Intuit, Inc.

Verint Systems, Inc.

Yair Horesh

Yehezkel Shraga Resheff

Alexander Zhicharevich

Shimon Shahar

Oren Sar Shalom

Aviv Ben Arie

Itay Margolin

Roni Romano

Adi Shalev

Daniel Ben David

Shlomi Medalion

Sheer Dangoor

Elik Sror

Noah Eyal Altman

Jeremie Dreyfuss

Galia Zacay

Ido Meir Mintz

Elhanan Mishraky

Saar Carmi

Omer Wosner

Shiomi Medalion

Aleksandr Kim

Liron Hayman

Sigalit Bechler

Natalie Bar Eliyahu

Tzvika Barenholz

Miriam Hanna Manevitz

Talia Tron

Noga Noff

Rotem Rozenblum

Gilaad Dital

Omer Ziv

Yaron Sheffer

Margarita Vald

Rami Cohen

Nirmala Ranganathan

Noa Haas

Yaakov Tayeb

Yonatan Ben-Simhon

Or Basson

Yaron Moshe

Yerucham Meir Berkowitz

Omer Zalmanson

Elad Shmidov

Uri Segal

Hadas Baumer

Daniel Vaisman

Yehezkal Shraga Resheff

Nir Keret

Yagil Ovadia

Alexsandr Kim

Natalie Grace De Shetler

Aviv Ben-arie

Nitzan Gado

Osnat Haj Yahia

Kfir Aharon

Hagai Fine

Yonantan Ben-Simhon

Meital Iulia Bojan

Shir Meir Lador

Hadar Kedem

Aspects of the present disclosure provide techniques for detecting latent bias in machine learning models. Embodiments include receiving a data set comprising features of a plurality of individuals. Embodiments include receiving identifying information for each individual of the plurality of individuals. Embodiments include predicting, for each respective individual of the plurality of individuals, a probability that the respective individual belongs to a given class based on the identifying information for the given individual. Embodiments include providing, as inputs to a machine learning model, the features of the plurality of individuals from the data set. Embodiments include receiving outputs from the machine learning model in response to the inputs. Embodiments include determining whether the machine learning model is biased against the given class based on the outputs and the probability that each respective individual of the plurality of individuals belongs to the given class.

Machine learning model bias detection

A system for vendor deduplication. The system creates embeddings of entity names and creates an initial entity graph comprising entities whose embeddings are related. The initial entity graph includes nodes representing the entity names linked together by edge weights indicating their similarity. The system adjusts the edge weights between the entity names according to transactional data related to the entities to create a final entity graph and merges the entity names as a common entity based on the adjusted edge weights in the final entity graph.

Two-fold vendor de-duplication via node embeddings

Systems and methods are disclosed for training a text classification model based on long text and known semantics as training data. With a text classification model limited as to the amount of text that may be input at one time, long text that is greater than the limit may be segmented into smaller segments that are less than the limit (such as into sentences). Each segment of the long text is compared with sample segments with known impact of specific semantics to associate the long text segments with the specific semantics. To compare the long text segments with sample segments, an embedding model generates an embedding from each of the segments so that the embeddings may be compared. With the long text segments associated with specific semantics, the long text segments and the associated semantics are used as training data to train a text classification model.

Training of a text classification model from long text and based on text semantics

Certain aspects of the disclosure provide a method for detecting data collection errors by processing error data with a plurality of regression models to generate a plurality of predicted error rates over a plurality of time intervals. The method includes determining an error mode by applying a set of policy rules optimized for determining the error mode to the plurality of predicted error rates.

Real-time remote system shutdown prediction

A system includes a categorization application including a personal categorization model for execution on a user device, the personal categorization model configured to assign a personal assigned category and a personal assigned categorization confidence value to transaction record. The system also includes a global categorization model executing on a central server. The global categorization model is configured to receive the transaction record from the user device, and process the transaction record through the global categorization model to obtain a global assigned category and a global assigned categorization confidence value. The system is configured to select a transaction category from the personal assigned category and the global assigned category using the personal assigned categorization confidence value and the global assigned categorization confidence value, and categorize the transaction with the transaction category.

Multi-device, multi-model categorization system

A method including extracting data from disparate data sources. The data includes data pairs including a corresponding data point and a corresponding time associated with the corresponding data point. The method also includes extracting insights from the data at least by identifying a trend in the data pairs. The method also includes forming a model vector including the insights and an additional attribute to the insights. The additional attribute characterizes the insights. The additional attribute includes at least user feedback including a user ranking of a ranked subset of the insights from a user. The method also includes inputting the model vector into a trained insight machine learning model to obtain a predicted ranking of the insights. The method also includes selecting, based on the predicted user ranking, a pre-determined number of insights to form predicted relevant insights. The method also includes reporting the predicted relevant insights.

Computer prediction of relevant data from multiple disparate sources

This disclosure relates to systems and methods for identifying fraudulent email addresses associated with an electronic payment service. In some implementations, a computing device receives an email with a prefix having a number of characters and characterized by a prefix length indicative of the number of characters in the prefix. The computing device identifies each of a number of bigrams is identified within the prefix, and determines a row and column distance for each bigram between two consecutive characters of the bigram as positioned on a keyboard. The computing device calculates a Euclidean distance between the two consecutive characters of the bigram based on the row and column distances, and determines a normalized distance based on the prefix length and an average of the Euclidean distances calculated for the number of bigrams in the prefix. The normalized distance is compared with a value to classify the email as suspicious or as not suspicious.

