Abstract: A non-transitory computer-readable recording medium stores a model generation program for causing a computer to execute a process including: acquiring results output from a first machine learning model in response to input of a first plurality of pieces of data to the first machine learning model; selecting a second plurality of pieces of data from the first plurality of pieces of data, based on the results; and generating a second machine learning model by executing machine learning of the first machine learning model, by using the second plurality of pieces of data as input, with some parameters being fixed among a plurality of parameters included in the first machine learning model.

Abstract: Disclosed is an approach for performing auto-classification of documents. A machine learning framework is provided to analyze the document, where labels associated with certain documents can be propagated to other documents.

Abstract: A machine learning model is accessed that is configured to use one or more parameters to process images to generate labels. The machine learning model is executed to transform at least part of each of at least one digital pathology image into a plurality of predicted labels; and generate a confidence metric for each of the plurality of predicted labels. An interface is availed that depicts the at least part of the at least one digital pathology image and that differentially represents predicted labels based on corresponding confidence metrics. In response to availing of the interface, label input is received that confirms, rejects, or replaces at least one of the plurality of predicted labels. The one or more parameters of the machine learning model are updated based on the label input.

Abstract: Feature engineering, for example, in automated machine learning, can include receiving streaming data representing at least one attribute detected by a sensor over time. Long term point statistics associated with the streaming data can be computed. The streaming data can be quantized into intervals of time windows and short term point statistics based on the intervals can be computed. The long term point statistics and the short term point statistics can be normalized. Dynamic time warping can be applied across the normalized long term point statistics and short term point statistics. A pair of probability distributions can be generated associated with the dynamic time warped normalized long term point statistics and short term point statistics. Based on distance between the mean values of the probability distributions, machine learning input features can be produced. The machine learning input features can be fed to train a machine learning model for detecting anomaly.

Abstract: In an example embodiment, a random forest machine learning algorithm is used to create and/or identify rules to apply to an individual entity in a computer system that has a plurality of entities, each with a number of rules. More precisely, rule predicates are used as features of a random forest model built to predict a particular outcome (e.g., a transaction that is fraudulent). Hyperparameters of the random forest model are varied and iterated. A classifier is used to calculate feature importance for all features in the training data. Feature importance may be calculated using permutation feature importance. The N “most important” features are then found from this set. The N “most important” features are then used to find rules above a certain precision and recall rate. These rules may then be backtested and the best rules can be used to generate additional rules.

Abstract: An artificial intelligence (AI) feedback method includes: providing input1 and input2 to a script extractor and executor, providing, by the script extractor and executor, an initial condition to an AI, generating, by the AI, a two-dimensional (2D) information table from the initial condition, adding the popularity frequency value or ranking value to corresponding coordinates in the 2D information table and matching knowledge by an interaction between the AI and the script extractor and executor, generating, by the AI, a 2D pseudolinear transformation table from the 2D information table, and performing, by the AI, pseudolinear transformation multiple times to form a 2D unique characteristic table and deriving information of coordinates whose characteristic vector is not changed as a deep truth value.

Abstract: A method, system, and computer program product are configured to create a tuned data record matching model by adjusting values of one or more parameters in a data record matching model based on a second training data set labeled at a data record level, wherein the data record matching model is initially trained using a first training data set labeled at an attribute level.

Abstract: Some embodiments provide a non-transitory machine-readable medium that stores a program. The program may receive a set of data from a data source. The program may generate a plurality of time series data based on the set of data. The program may determine a subset of the plurality of time series data as anomalies. The program may provide notifications indicating that the subset of the plurality of time series data are anomalies.

Abstract: The disclosed technology is generally directed to data classification. In one example of the technology, training data and a ground truth that indicates sensitive data within the training data is received. Based at least on the training data, natural language processing is used to learn features. The features include a naming feature that is associated with names of data resources in the training data. Based at least on the training data and the ground truth, using supervised learning, a model that is a heuristic model and/or a machine learning model is created. Input data information that is associated with input data is received. The model is used to determine a data resource sensitivity estimator (DRSE) value for each portion of the input data. The determination is based on the combination of features for the input data. Potentially sensitive data within the input data is flagged based on the DRSE values.

Abstract: Aspects described herein may allow unmatched records in databases be matched automatically. For example, a computing device may receive source records and target records to be matched together. The computing device may determine, based on a machine learning model and for each of the plurality of target records, a distance value between the respective target record, and a subset of the plurality of source records. A matched record that identifies the subset of the plurality of source records and a selected one or more target records may be generated. The matched record may be configured to update records in a database.

Abstract: Taxonomy management may be performed for generating and executing data labeling jobs. Data labeling taxonomy versions are managed and reused for data labeling applications generated and executed by a data labeling system. Instead of building labeling jobs through the use of manually managed configuration files that provide a taxonomy, a single taxonomy can be registered, updated, and used multiple times for various different types of labeling scenarios, including chaining jobs and continual learning loops.

