Abstract: Systems and methods for dynamic user profile projection are provided. One or more aspects of the systems and methods includes computing, by a prediction component, a predicted number of lookups for a future time period based on a lookup history of a user profile using a lookup prediction model; comparing, by the prediction component, the predicted number of lookups to a lookup threshold; and transmitting, by a projection component, the user profile to an edge server based on the comparison.

Abstract: Systems and methods for dynamic user profile management are provided. One aspect of the systems and methods includes receiving, by a lookup component, a request for a user profile; computing, by a profile component, a time-to-live (TTL) refresh value for the user profile based on a lookup history of the user profile; updating, by the profile component, a TTL value of the user profile based on the request and the TTL refresh value; storing, by the profile component, the user profile and the updated TTL value in the edge database; and removing, by the edge database, the user profile from the edge database based on the updated TTL value.

Abstract: A computer-implemented method according to one embodiment includes receiving, at a deep neural network (DNN), a plurality of sentences each having an associated label; training the DNN, utilizing the plurality of sentences and associated labels; and producing a linguistic expression (LE) utilizing the trained DNN.

Abstract: One embodiment provides for a method for evaluation of an artificial intelligence (AI) service, the method includes partitioning, by a processor, data into in-domain data and out-of-domain data. The processor defines held-out data from both of the in-domain data and the out-of-domain data for evaluation by each of domain and sub-domain based on building a taxonomy of both domains and sub-domains for the AI service. The held-out data is excluded from training data used for training the AI service. The processor further determines distribution underlying performance metrics for the held-out data using bootstrap validation processing. The processor also determines performance guarantees for multiple settings conditioned on multiple characteristics of an application scenario for the held-out data of the taxonomy based on the underlying performance metrics.

Abstract: In some embodiments, a model training system trains a sample generation model configured to generate synthetic data entries for a dataset. The sample generation model includes a prior model for generating an estimated latent vector from a partially observed data entry, a proposal model for generating a latent vector from a data entry of the dataset and a mask corresponding to the partially observed data entry, and a generative model for generating the synthetic data entries from the latent vector and the partially observed data entry. The model training system trains the sample generation model to optimize an objective function that includes a first term determined using the synthetic data entries and a second term determined using the estimated latent vector and the latent vector. The trained sample generation model can be executed on a client computing device to service queries using the generated synthetic data entries.

Abstract: One embodiment provides for a method for evaluation of an artificial intelligence (AI) service, the method includes partitioning, by a processor, data into in-domain data and out-of-domain data. The processor defines held-out data from both of the in-domain data and the out-of-domain data for evaluation by each of domain and sub-domain based on building a taxonomy of both domains and sub-domains for the AI service. The held-out data is excluded from training data used for training the AI service. The processor further determines distribution underlying performance metrics for the held-out data using bootstrap validation processing. The processor also determines performance guarantees for multiple settings conditioned on multiple characteristics of an application scenario for the held-out data of the taxonomy based on the underlying performance metrics.

Abstract: One embodiment provides for a method for evaluation of an artificial intelligence (AI) service, the method includes partitioning, by a processor, data into in-domain data and out-of-domain data. The processor defines held-out data from the in-domain data and the out-of-domain data for evaluation by domain and sub-domain based on building a taxonomy of domains and sub-domains for the AI service. The processor further determines distribution underlying performance metrics for the held-out data using statistical processing. The processor also determines performance guarantees for multiple settings conditioned on multiple characteristics of an application scenario for the held-out data of the taxonomy based on the underlying performance metrics. The processor further provides confidence intervals based on the performance guarantees.

Abstract: In some embodiments, a model training system trains a sample generation model configured to generate synthetic data entries for a dataset. The sample generation model includes a prior model for generating an estimated latent vector from a partially observed data entry, a proposal model for generating a latent vector from a data entry of the dataset and a mask corresponding to the partially observed data entry, and a generative model for generating the synthetic data entries from the latent vector and the partially observed data entry. The model training system trains the sample generation model to optimize an objective function that includes a first term determined using the synthetic data entries and a second term determined using the estimated latent vector and the latent vector. The trained sample generation model can be executed on a client computing device to service queries using the generated synthetic data entries.

Abstract: Embodiments are provided for analysis of errors of a predictive model. In some embodiments, a system can include a processor that executes computer-executable components stored in memory. The computer-executable components can include an overview component that causes a client device to present first data identifying an error corresponding to a cell of a confusion matrix for a classification model, the error representing a mismatch between a first label generated by the classification model and a second label corresponding to a ground-truth observation. The computer-executable components also can include an element view component that receives second data defining a root cause of the error. The computer-executable components can further include an error annotation component that can embed the second data into a first data structure containing the first data, resulting in a first annotated data structure.

Abstract: Methods, systems, and computer program products for table recognition in PDF documents are provided herein. A computer-implemented method includes discretizing one or more contiguous areas of a PDF document; identifying one or more white-space separator lines within the one or more discretized contiguous areas of the PDF document; detecting one or more candidate table regions within the one or more discretized contiguous areas of the PDF document by clustering the one or more white-space separator lines into one or more grids; and outputting at least one of the candidate table regions as a finalized table in accordance with scores assigned to each of the one or more candidate table regions based on (i) border information and (ii) cell structure information.

