Abstract: To generate data for populating/updating accounting databases based on digitized accounting source documents, access to an entity database comprising identifiers of entities associated with an accounting database and to a digital template library comprising processing templates for processing digitized accounting source documents is provided. Each entity in the entity database is associated with one processing template.

Abstract: A computing device may determine a plurality of popular topics for a time period. The computing device may determine a plurality of content items associated with at least one of the plurality of popular topics. The computing device may determine an output time for each of the plurality of content items. The computing device may determine a ranking order for the plurality of content items. The ranking order may be based on the output time for each of the plurality of content items. The computing device may output or cause to be output an indicator of at least a portion of the plurality of content items in the ranking order.

Abstract: An embodiment for managing data using machine learning models and information governance. The embodiment may automatically detect a data analysis request made within a system and identify subject datasets. The embodiment may automatically conduct shallow term assignments on each row and column of data in the subject datasets and automatically match the shallow term assignments for each row and column with a stored set of ranked terms, and automatically flag rows or columns matching with ranked terms above a predetermined threshold ranking for further analysis. The embodiment may automatically and continuously monitor and detect irrelevant metadata types to prevent subsequent analysis and storage of data including the irrelevant metadata types. The embodiment may automatically generate a criticality ranking for stored analysis datasets.

Abstract: Data includes a first dataset with a first set of attributes and a second dataset with a second set of attributes is received. The first dataset and the second dataset are associated with an industrial asset. A first attribute value of a first attribute in the first set of attributes is assigned a first priority value and a second attribute value of the first attribute in the second set of attributes is assigned a second priority value. A third dataset that includes a third set of attributes of the industrial asset is generated. The third set of attributes includes the first attribute. An attribute value between the first attribute value and the second attribute value is selected based on the first priority value and the second priority value. The first attribute of the third set of attributes of the third dataset is set to the selected attribute value.

Abstract: A cognate specification system manages and enforces digital, canonical representations of entities that are added to and in the system by creating a composite entity key (“EntityKey”) that uniquely and immutably identifies each entity within the system, and that is created based upon the attributes of the entity rather than being arbitrarily assigned. Entities may be organized into various types or cognates, each associated with distinct sets of attributes. A physical or digital referent may be added to the system as an entity by specifying attributes. In such a system, any entity, whether a physical or digital referent, a process, or other type, is uniquely and verifiably identifiable, and may be linked to or associated with other entities, allowing users of the system to produce and access specification granularity beyond traditional approaches focused on systems interoperability.

Abstract: Disclosed herein are systems and methods for intelligent metadata management and data lineage tracing. In exemplary embodiments of the present disclosure, a data element can be traced throughout multiple applications, platforms, and technologies present in an enterprise to determine how and where the specific data element is utilized. The data element is traced via a hierarchical key that defines it using metadata. In this way, metadata is interpreted and used to trace data lineage from one end of an enterprise to another.

Abstract: Each year, almost 10% of claims are denied by payers (i.e., health insurance plans). With the cost to recover these denials and underpayments, predicting payer response (likelihood of payment) from claims data with a high degree of accuracy and precision is anticipated to improve healthcare staffs' performance productivity and drive better patient financial experience and satisfaction in the revenue cycle (Barkholz, 2017). However, constructing advanced predictive analytics models has been considered challenging in the last twenty years. That said, we propose a (low-level) context-dependent compact representation of patients' historical claim records by effectively learning complicated dependencies in the (high-level) claim inputs. Built on this new latent representation, we demonstrate that a deep learning-based framework, Deep Claim, can accurately predict various responses from multiple payers using 2,905,026 de-identified claims data from two US health systems.

Abstract: Systems and methods for unifying business records data from different database sources into a common format in the same repository are disclosed. In one embodiment, a process includes receiving, by an integration platform system, a first set of business records data from a customer records system, where the business records data includes records in a first format, identifying a first transformation rule corresponding to the first format using the integration platform system, where the first transformation rule includes information for converting data from the first format to a unified format, converting each record in the first set of business records from the first format to the unified format by the integration platform system using the first transformation rule, and storing the first set of converted records in the unified format in a business records database.

Abstract: Examples of the present disclosure provide apparatuses and methods for direct data transfer. An example method comprises transferring data between a first device and a second device, wherein the first device is a bit vector operation device, and transforming the data using a data transform engine (DTE) by rearranging the data to enable the data to be stored on the first device when transferring the data between the second device and first memory device.

Abstract: A computer-implemented method can comprise establishing programmatic connections to a digitally stored first database comprising over one million records, each of the records comprising columns; reading a configuration file that specifies tables in the database; for each particular table, forming and submitting a plurality of queries to the database, each of the queries specifying data aggregation operations, and in response thereto, receiving result sets of records of the database; calculating metadata metrics that characterize columns of the records in the result sets and storing the metadata metrics in tables for string column statistics, numeric column statistics, date column statistics, based upon a particular data type among different data types of the columns; generating presentation instructions which when rendered cause displaying one or more graphical visualizations in a graphical user interface.