Detecting fraud by calculating email address prefix mean keyboard distances using machine learning optimization

Aspects of the present disclosure provide techniques for detecting sensitive information. Embodiments include determining a first likelihood that a record contains at least a given type of sensitive information using a first detection technique that involves providing one or more inputs to a machine learning model based on the record and receiving the first likelihood as an output from the machine learning model based on the one or more inputs. Embodiments include determining a second likelihood that the record contains at least the given type of sensitive information using a second detection technique comprising a search of the record. Embodiments include applying a policy to determine whether the record contains sensitive information based on the first likelihood and the second likelihood. Embodiments include performing one or more actions based on whether the record contains sensitive information.

Detecting sensitive information in records using context and decoys

A method may including obtaining a voice transcript corpus and a chat transcript corpus, extracting voice transcript sentences from the voice transcript corpus and chat transcript sentences from the chat transcript corpus, encoding, by a series of neural network layers, the voice transcript sentences to generate voice sentence vectors, encoding, by the series of neural network layers, the chat transcript sentences to generate chat sentence vectors, determining, for each voice sentence vector, a matching chat sentence vector to obtain matching voice-chat vector pairs, and adding, to a parallel corpus, matching voice-chat sentence pairs using the matching voice-chat vector pairs. Each of the matching voice-chat sentence pairs may include a voice transcript sentence and a matching chat transcript sentence. The method may further include training a disfluency remover model using the parallel corpus.

Disfluency removal using machine learning

A method includes processing a set of query texts to identify a set of expressions, where each expression references a set of columns of datetime data in a datastore. The method also includes training a statistical model to determine a distribution of the datetime data for each column that was identified. The method further includes processing the set of expressions to generate a directed graph including more than one nodes and a plurality of edges, where each node represents one of the columns or a transformation applied by one of the expressions to one of the columns. The method additionally includes generating a weight for edges of the directed graph according to a distribution of the datetime data in the columns and a usage index of a corresponding expression.

Tables time zone adjuster

Certain aspects of the present disclosure provide techniques for generating a metric, include receiving a rule defining one or more text strings; determining a set of transactions based on a user attribute; determining a first subset of transactions; determining a second subset of transactions; generating a first categorical distribution based on each transaction of the first subset of transactions being associated with a transaction description containing at least one text string of the one or more text strings; calculating a first unity metric based on the first categorical distribution; generating a second categorical distribution based on each transaction of the second subset of transactions being associated with a transaction description that does not contain a text string of the one or more text strings; calculating a second unity metric based on the second categorical distribution; determining a reliability metric for the rule; and providing the reliability metric.

Automated categorization of data by generating unity and reliability metrics

A method implements efficient counterfactual search. The method includes receiving a request corresponding to an input vector, processing the input vector with a model to generate an output vector that does not correspond to a selected class, and processing the input vector using a component, of a plurality of components, to generate a counterfactual vector to the selected class. The plurality of components includes a number of dimensions that is less than a number of features of the input vector. The method further includes processing the counterfactual vector to generate a recommendation and presenting the recommendation.

Efficient counterfactual search

Described herein are example implementations for generating a perturbation seed for the perturbation of electronic data. A system obtains a plurality of datapoints (with one or more statistics calculated from the plurality of datapoints to be perturbed based on a perturbation seed). The system calculates one or more metrics from the plurality of datapoints. The system also generates, for each of the one or more metrics, a rounded metric by rounding the metric. The system further generates the perturbation seed. Generating the perturbation seed includes hashing the one or more rounded metrics. Rounding a metric may be to a defined place value (such as the second most significant place value), and a binary output of hashing the one or more rounded metrics may be converted to a number. The system may perturb one or more statistics based on the perturbation seed and output the one or more perturbed statistics.

Seed generation for electronic data perturbation

An entry validation system executed by a processor, may compare an entry into a user interface (UI) field with at least one range of valid entries. The at least one range of valid entries may be generated by a machine learning (ML) system by clustering known valid data using a clustering algorithm producing a lowest number of clusters that absorbs all of the known valid data within the clusters. In response to the comparing, the entry validation system may provide a determination of validity or invalidity to an application displaying the UI field. The determination of validity may cause the entry to be processed by the application, and the determination of invalidity may cause the application to reject the entry.

Automatic entry validation using density based clustering

Knowledge about a user is used to determine whether one or more messages received by the user are malicious. The knowledge about the user may be based on the user's financial history such as transaction records. Particularly, a classifier model is trained on a supervised approach using a dataset containing, for example, a categorization of incoming messages (e.g., password change message), the user's aggregated transaction records, message attributes, user attributes, and corresponding classification labels. After the training, the classifier model is deployed to determine whether an incoming message is malicious.

Malicious message classification using machine learning models

Growing community of inventors

Kfar-Saba, Israel

Yair Horesh