number of changes required to the word to be identical to the target word, and wherein the edit distance amount setting determines the number of changes to the word, and,
a set of phonetic distractors comprising one or more homophones of the target word, based upon the one or more skill scores of the learner-record.
2. The computer-implemented method of claim 1 , further comprising:
identifying, by the computer, in a dictionary source a plurality of words having an edit distance based upon the target word, wherein the edit distance of a word is an amount of letter-insertions and letter-deletions required for the word to be the target word.
3. The computer-implemented method of claim 2 , wherein the edit distance of each respective word in the set of orthographic distractors satisfies an edit distance amount setting, and wherein the edit distance amount setting limits the edit distance of the respective word based upon the one or more skill scores of the learner-record.
4. The computer-implemented method of claim 1 , further comprising:
extracting, by the computer, a plurality of target words from one or more resources; and
for each respective target word:
determining, by the computer, the word difficulty score based upon one or more characteristics of the target word;
generating, by the computer, the metadata of the target word, the metadata indicating the word difficulty score and a resource containing the target word; and
storing, by the computer, the target word and the metadata associated with the target word into the keyword store.
5. The computer-implemented method of claim 4 , further comprising:
receiving, by the computer, the resource of the one or more resources containing the respective target word of the a set of one or more target words from a computing device;
determining, by the computer, a resource difficulty score associated with the resource based upon one or more characteristics of the resource; and
storing, by the computer, into a resource store configured to store the one or more resources, at least a portion of the resource and metadata indicating the resource difficulty score.
6. The computer-implemented method of claim 1 , further comprising:
determining, by the computer, the definition of the target word based on context used within a resource from which the target word originates.
7. The computer-implemented method of claim 1 , further comprising:
identifying, by a computer, one or more synonyms of the target word in a dictionary data source, wherein the set of semantic distractors comprises at least one word having a synonymous definition related to the target word.
8. The computer-implemented method of claim 1 , further comprising:
identifying, by a computer, in a dictionary source, one or more antonyms of the target word, wherein the set of semantic distractors comprises at least one word having an antonymous definition related to the target word.
9. The computer-implemented method of claim 1 , further comprising:
identifying, by the computer, in a dictionary data source, the one or more homophones of the target word having a comparatively small edit distance amount identified in the dictionary source.
10. The computer-implemented method of claim 9 , further comprising:
excluding, by the computer, from the set of phonetic distractors a homonym identified in the dictionary source.
11. A system comprising:
a keyword store configured to store a plurality of keywords extracted from one or more resources and metadata associated with each respective keyword, the metadata associated with each respective keyword indicating a word difficulty score of the respective keyword;
a computer comprising a processor and non-transitory machine-readable storage instructing the processor, wherein the computer is configured to:
automatically select from the keyword store a set of one or more keywords based upon the word difficulty score of each respective keyword and one or more skill scores of a learner-record stored in a learner database;
generate a set of one or more identified orthographic distractors comprising one or more words of the plurality of words having the edit distance of each respective word satisfying an edit distance amount setting, wherein the edit distance of the word is a number of changes required to the word to be identical to the target word, and wherein the edit distance amount setting determines the number of changes to the word; and
generate a set of one or more syntactic distractors comprising a word having a same root word as the keyword and that is a grammatical variant of the keyword.
12. The system of claim 11 , wherein the edit distance amount setting limits the edit distance of the respective word relative to the keyword based upon the one or more skills scores of the learner-record.
13. The system of claim 11 , wherein the computer is configured to:
generate a set of one or more identified phonetic distractors comprising a homophone of the keyword.
14. The system of claim 11 , wherein the computer is further configured to:
generate a set of one or more semantic distractors comprising a synonym and an antonym of the keyword.
15. The system of claim 14 , wherein the computer is further configured to:
identify a context of the keyword according to metadata associated with the keyword identifying content of the resource, wherein the semantic distractors are identified based on the context of the keyword in the resource.
16. The system of claim 11 , wherein the computer is further configured to:
determine the edit distance according to one or more learner attributes from a learner profile, wherein the edit distance amount has a difficulty level comparable to an ability level of the learner for a language skill.
17. The system of claim 11 , wherein the computer is further configured to:
store a set of distractors in a distractor store database.
18. The system of claim 11 , wherein the computer is further configured to:
output a set of distractors of a distractor type to a user interface of a computing device associated with a content curator.
19. The system of claim 18 , wherein the computer is further configured to:
amend a set of distractors of a according to a request for an amendment received from the user interface of the content curator.
20. The system of claim 11 , wherein the computer is further configured to:
receive, from a software module, a request for a set of distractors of the distractor type and associated with a corpus of a new activity; and
output, to the software module, the set of distractors requested.