# US 2017/0256178 A1 · Systems and methods for calculating text difficulty > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2017/0256178 A1 - Document ID: US20170256178A1 - Status: Published application - Kind: A1 - Application number: 15/603,296 - Filed: 2017-05-23 - Published: 2017-09-07 - Earliest claimed: 2013-02-15 - Assignees: Voxy, Inc. - Applicants: Voxy, Inc. - Inventors: Katharine NIELSON; Kasey KIRKHAM; Na'im TYSON; Andrew BREEN - Web page: https://patentgrep.com/patent/US20170256178A1 ## CPC - G09B 19/06: Foreign languages (with audible presentation of material to be studied G09B5/04) - A61B 5/162: {Testing reaction times} - A61B 5/165: {Evaluating the state of mind, e.g. depression, anxiety} - A61B 5/374: Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves - A61B 5/411: {Detecting or monitoring allergy or intolerance reactions to an allergenic agent or substance} - A61B 5/486: {Biofeedback (using electroencephalography [EEG] A61B5/375)} - G06F 40/205: Parsing - G06F 40/253: Grammatical analysis; Style critique - G06F 40/263: Language identification - G09B 5/00: Electrically-operated educational appliances (working with questions and answers G09B7/00; simulators G09B9/00; advertising or displaying in general G09F) - G09B 5/02: with visual presentation of the material to be studied, e.g. using film strip - G09B 7/08: characterised by modifying the teaching programme in response to a wrong answer, e.g. repeating the question or supplying further information - A61B 5/0002: {Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network (endoradiosondes A61B5/07)} - A61B 5/1124: {Determining motor skills} - A61B 5/378: Visual stimuli - A61B 5/4064: {Evaluating the brain (for intracranial pressure A61B5/031; for cerebral blood gases A61B5/14553; using EEG A61B5/369)} - A61B 5/4076: {Diagnosing or monitoring particular conditions of the nervous system} - A61B 5/4088: {Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia} - A61B 5/4842: {Monitoring progression or stage of a disease} - G09B 7/00: Electrically-operated teaching apparatus or devices working with questions and answers (mechanically operated G09B3/00; computing arrangements G06F) - G09B 19/04: Speaking (with audible presentation of the material to be studied G09B5/04) - G09B: EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS (devices for psychotechnics or for testing reaction times A61B5/16; games, sports, amusements A63; projectors, projector screens G03B) - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## 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 may correspond to language learners' proficiency levels. ## Claims ### Claim 1 A computer-implemented method comprising: extracting, by a computer, text from one or more resources of a corpus stored in a database, the corpus comprising a plurality of resources, each resource containing text; identifying, by the computer, a set of lexical features in the text of a resource of the corpus according to a lexical feature bank file indicating one or more semantic features and one or more syntactic features, wherein the set of lexical features comprises at least one semantic feature in the bank file and at least one syntactic feature in the bank file; storing, by the computer, into a database configured to store the corpus comprising the plurality of resources and metadata associated with each resource, metadata containing a first value associated with each of the lexical features in the set of lexical features according to the bank file, wherein the bank file indicates each value associated with each respective lexical feature; determining, by the computer, a text difficulty score for the text using each of the first values associated with each of the features in the set of lexical features; outputting, by the computer, the text difficulty score to a user interface of a computing device of a content curator; transmitting, by the computer, a set of values comprising the first values of each feature in the set of lexical features to the computing device of the content curator; and updating, by the computer, the text difficulty score of the text according to a score adjustment received from the user interface of the content curator. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, further comprising: updating, by the computer, the set of values based on one or more amendments received from the user interface of the content curator. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, further comprising: storing, by the computer, into the database storing the plurality of resources, metadata containing the text difficulty score associated with the text of each resource. ### Claim 4 (depends on 3) The computer-implemented method of claim 3, further comprising: generating, by the computer, a portion of one or more lessons associated with a learner profile using the text of a resource selected from the database based upon metadata indicating one or more skill scores of the learner profile stored in a profile database and the metadata containing the text difficulty score of each of the resources of the corpus in the database. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein the resource is one or more portions of a document resource containing the text. ### Claim 6 The computer-implemented method of 1, wherein the resource contains and audio track, and wherein the text is a transcript of the audio track. ### Claim 7 The computer-implemented method of 1, further comprising: receiving, by the computer, the resource from a user interface of a computing device; and parsing, by the computer, the text into one or more portions of text, wherein each portion of text is compared against the lexical feature bank. ### Claim 8 The computer-implemented method of 7, further comprising: receiving, by the computer, the resource from a resource store storing one or more resources. ### Claim 9 The computer-implemented method of 8, further comprising: receiving, by the computer, one or more training resources containing text, wherein the one or more semantic and syntactic features of the text are labeled with metadata indicating the features and an associated first weighted value associated with each of the features; and generating, by the computer, the lexical feature bank comprising each of the semantic