A customer who is contacting customer care via a support session regarding a problem is classified into a customer category of multiple customer categories based at least on customer account information of the customer. A customer care topic in a predetermined set of multiple customer care topics that correspond to the problem is then identified via machine learning. A topic script that corresponds to the customer category of the customer for the customer care topic in the predetermined set of customer care topics is further retrieved or generated, in which the topic script includes one or more topic issues related to the customer care topics. The topic script is provided for presentation to a customer service representative (CSR) to prompt the CSR to discuss the one or more topic issues related to the customer care topic with the customer.

Structured documents are processed using convolutional neural networks. For example, the processing can include receiving a rendered form of a structured document; mapping a grid of cells to the rendered form; assigning a respective numeric embedding to each cell in the grid, comprising, for each cell: identifying content in the structured document that corresponds to a portion of the rendered form that is mapped to the cell, mapping the identified content to a numeric embedding for the identified content, and assigning the numeric embedding for the identified content to the cell; generating a matrix representation of the structured document from the numeric embeddings assigned to the cells of the grids; and generating neural network features of the structured document by processing the matrix representation of the structured document through a subnetwork comprising one or more convolutional neural network layers.

Techniques for generating an ontology based on biomarker information associated with persons to facilitate improving clinical predictions relating to medical conditions are presented. An ontology generator component (OGC) can extract clinical features associated with patients and their associated times from medical records or databases to develop clinical profiles associated with the patients and relating to a medical condition. OGC can develop an ontology relating to the medical condition, including progression and severity of biomarkers associated with the medical condition, based on the clinical profiles and domain knowledge information relating to the medical condition. OGC can determine global features relating to progression and severity associated with the medical condition based on the ontology. At a forecasting point, the global features can be extracted from the ontology and applied to a prediction model to enhance prediction of onset of, or progression of, the medical condition for a patient.

Techniques for generating an ontology based on biomarker information associated with persons to facilitate improving clinical predictions relating to medical conditions are presented. An ontology generator component (OGC) can extract clinical features associated with patients and their associated times from medical records or databases to develop clinical profiles associated with the patients and relating to a medical condition. OGC can develop an ontology relating to the medical condition, including progression and severity of biomarkers associated with the medical condition, based on the clinical profiles and domain knowledge information relating to the medical condition. OGC can determine global features relating to progression and severity associated with the medical condition based on the ontology. At a forecasting point, the global features can be extracted from the ontology and applied to a prediction model to enhance prediction of onset of, or progression of, the medical condition for a patient.

Certain aspects of the disclosure are directed to monitoring user-data communications corresponding to a user-generated message. According to a specific example, user-data communications, which are addressed to a client among a plurality of remotely-situated client entities, are directed to a message recording system. Each of the plurality of remotely-situated client entities are respectively configured and arranged to interface with a data communications server providing data communications services on a subscription basis. During recording of a message associated with the user-data communications and on the message recording system, speech characteristic parameters of the message may be analyzed, and a sentiment score and a criticality score for the message, may be determined. During the recording of the message, the user-data communications may be routed based on the determined sentiment score and criticality score.

Solution automation & interface analysis components can be implemented in many ways, such as by specifying input/outputs & training a learning (generate, test & update) algorithm on the input/output data to generate a prediction function, to replace code connecting input & outputs.

Alternatively, additional specific example structure (like code/configuration/data) implementations to connect input/outputs of sub-tasks like core interaction functions & problem-solving intents to implement solution automation & interface analysis are included in the specification of this invention.

Solution automation & interface analysis components can be implemented in many ways, such as by specifying input/outputs & training a learning (generate, test & update) algorithm on the input/output data to generate a prediction function, to replace code connecting input & outputs.

Alternatively, additional specific example structure (like code/configuration/data) implementations to connect input/outputs of sub-tasks like core interaction functions & problem-solving intents to implement solution automation & interface analysis are included in the specification of this invention.

A goal management system receives input of a qualitative first goal related to a body of a user (step S111), identifies a quantitative second goal related to the body of the user from the first goal thus received (step S112 to step S117), and presents the second goal thus identified (step S118). The first goal is converted into the quantitative goal for at least one of a plurality of feature amounts related to the body, thereby identifying the second goal including at least one goal obtained by such conversion. The first goal is converted into the quantitative goal for each feature amount corresponding to a meaning obtained by linguistic analysis. When there are a plurality of meanings obtained by linguistic analysis of the first goal, the first goal is converted into the quantitative goal for each feature amount on the basis of a range of each feature amount per meaning. This makes it possible to indicate a quantitative goal related to the body without receiving input of a goal that is a quantitative numerical value related to the body.

Systems and methods are disclosed for flexibly applying a query term to heterogeneous data. A query system can receive a query that includes a data-determinant query term. As the system executes the query it can generate interim search results. As the system query processes the interim search results based on the query, it can apply the data-determinant query term to records of the interims search results based on the structure of the records.

A system for assessing text content of a product. The system includes a computing device having a processor and a storage device storing computer executable code. The computer executable code, when executed at the processor, is configured to: provide text contents and confounding features of products; train a first regression model using the text content and the confounding features of the products; train the second regression model using the confounding features; operate the first regression model using the text contents and the confounding features to obtain a total loss; operate the second regression model using the confounding features of to obtain a partial loss; subtract the total loss from the partial loss to obtain a residual loss; use the residual loss to evaluate models and parameters for the regression models; and use the first regression model to obtain log odds of the words indicating importance of the words.