Intuit, Inc.

Saikat Mukherjee

Karpaga Ganesh Patchirajan

Cem Unsal

Mritunjay Kumar

Esmé Manandise

Anu Sreepathy

Sudhir Agarwal

Anu Singh

Roger Charles Meike

William Taylor Laaser

Per-Kristian G Halvorsen

Conrad De Peuter

Gang Wang

David Alan Hanekamp, Jr

Michael A Artamonov

Kevin Michael Furbish

Michael Radwin

Per-Kristian Halvorsen

Systems and methods for lean parsing are disclosed. An example method is performed by one or more processors of a system and includes retrieving form data including first sentence segments and second sentence segments, determining a first predicate structure for each of the sentence segments based on a set of operators within the first set of sentence segments, identifying known tokens within the second set of sentence segments, each of the known tokens appearing on a list of predetermined tokens, identifying new tokens within the second set of sentence segments, each of the new tokens not on the list, mapping each known and new token to at least one operator, determining a second predicate structure for each sentence segment based on the mapping, and generating a predicate argument structure incorporating the first and second predicate structures, the predicate argument structure ready for mapping to at least one machine executable function.

Lean parsing: a natural language processing system and method for parsing domain-specific languages

Systems and methods for recognizing domain specific named entities are disclosed. An example method may be performed by one or more processors of a text incorporation system and include extracting a number of terms from a text under consideration, identifying, among the number of terms, a set of unmatched terms that do not match any of a plurality of known terms, passing each respective unmatched term to a vectorization module, embedding a vectorized version of each respective unmatched term in a vector space, comparing each vectorized version to known term vectors, passing, to a machine learning model, candidate terms corresponding to known term vectors closest to the vectorized versions, identifying, using the machine learning model, a best candidate term for each respective unmatched term, mapping the best candidate terms to unmatched terms in the text under consideration, and incorporating the text under consideration into the system based on the mappings.

System and method for recognizing domain specific named entities using domain specific word embeddings

A method and system learn functions to be associated with data fields of forms to be incorporated into an electronic document preparation system. The functions are essentially sets of operations required to calculate the data field. The method and system receive form data related to a data field that expects data values resulting from performing specific operations. The method and system utilize machine learning and training set data to generate, test, and evaluate candidate functions to determine acceptable functions.

System and method for automatic learning of functions

A method and system to learn new forms to be incorporated into an electronic document preparation system, or to learn the behavior of existing systems, receive form data related to a new form having a plurality of data fields that expect data values based on specific functions. The method and system gather training set data including previously filled forms having completed data fields corresponding to the data fields of the new form. The method and system include multiple analysis modules that each generate candidate functions for providing data values for the data fields of the new form. The method and system evaluate the candidate functions from each analysis technique and select the candidate functions that are most accurate based on comparisons with the training set data.

System and method for automatically generating calculations for fields in compliance forms

This disclosure describes converting computer-executable predicate-argument structures for a specific field to field-specific predicated-argument structures to improve execution. In some implementations, a method can be performed by one or more processors of a computing device, and can include receiving one or more predicate-argument structures (PASs) associated with taxation-specific text and converting the one or more PASs into one or more tax-specific predicate-argument structures (TPASs). Converting the one or more PASs to one or more TPASs may include one or more of: defining terms in a segment based on a definition of the term from a different segment or line description (including from a different document); reordering nodes, replacing nodes, or removing nodes of a segment (such as based on one or more single segment tree traversal rules); or combining multiple PASs for multiple segments of a single line description based on one or more multiple segment tree traversal rules.

Document text extraction to field-specific computer executable operations

A method may include acquiring, from an initial document having a document type, initial document elements and initial attributes, deriving initial features for the initial document elements using the initial attributes, detecting initial form components using the initial features, clustering the initial form components into initial line objects of an initial structured representation by applying an unsupervised machine learning model to the geometric attributes of the initial document elements, acquiring, from a next document having the document type, next document elements and next attributes describing the next document elements, deriving next features for the next document elements using the next attributes, detecting next form components using the next features, determining that the initial form components and the next form components are different, clustering the next form components into next line objects of a next structured representation, and replacing the initial structured representation with the next structured representation.

Generating structured representations of forms using machine learning

A method and system parses natural language in a unique way, determining important words pertaining to a text corpus of a particular genre, such as tax preparation. Sentences extracted from instructions or forms pertaining to tax preparation, for example are parsed to determine word groups forming various parts of speech, and then are processed to exclude words on an exclusion list and word groups that don't meet predetermined criteria. From the resulting data, synonyms are replaced with a common functional operator and the resulting sentence text is analyzed against predetermined patterns to determine one or more functions to be used in a document preparation system.

Generating a difference between a first and second plurality of lines of text in structured machine-readable format may include determining, by at least one processor, a line of the second plurality of lines that constitutes a best match for a line of the first plurality of lines. The line of the first plurality of lines and its respective best match may be associated with a similarity score. The at least one processor may compare the similarity score to a threshold value. In response to determining that the similarity score is greater than or equal to the threshold value, the at least one processor may compute, the textual difference between the line of the first plurality of lines and its best match. In response to computing the textual difference, the at least one processor may analyze the textual difference to identify a non-meaningful change. In response to identifying a non-meaningful change, the at least one processor may record the textual difference in a delta with a flag indicating the presence of the non-meaningful change.

System and method of generating deltas between documents

A method and system learns functions to be associated with data fields of forms to be incorporated into an electronic document preparation system. The functions are essentially sets of operations required to calculate the data field. The method and system receive form data related to a data field that expects data values resulting from performing specific operations. The method and system utilize machine learning and training set data to generate, test, and evaluate candidate functions to determine acceptable functions.

A natural language processing method and system utilizes a combination of rules-based processes, vector-based processes, and machine learning-based processes to identify the meaning of terms extracted from data management system related text. Once the meaning of the terms has been identified, the method and system can automatically incorporate new forms and text into a data management system.

A method may include acquiring, from a document, document elements and attributes describing the document elements. One or more of the attributes may be geometric attributes describing a placement of the corresponding document element within the document. The method may further include deriving features for the document elements using the attributes, detecting form components using the features, clustering the form components into line objects of a structured representation by applying an unsupervised machine learning model to the geometric attributes of the document elements, and populating a compliance form using the structured representation.

A method involves identifying account data of an entity for a present time period, where the account data includes more than one first data value, creating a comparison group for the entity. The comparison group includes more than one second data value, the account data of the entity includes the second data values, and the second data values originate from a prior time period. The method further involves selecting, from the first data values, a subset of the first data values, selecting, from the second data values, a subset of the second data values, identifying, by accessing a library including anomaly detection methods, an anomalous value within the subset of the first data values by comparing the subset of the first data values with the subset of the second data values, selecting an action in response to identifying the anomalous value within the subset of first data values, and initiating the action.

Detecting anomalous values in small business entity data

A method and system generates sufficient sample data sets for efficiently and accurately testing a new calculation for preparing a portion of an electronic document for users of an electronic document preparation system. The method and system prepares the new calculation and gathers historical user data related to previously prepared electronic documents for a large number of historical users. The method and system generates a representative value for each historical user data, based on the sections of a previous version of electronic document preparation software which were executed for each historical user in preparing electronic documents for the historical users. The method and system groups the historical users based on the hash values which indicates the historical user's behavior in the software. The groups are then sampled by selecting a small number of historical users from each group.

System and method for identifying a subset of total historical users of a data management system to represent a full set of test scenarios based on prehashing of code coverage information

Growing community of inventors

Fremont, CA, United States of America

Saikat Mukherjee