and syntactic features labeled in the one or more training resources and each of the associated first weighted values. ### Claim 10 The computer-implemented method of 9, further comprising: assigning, by the computer, a second value to each of the features in the set of lexical features, wherein the computer determines the text difficulty score for the text using each of the first values and each of the second values associated with each of the lexical features. ### Claim 11 The computer-implemented method of 1, further comprising: comparing, by the computer, the text difficulty score associated with the resource against one or more stored text difficulty scores associated with one or more resources stored in the resource store; and determining, by the computer, a skill level associated with the resource according to the text difficulty score associated with the resource relative to the one or more stored text difficulty scores. ### Claim 12 A system comprising: a database comprising non-transitory machine-readable storage media configured to store a corpus comprising one or more resources, and metadata associated with each of the one or more resources in the corpus; a learner database comprising a non-transitory machine-readable storage media configured to store one or more learner profiles; and a computer comprising a processor and a non-transitory machine-readable storage containing a text difficulty calculator module executed by the processor, wherein the computer is configured to: identify in text of a resource of the corpus a set of lexical features according to a lexical feature bank file, the set of lexical features including one or more semantic features and one or more syntactic features; store, into the database storing the resource of the corpus, metadata associated with each resource, the metadata containing a value associated with each of the identified semantic and syntactic features in the text according to the bank file, and metadata containing a statistical model associated with one or more resources of the corpus; determine a text difficulty score associated with the resource using each of the values assigned to the identified semantic and syntactic features; determine a text difficulty score of a new resource containing one or more lexical features, the text difficulty score of the new resource based upon the statistical model associated with the one or more resources of the corpus and the one or more lexical features of the new resource; output the text difficulty score to a computing device of a content curator; and adjust the text difficulty score according to an input received from the computing device of the content curator. ### Claim 13 (depends on 12) The system of claim 12, wherein the computer is further configured to: select from the database storing the plurality training resources, the new resource based upon one or more skill scores of a learner profile in the learner profile database and the text difficulty score in metadata of the new resource; and ### Claim 14 (depends on 13) The system of claim 13, wherein the computer is further configured to: generate one or more portions of one or more language lessons associated with the learner profile using text of the new resource selected from the database. ### Claim 15 (depends on 12) The system of claim 12, wherein the computer is further configured to: assign a weight to each of the assigned values; and determine a weighted value for each of the identified syntactic and semantic features, wherein the text difficulty score is determined using the weighted values. ### Claim 16 (depends on 12) The system of claim 12, wherein the computer is further configured to: group the resource with a set of one or more resources having a comparable text difficulty score; and categorize resources into one or more sets of resources based on a difficulty category. ### Claim 17 A computer-implemented method comprising: extracting, by a computer, one or more linguistic features having a weighted value from a plurality of training resources containing text, wherein each linguistic feature of the text of each training resource is associated with a metadata label indicating the weighted value for each respective feature and containing a text difficulty score of the text; determining, by the computer, a vector value associated with each training resource based on each of the weighted values of each of the extracted one or more linguistic features; generating, by the computer, from metadata associated with each of the training resources stored in a database, a statistical model using the vector values associated with each training resource, wherein the statistical model represents a correlation between a set of features selected for extraction, a set of weighted values assigned to the set of features selected for extraction, and a set of text difficulty scores associated with the training resources; and in response to the computer receiving from a second computer a new resource to include in the plurality of training resources stored in the database: extracting, by the computer, one or more linguistic features having a weighted value from the new resource; determining, by the computer, a vector value for the new resource based upon the set of extracted linguistic features; and generating, by the computer, metadata containing a text difficulty score for the new resource based upon the vector value for the new resource and the statistical model. ### Claim 18 (depends on 17) The computer-implemented of claim 17, further comprising: storing, by the computer, the plurality of training resources and the metadata into a database configured to store the plurality of training resources and the metadata. ### Claim 19 (depends on 17) The computer-implemented of claim 17, further comprising: selecting, by the computer, from the database storing the plurality training resources, the new resource based upon one or more skill scores of a learner profile in a learner profile database and the text difficulty score in the metadata of the new resource; and ### Claim 20 (depends on 19) The computer-implemented of claim 19, further comprising: generating, by the computer, one or more portions of one or more language lessons associated with the learner profile using the text of the new resource.