Abstract: A centralized database management system performs data synchronization with lower bandwidth consumption and higher efficiency using a resizable invertible bloom filter. The system may include a resizable invertible bloom filter module that constructs and maintains invertible bloom filters that are resizable based on a number of differences between different snapshots. The resizable invertible bloom filter module may maintain a list of possible sizes for a resizable invertible bloom filter. The resizable invertible bloom filter module may determine and maintain a list of applicable partition sizes, each partition size being a product of a divisor and a resizing factor. If the number of differences exceeds the number of expected differences and results in failure in decoding, the system may retry a larger size in a set of predetermined sizes. The system may continue to try larger sizes until a minimal size required for successful decoding is found.

Abstract: An information processing apparatus, establishing data synchronization with a second information processing apparatus, includes a processor configured to: register, in information processing apparatus update log information, identification information on groups each including related data among data updated in the information processing apparatus; in response to reception of second information processing apparatus update log information including identification information on a group including related data among data updated in the second information processing apparatus, perform a synchronization process by checking the second information processing apparatus update log information against the information processing apparatus update log information, the synchronization process establishing synchronization on data forming each synchronization target group, the synchronization target group being one of the groups and as a synchronization target by the identification information thereof; and delete the identifi

Abstract: The systems and methods described here can reduce the storage space required (memory and/or disk) to store certain types of data, provide efficient (fast) creation, modification and retrieval of such data, and support such data within the framework of a multi-version database. In some embodiments, the systems and methods can store each field of a set of records as a vector of values, e.g., a data vector. A set of records can be represented using a vector id vector, or “vid” vector, wherein each element of the vid vector contains a reference to the memory location of a data vector. A header table can store associations between labels and “vid” vectors that pertain to those labels. Identical data vectors can be re-used between different record sets or vid vectors needing that vector, thus saving space.

Abstract: The present disclosure provides methods, systems and apparatus for efficient data management in distributed database system. According to an aspect, a method is provided. The method includes relocating a first table and a second table of a database to a first node of the database, wherein the first table and a second table (or one or more tables which have a colocation relationship with the first table) are associated with one or more distribution keys corresponding to one or more columns of the first table and the second table. The method further includes defining a colocation relationship between the first table and the second table based on a distribution key of the one or more distribution keys. The colocation relationship may indicate that data of the first table and the second table that correspond according to the distribution key are to reside on a same node of the database.

Abstract: An approach is disclosed herein that retrieves data from a data set that includes first column data comprising a first data type and a second data type. The approach structures the first column data into second column data and third column data based on the first data type and the second data type. The approach determines a first semantic category and a second semantic category for the first data type and the second data type, and then determines a first privacy category and a second privacy category based on the first semantic category and the second semantic category. The approach anonymizes the second column data and the third column data to produce anonymized data based on the first privacy category and the second privacy category, respectively. In turn, the approach generates an anonymized view of the data set using the anonymized data.

Abstract: A computing device displays, in a graphical user interface, a canvas region that includes a first scene. The first scene includes a first visualization card. The device also displays, in the graphical user interface, a recommendation region that includes one or more data visualizations. Each data visualization corresponds to a respective metric of one or more metrics of a data source. The respective metric has a respective temporal attribute. The device receives user selection to add a first data visualization, corresponding to a first metric of the one or more metrics, to the first visualization card. In response to the user selection, the device updates display of the first visualization card to include the first data visualization.

Abstract: A method for determining a spatial two-tuple is performed by a computer device. The method comprises: obtaining a given spatial distance and multiple spatial objects, wherein each spatial object belongs to a first spatial object set or belongs to a second spatial object set (101); determining a global area according to a minimum bounding rectangle (MBR) and an extended minimum bounding rectangle (EMBR) corresponding to the first spatial object set and an MBR and an EMBR corresponding to the second spatial object set (102); performing subspace division on the global area to obtain multiple subspaces (103); partitioning the multiple spatial objects according to a positional relationship between each spatial object and each subspace to determine spatial objects comprised in each subspace (104); and according to a spatial distance between the spatial objects of each subspace, determining a spatial two-tuple comprised in each subspace that matches the given spatial distance (105).

Abstract: A system to quantify aggregate alignment of segmented text with an evaluator population, with a data processing system comprising memory and one or more processors, can segment a first extended text object into one or more evaluation text objects associated with a population reference, identify one or more text frame objects corresponding to the evaluation text objects, the text frame objects being associated with a second extended text object, generate, based on the text frame objects, one or more context identifier objects corresponding to the evaluation text objects, and generate a condensed text object including one or more of the evaluation text objects, the evaluation text objects being positioned in the condensed text object in response to output of a first machine learning model trained with input including at least one of the first extended text objects, the evaluation text objects, the context identifier objects, and the text frame objects.

