Abstract: A predictive computational model is generated through a hybrid process. In the hybrid process, a trained predictive computational model is automatically generated based on a dataset, where the predictive computational model is trained to generate an output based on new data. The automatic process uses a pipeline to train the model and makes decisions in the steps of the pipeline. After the model is automatically trained, a representation of the pipeline is presented to a user in a user interface. The user interface allows the user to modify at least some decision made in the automatic machine learning process. One or more modifications are received from the user through the user interface and are used to refine the trained model. The refined model is deployed to generate an output based on new data.

Abstract: A feature engineering tool automatically generates a group of features for generating a machine learning model. The feature engineering tool selects primitives for a dataset and synthesizes a plurality of features based on the selective primitives and the dataset. The feature engineering tool iteratively applies the plurality of features to a different portion of the dataset to evaluate usefulness of the plurality of features. Based on the evaluation, it removes some of the plurality of features to obtain the group of features. The feature engineering tool also determines the importance factor for each feature in the group. The machine learning model is generated based on the features and their importance factors and can be used to make a prediction based on new data.

Abstract: A document to be analyzed and a set of key names to be extracted from the document are received. A set of strings of characters contained within the document and a location for each string of character are identified. Moreover, a document graph for the document is generated. The document graph includes a set of nodes and a set of edges. Each node of the set of nodes corresponds to a string of characters of the set of strings of characters. Each edge of the set of edges connects two or more nodes together. Additionally, based on the document graph and the received set of key names, a set of keys are identified. Furthermore, a set of values are extracted from the document, and a set of key-value pairs are generated based on the identified set of keys and the extracted set of values.

Abstract: A model is trained through a hybrid machine learning process. In the hybrid machine landing process, an automatic machine learning process is performed on a dataset to generate a model for making a prediction. The automatic machine learning process uses a pipeline to train the model and makes decisions in the steps of the pipeline. After the model is trained through the automatic machine learning process, a representation of the pipeline is generated and presented to a user in a user interface. The user interface allows the user to modify at least some decision made in the automatic machine learning process. One or more modifications are received from the user through the user interface and are used to refine the trained model. The refined model is deployed to make the prediction based on new data.

Abstract: A document to be analyzed and a set of key names to be extracted from the document are received. A set of strings of characters contained within the document and a location for each string of character are identified. Moreover, a document graph for the document is generated. The document graph includes a set of nodes and a set of edges. Each node of the set of nodes corresponds to a string of characters of the set of strings of characters. Each edge of the set of edges connects two or more nodes together. Additionally, based on the document graph and the received set of key names, a set of keys are identified. Furthermore, a set of values are extracted from the document, and a set of key-value pairs are generated based on the identified set of keys and the extracted set of values.

Abstract: A module manages usage of a memory in a computer by scheduling data processing tasks. The module determines whether a memory associated with a computer is available for performing a data processing task. In response to determining that the memory is not available for performing the data processing task, the module generates a factory object for the data processing task. The factory object specifies instructions to produce the data processing task. After the factory object is generated, the module receives a request to perform the data processing task. The module produces the data processing task by using the factory object in response to the request and determine again whether the memory is available for performing the data processing task. In response to determining that the memory is now available for performing the data processing task, the module performs the data processing task by using the memory.

Abstract: A feature engineering tool automatically generates a group of features for training a machine learning model. The feature engineering tool selects primitives for a dataset and synthesizes a plurality of features based on the selective primitives and the dataset. The feature engineering tool iteratively applies the plurality of features to a different portion of the dataset to evaluate usefulness of the plurality of features. Based on the evaluation, it removes some of the plurality of features to obtain the group of features. The feature engineering tool also determines the importance factor for each feature in the group. The machine learning model is generated based on the features and their importance factors and can be used to make a prediction based on new data.

Abstract: A data analytics application receives a workflow that includes a sequence of tools. Each tool performs a data analytics function. The data analytics application processes a data file using the sequence of tools to generate a result item representing an outcome of the processing of the data file. The data analytics application stores one or more metadata files, each of which includes data generated by an interactive tool in the sequence during the processing of the data file. The data analytics application receives a user input through an interactive element associated with an interactive tool in the sequence. The interactive element can modify an operation of the interactive tool based on the user input. The data analytics application retrieves the metadata file for the interactive tool and processes the metadata file by using a subset of the sequence of tools and the user input to generate a different result item.

