Intuit, Inc.

Meng Chen

Lei Pei

Nhung Ho

Bei Huang

Linxia Liao

Ngoc Nhung Thi Ho

Jing Zhang

Ngoc Nhung Ho

Nivethitha Kumar

Neo Yuchen

Shanshan Tuo

Divya Beeram

Wan Yu Zhang

Lulu Cheng

Tomer Tal

Wenting Cai

Kavita Sundar

Crystal Meng

Zachary Grove Jennings

Yu-Chung Hsiao

Wei W Wang

Juan Liu

Giovanni Seni

Heather Simpson

Zhicheng Xue

Xiangling Meng

Yueyue Gu

Certain embodiments of the present disclosure provide techniques training a user detection model to identify a user of a software application based on voice recognition. The method generally includes receiving a data set including a plurality of voice interactions with users of a software application. For each respective recording in the data set, a spectrogram representation is generated based on the respective recording. A plurality of voice recognition models are trained. Each of the plurality of voice recognition models is trained based on the spectrogram representation for each of the plurality of voice recordings in the data set. The plurality of voice recognition models are deployed to an interactive voice response system.

Method and system for user voice identification using ensembled deep learning algorithms

Certain aspects of the present disclosure provide techniques for generating a user interface to prompt users of a software application to perform an action in the software application. The method generally includes generating historical transaction time gap data for transactions in the account. A probability distribution is generated based on the historical time gap data. The probability distribution represents a probability that a transaction related to the account has been performed after an elapsed time from a previous transaction. A probability that an unrecorded transaction exists for an account based on the probability distribution and a time difference between a most recent transaction and a current time. The probability that an unrecorded transaction exists is determined to exceed a threshold probability, and a user interface is generated and displayed to a user of the software application including a prompt for the user to enter new transactions for the account.

Method and system for generating user interfaces to prompt users to perform an activity in a software application based on transaction time analysis

The present disclosure provides a composite machine learning system for a transaction labeling service. A transaction labeling service receives at least one descriptive string describing a transaction associated with a user. The service identifies a preliminary grouping from a generalized scheme. The service extracts a set of N-grams from the descriptive string and converts the N-grams and the preliminary grouping into a set of features. A machine learning model determines a label from a labeling scheme for the transaction based on the features. User input related to the label includes an accuracy indicator and a reliability indicator. If the reliability indicator satisfies a reliability condition, a set of training data for the machine learning model is updated based on the descriptive string and the label. The machine learning model is then trained against the updated set of training data.

Composite machine learning system for label prediction and training data collection

Systems of the present disclosure generate database queries for financial information requested in a natural-language form. A natural-language processing (NLP) financial aggregator receives a request for financial information and extracts NLP features of the request, including keywords. The NLP financial aggregator determines a type of the request based on the features and creates a query in a database-query language for the financial information based on the type and on the features. The NLP financial aggregator submits the query to a database where the financial information is stored. The software then receives the financial information from the database and sends the information to the user.

System to convert natural-language financial questions into database queries

Certain aspects of the present disclosure provide techniques for vectorization of transactions including: receiving electronic transaction information of one or more transactions of a user; for each transaction of the one or more transactions: segmenting the electronic transaction information of the transaction into one or more transaction words; generating a second transaction description related to the transaction; and identifying a category of the transaction; generating, based on the corresponding identified categories of the one or more transactions, a set of transaction history data of the user; providing the set of transaction history data of the user as an input to a machine learned model trained to output a set of word embedding vectors; determining, based on an output of the machine learned model comprising a set of word embedding vectors, a set of similar merchants; and providing the set of similar merchants for display to the user.

Vectorization of transactions

A method and system identify assistance offerings that are likely to be relevant to users of a data management system. The method and system utilize a multivariate random forest regression machine learning process to train an assistance offerings recommendation model to recommend relevant assistance offerings to users of the data management system. The multivariate random forest regression machine learning process replaces zero values in the training set data with negative numbers to increase the accuracy of the machine learning process.

Method and system for recommending assistance offerings

A method that predicts business income from user transaction data. A multinomial classifier is trained, using a vector of features from data related to a historical transaction and a label associated with the historical transaction, to generate a probability that the historical transaction belongs to a specific classification with respect to income. Data related to a new transaction is split into a set of unigrams. A new vector of features is generated from the data related to the new transaction. The new vector includes a set of values that correspond and are assigned to the set of unigrams. A classification with respect to income is determined for the new transaction by applying the multinomial classifier to the new vector. The new transaction is labeled with the classification. One or more fields of a form that is maintained by an online service is populated using the classification.

Method for predicting business income from user transaction data

Systems and methods that may be used to provide guidance and or tag suggestions to a user of an electronic accounting system and or service that overcome the shortcomings associated with user-defined tags.

System and method for providing global tag suggestions based on user information and transaction data

A method is disclosed. The method includes: obtaining a help request associated with a user operating an application and a click stream of the user within the application; generating a feature data structure based on the help request and the click stream; generating, by applying the feature data structure to a machine learning model, a topic data structure including a plurality of scores corresponding to a plurality of topics; obtaining a plurality of topic distributions corresponding to a plurality of articles; identifying, by applying the topic data structure to the plurality of topic distributions, a subset of the plurality of articles for the user; and displaying, in response to the help request, a graphical user interface (GUI) including a plurality of links to the subset of the plurality of articles.

Machine learning suggested articles for a user

A method includes obtaining data related to a plurality of historical transactions, where each historical transaction is associated with a label based on a click stream created by the first user, generating a vector of features from the data related to each historical transaction, training, using the vectors and labels, a multinomial classifier to generate a probability that a specific transaction belongs to a specific classification with respect to income, obtaining data related to a new transaction from a financial stream for a second financial account of a second user of the financial service, generating a new vector of features from the data related to the new transaction, determining a classification with respect to income for the new transaction, and presenting the classification to the second user for review in a view of a graphical user interface.

The present disclosure provides a composite machine-learning system for a transaction labeling service. A transaction labeling service receives at least one descriptive string describing a transaction associated with a user. The service identifies a preliminary grouping from a generalized scheme. The service extracts a set of N-grams from the descriptive string and converts the N-grams and the preliminary grouping into a set of features. A machine-learning model determines a label from a labeling scheme for the transaction based on the features. User input related to the label includes an accuracy indicator and a reliability indicator. If the reliability indicator satisfies a reliability condition, a set of training data for the machine-learning model is updated based on the descriptive string and the label. The machine-learning model is then trained against the updated set of training data.

Composite machine-learning system for label prediction and training data collection

Method and system for recommending assistance offerings using multivariate random forest leveraging negative samples

The invention relates to a method for generating a topic name for accounts grouped under a topic. The method includes obtaining account names associated with the accounts, generating a plurality of n-grams from the account names, and for each n-gram, obtaining a quality score based on relevance and meaningfulness of the n-gram. The relevance is determined using at least one relevance score that reflects how representative the n-gram is for the plurality of n-grams, and the meaningfulness is determined based on whether the n-gram exists in a knowledge base. The method further includes assigning one or more of the n-grams to the topic as the topic name, based on the quality score generated for each of the n-grams.

Growing community of inventors

Mountain View, CA, United States of America

Meng Chen