Abstract: Embodiments of the disclosed technologies receive a first-party trained model and a first-party data set from a first-party system into a protected environment, receive a first third-party data set into the protected environment, and, in a data clean room, joining the first-party data set and the first third-party data set to create a joint data set for the particular segment, tuning a first-party trained model with the joint data set to create a third-party tuned model, sending model parameter data learned in the data clean room as a result of the tuning to an aggregator node, receiving a globally tuned version of the first-party trained model from the aggregator node, applying the globally tuned version of the first-party trained model to a second third-party data set to produce a scored third-party data set, and providing the scored third-party data set to a content distribution service of the first-party system.

Abstract: A system and method for accelerating an adaptation of one or more machine learning models of a data handling and governance service includes implementing a data handling and governance platform digitally accessible by a target subscriber of the data handling and governance service, wherein the data handling and governance platform interfaces with a plurality of distinct subscriber-agnostic digital content machine learning classification models of the data handling and governance service; identifying a target subscriber-agnostic digital content machine learning classification model; adapting the target subscriber-agnostic digital content machine learning classification model to a subscriber-specific digital content machine learning classification model based on a training of the target subscriber-agnostic digital content machine learning classification model with at least one training corpus comprising subscriber-specific training data samples; and implementing the subscriber-specific digital content machine l

Abstract: A multi-class classifier (MCC) is trained using annotated data. The annotated data comprises instances of sample data and associated label data. Creation of the annotated data and subsequent active learning by the MCC uses resources. A target-aware active learning system selects sample data for addition to an annotation queue based on factors such as current accuracy of a particular class determination and priority of that class. As each instance in the sample data in the annotation queue is annotated and used for subsequent training, accuracy of particular classes is improved until a specified accuracy for that class is attained. By being selective in the ordering of instances in the annotation queue, overall resource usage and corresponding costs associated with creating annotated data and training is reduced. Overall accuracy for all classes is improved using a smaller overall set of annotated data compared to naïve approaches.

Abstract: A computer system is provided, which is capable of evaluating the degree of influence of training data on prediction accuracy of a decision tree type machine learning model, while suppressing increase in processing time thereof. A similarity score calculating unit uses a tree structure of a trained model of a target predictor to calculate, for each of training data used for learning this trained model, a similarity score in which is evaluated similarity between the training data in the trained model and other training data. An evaluating unit selects target data that is training data that is a target of evaluation from the training dataset on the basis of the similarity score, and calculates an influence score in which the degree of influence of the target data on accuracy of the trained model is evaluated.

Abstract: A data classification device includes: a data acquisition unit configured to acquire data to be classified; a classification unit configured to classify the data into one of a plurality of classes by using a learned model learned using a neural network; a similarity calculation unit that calculates the similarity between the data and learned data used to generate the learned model; and a determination unit configured to determine whether the data belongs to the classes on the basis of the similarity, and, when the data does not belong to any of the classes, determine that the data belongs to an unknown class.

Abstract: Embodiments of computer-implemented systems and methods for extracting tables from documents are described. Document layout data is generated based on a plurality of glyphs associated with document lines, comprising data identifying: a plurality of text segments within each line; a plurality of text segment links; a plurality of text blocks of one or more of the text segments; and a plurality of text block links. A document is generated including at least one editable document table corresponding to at least one document table identified based on the plurality of glyphs associated with document lines.

Abstract: Intent-based automation that discovers automatable tasks and/or determines task variants in data is disclosed. Task capture data may be utilized to determine task variants in task mining data. Semantic understanding of user actions by artificial intelligence (AI)/machine learning (ML) model(s), for example, may be applied to determine the intent of the user rather than only focusing on what actions the user is performing on the computing system. Application logs and semantic understanding may be used to facilitate a more accurate determination of what the user actually intends to do. Task capture for individual user flows may be performed. Once these are captured, task capture algorithms and AI/ML models are used to determine which parts of the flows are similar and/or match and which parts are unique. The path through these flows can then be followed to build a process graph that includes decision points representing the unique flows.

Abstract: The present invention relates to a method, a service server, and a computer-readable medium for selectively extracting data for labeling, and more particularly, to a method, a service server, and a computer-readable medium for selectively extracting data for labeling, capable of selectively providing data for effectively training a machine learning model that is desired to be achieved by a user by deriving a plurality of feature vectors for a plurality of data included in an original dataset, applying a preset rule to the feature vectors to select a preset number of feature vectors among the feature vectors, and plotting the selected feature vectors and unselected feature vectors on a plane of three or less dimensions to provide a plotting result to the user.

Abstract: The present teachings include techniques and systems for processing spreadsheets at the level of a worksheet or another collection of cells, including the fragmenting of spreadsheets. For example, a spreadsheet may include one or more input worksheets, output worksheets, and/or transient worksheets, where transient worksheets utilize input data in input worksheets to facilitate creation of output data in output worksheets. In particular, fragmenting a spreadsheet may include determining, for each output worksheet, what input and transient worksheets are needed to create output data in the output worksheet, and then creating a plurality of fragments therefrom. Thus, fragmenting a spreadsheet may include a temporal and/or physical decoupling of the spreadsheet's worksheets. This can be useful for, inter alia, understanding the relationships between a plurality of worksheets, isolating certain processing tasks, editing or interchanging worksheets, and/or for improving computational performance, e.g.