Abstract: A data analytics application receives a workflow that includes a sequence of tools. Each tool in the sequence performs a data analytics function. The data analytics application generates an interactive element for a tool in the sequence based on the data analytics function of the tool in the workflow. The interactive element is applied to the tool to update the workflow. The interactive element may be affiliated to the tool or be added into the analytics workflow as a new tool. The updated workflow is provided for processing a data file. The updated analytics workflow generates a result item representing an outcome of the processing of the data file. The interactive element applied to the tool is adapted to receive a user input and to trigger an operation of the tool based on the user input to generate a different result item.

Abstract: A data analytics system stores a data file that includes an ordered set of data blocks. The data blocks can be parsed out of order. An error management module of the data analytics system detects a parse error occurring during parsing of a data block and generates an error message for the parse error. The error message includes unresolved location information indicating a location of the detected parse error in the data block. The error management module resolves the unresolved location information after determining that one or more additional data blocks preceding the data block in the ordered set have been parsed. The error management module generates resolved location information that indicates a location of the parse error in the data file. The error management module updates the error message with the resolved location information and outputs the updated error message.

Abstract: A data analytics system stores a plurality of sequential data blocks in a block cache. Each data block contains one or more data records. Each data record of a data record contains one or more values and a record marker marking the end of the data record. The data analytics system allocates a data block in the block cache to a worker thread. The worker thread parses the data block to identify the one or more data records contained within the data block. The data analytics system determines whether the data block was correctly parsed. Responsive to determining that the data block was correctly parsed, the data analytics system provides the one or more data records contained within the data block to a data analytics workflow.

Abstract: Data records are joined using a computer. Data records in a first plurality of data records and a second plurality of data records are hashed. The data records in the first and second pluralities are respectively assigned to first and second groupings based on the hashes. Associated pairs of groupings from the first and second groupings are provided to a thread executing on a computer processor, and different pairs are provided to different threads. The threads operate on the pairs of groupings in parallel to determine whether to join the records in the groupings. A thread joins two data records under consideration if the hashes associated with the data records match. The joined data records are output.

Abstract: A data analytics system stores a data file that includes an ordered set of data blocks. The data blocks can be parsed out of order. An error management module of the data analytics system detects a parse error occurring during parsing of a data block and generates an error message for the parse error. The error message includes unresolved location information indicating a location of the detected parse error in the data block. The error management module resolves the unresolved location information after determining that one or more additional data blocks preceding the data block in the ordered set have been parsed. The error management module generates resolved location information that indicates a location of the parse error in the data file. The error management module updates the error message with the resolved location information and outputs the updated error message.

Abstract: A module manages memory in a computer. The module monitors usage of a primary memory associated with the computer. The primary memory stores memory blocks in a ready state. In response to primary memory usage by the memory blocks in the ready state exceeding a ready state threshold, the module compresses at least some of the memory blocks in the ready state to form memory blocks in a ready and compressed state. In response to primary memory usage by the memory blocks in the ready and compressed state exceeding a release threshold, the module releases at least some of the memory blocks in the ready and compressed state. In response to primary memory usage by the memory blocks in the compressed state exceeding a compressed threshold, the module transfers at least some memory blocks in the compressed state to a secondary memory associated with the computer.

Abstract: A module manages memory in a computer. The module monitors usage of a primary memory associated with the computer. The primary memory stores memory blocks in a ready state. In response to primary memory usage by the memory blocks in the ready state exceeding a ready state threshold, the module compresses at least some of the memory blocks in the ready state to form memory blocks in a ready and compressed state. In response to primary memory usage by the memory blocks in the ready and compressed state exceeding a release threshold, the module releases at least some of the memory blocks in the ready and compressed state. In response to primary memory usage by the memory blocks in the compressed state exceeding a compressed threshold, the module transfers at least some memory blocks in the compressed state to a secondary memory associated with the computer.

Abstract: A data analytics system stores a plurality of sequential data blocks in a block cache. Each data block contains one or more data records. Each data record of a data record contains one or more values and a record marker marking the end of the data record. The data analytics system allocates a data block in the block cache to a worker thread. The worker thread parses the data block to identify the one or more data records contained within the data block. The data analytics system determines whether the data block was correctly parsed. Responsive to determining that the data block was correctly parsed, the data analytics system provides the one or more data records contained within the data block to a data analytics workflow.

Abstract: Data records are joined using a computer. Data records in a first plurality of data records and a second plurality of data records are hashed. The data records in the first and second pluralities are respectively assigned to first and second groupings based on the hashes. Associated pairs of groupings from the first and second groupings are provided to a thread executing on a computer processor, and different pairs are provided to different threads. The threads operate on the pairs of groupings in parallel to determine whether to join the records in the groupings. A thread joins two data records under consideration if the hashes associated with the data records match. The joined data records are output.