Abstract: A computer-implemented method according to one embodiment includes receiving, at a deep neural network (DNN), a plurality of sentences each having an associated label; training the DNN, utilizing the plurality of sentences and associated labels; and producing a linguistic expression (LE) utilizing the trained DNN.

Abstract: One embodiment provides for a method for evaluation of an artificial intelligence (AI) service, the method includes partitioning, by a processor, data into in-domain data and out-of-domain data. The processor defines held-out data from the in-domain data and the out-of-domain data for evaluation by domain and sub-domain based on building a taxonomy of domains and sub-domains for the AI service. The processor further determines distribution underlying performance metrics for the held-out data using statistical processing. The processor also determines performance guarantees for multiple settings conditioned on multiple characteristics of an application scenario for the held-out data of the taxonomy based on the underlying performance metrics. The processor further provides confidence intervals based on the performance guarantees.

Abstract: Methods, systems, and computer program products for table recognition in PDF documents are provided herein. A computer-implemented method includes discretizing one or more contiguous areas of a PDF document; identifying one or more white-space separator lines within the one or more discretized contiguous areas of the PDF document; detecting one or more candidate table regions within the one or more discretized contiguous areas of the PDF document by clustering the one or more white-space separator lines into one or more grids; and outputting at least one of the candidate table regions as a finalized table in accordance with scores assigned to each of the one or more candidate table regions based on (i) border information and (ii) cell structure information.

Abstract: According to one embodiment of the present invention, a computer-implemented method of performing analytics on a large quantity of data accommodated by an external mass storage device is provided. The analytics may be divided into a set of modules, wherein each module is selectively executed and comprises a script for a parallel processing engine to perform a corresponding atomic operation on the analytics. A user selection is received of one or more modules to perform desired analytics on the large quantity of data from the external mass storage device, and the selected modules execute scripts for the parallel processing engine to perform the corresponding atomic operations of the desired analytics.

Abstract: According to one embodiment of the present invention, a computer-implemented method of performing analytics on a large quantity of data accommodated by an external mass storage device is provided. The analytics may be divided into a set of modules, wherein each module is selectively executed and comprises a script for a parallel processing engine to perform a corresponding atomic operation on the analytics. A user selection is received of one or more modules to perform desired analytics on the large quantity of data from the external mass storage device, and the selected modules execute scripts for the parallel processing engine to perform the corresponding atomic operations of the desired analytics.

Abstract: One embodiment provides a method for developing a text analytics program for extracting at least one target concept including: utilizing at least one processor to execute computer code that performs the steps of: initiating a development tool that accepts user input to develop rules for extraction of features of the at least one target concept within a dataset comprising textual information; developing, using the rules for feature extraction, an evaluation dataset comprising at least one document annotated with the at least one target concept to be extracted by the text analytics program; creating, using the rules for feature extraction, a rule-based annotator to extract the at least one target concept; training, using the evaluation dataset, a machine-learning annotator to extract the at least one target concept within the dataset; combining the rule-based annotator and the machine learning annotator to form a combined annotator; evaluating, using the evaluation dataset, extraction performance of the combine

Abstract: Parallel execution of machine learning programs is provided. Program code is received. The program code contains at least one parallel for statement having a plurality of iterations. A parallel execution plan is determined for the program code. According to the parallel execution plan, the plurality of iterations is partitioned into a plurality of tasks. Each task comprises at least one iteration. The iterations of each task are independent. Data required by the plurality of tasks is determined. An access pattern by the plurality of tasks of the data is determined. The data is partitioned based on the access pattern.

Abstract: One embodiment provides a method for developing a text analytics program for extracting at least one target concept including: utilizing at least one processor to execute computer code that performs the steps of: initiating a development tool that accepts user input to develop rules for extraction of features of the at least one target concept within a dataset comprising textual information; developing, using the rules for feature extraction, an evaluation dataset comprising at least one document annotated with the at least one target concept to be extracted by the text analytics program; creating, using the rules for feature extraction, a rule-based annotator to extract the at least one target concept; training, using the evaluation dataset, a machine-learning annotator to extract the at least one target concept within the dataset; combining the rule-based annotator and the machine learning annotator to form a combined annotator; evaluating, using the evaluation dataset, extraction performance of the combine

Abstract: A method includes receiving one or more natural language dependency parse trees as input. A hardware processor is used for processing the dependency parse trees by creating a mapping from nodes of the one or more dependency parse trees into actions, roles and contextual predicates. The mapping is used for information extraction. The actions include the verbs along with attributes of the verbs. The roles include arguments for the verbs. The contextual predicates include modifiers for the verbs.

Abstract: Described herein are methods, systems and computer program products for entity resolution. Entity resolution, also known as entity matching or record linkage, seeks to identify equivalent data objects between or among datasets. An example method includes creating a deterministic model by defining an entity to be resolved, selecting two datasets for comparison, defining matching predicates for attributes of the datasets to select a set of candidate matches, and defining a precedence rule for the candidate matches to select a subset of the candidate matches. The method includes running the deterministic model on the two datasets. Running the deterministic model includes applying the matching predicates and the precedence rule to data in the datasets that correspond to the attributes. The method also includes applying a cardinality rule to results of the running, and outputting the matching candidates for which the cardinality rule is satisfied.