Abstract: The present invention is a computer-implemented system and method for creating a user profile based on multimedia files. The method includes storing multimedia files associated with one or more users in at least one computing device. The method includes obtaining content data from the multimedia files stored in the computing device through a content extraction module. The content extraction module performs steps to obtain the content data from multimedia files. A mobile application accesses the files stored in the computing device. Files are extracted and processed by utilizing one or more machine learning algorithms and retrieving content data and file metadata for each multimedia file from the processed data. Content-based user attributes are created by analyzing the aggregated content data and file metadata, Content-based user attributes are transmitted to an application server and stores the content-based user attributes to a user profile database.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for jointly training an image embedding model and a text embedding model. In one aspect, a method comprises: processing data from a historical query log of a search system to generate a candidate set of training examples, wherein each training example comprises: (i) a search query comprising a sequence of one or more words, (ii) an image, and (iii) selection data characterizing how often users selected the image in response to the image being identified by a search result for the search query; selecting a plurality of training examples from the candidate set of training examples; and using the training data to jointly train the image embedding model and the text embedding model.

Abstract: Examples described herein involve facilitating playback of media content via a webpage. An example implementation involves a computing device: identifying one or more playback zones including at least one respective playback device; transmitting, via a web browser, a webpage request; receiving a webpage that includes a link to a media item; modifying the received webpage to include a representation associated with the link; causing, via the web browser, display of the modified webpage; responsive to a selection of the representation, causing display of the option; and responsive to a selection of the option, causing display of a list of the one or more playback zones, wherein at least one playback zone of the one or more playback zones is selectable to cause the media item to be added to the playback queue associated with the at least one playback zone.

Abstract: Systems and methods are provided for performing random walk graph computing. One method may comprise generating a subset of walkers on a graph, maintaining the generated subset of walkers in a walker pool in a memory, loading a coarse-grained block of the graph from a non-volatile storage into a block buffer of the memory, generating pre-sampled edges for vertices in the coarse-grained block, storing the pre-sampled edges into a pre-sampled edge buffer allocated for the coarse-grained block and moving one or more walkers of the generated subset of walkers using the pre-sampled edges stored in the pre-sampled edge buffer. The generated subset of walkers may have an initial number determined based on a memory space allocated to the walker pool.

Abstract: Systems and methods are provided for performing temporal graph computing. One method may include ordering out edges in a candidate edge set of a vertex in deceasing time, grouping the out edges into a plurality of trunks with at least one of the plurality of trunks being a multi-edge trunk having two or more edges, generating a plurality of alias tables to record content of the plurality of trunks, performing inverse transform sampling on the plurality of trunks to choose a first trunk of interest, determining that the first trunk of interest is a complete multi-edge trunk with a plurality of edges arranged in a plurality of buckets and each bucket has an average weight of a total weight of the plurality of edges and sampling an edge from the plurality of edges by alias table sampling.

Abstract: The present specification relates to the field of digital information, and in particular, to a flow graph calculation and storage method and system. A flow graph includes nodes and edges, and the method includes one or more rounds of processing respectively corresponding to one or more time periods. A round of processing includes: obtaining a graph data segment of a current subgraph based on newly added flow information in a current time period and storing the graph data segment, so as to implement distributed storage of the current subgraph in collaboration with another calculation unit; and for a node in the graph data segment of the current subgraph, obtaining a flow message of data corresponding to an ingress edge of the node from a source node on the ingress edge and storing the flow message, where the flow message includes a source link of the data.

Abstract: A system and method for technology analysis utilizing high-performance, scalable, multitenant, dynamically specifiable, knowledge graph information storage and utilization. The system uses an in-memory associative array for high-performance graph storage and access, with a non-volatile distributed database for scalable backup storage, a scalable, distributed graph service for graph creation, an indexing search engine to increase searching performance, and a graph crawler for graph traversal. One or more of these components may be in the form of a cloud-based service, and in some embodiments the cloud-based services may be containerized to allow for multitenant co-existence with no possibility of data leakage or cross-over. The system uses a cyber-physical graph to represent an enterprise's cyber-physical system and can provide graph analysis, graph security, and graph fusion related tasks to identify potential operational risks.

Abstract: A method includes: receiving a user query; generating first embedding data for the user query via a language agnostic machine learning embedding model; and predicting a first intent of the user query based on the first embedding data.

Abstract: A computing server may receive a target data instance. The computing server may scan through a data store to identify a related data instance that shares one or more matched data strings with the target data instance. The computing server may determine one or more windows of data locality to which the one or more matched data strings belong. The computing server may determine one or more data group labels for the one or more windows of data locality that includes the one or more matched data strings. The computing server may store data group information of the matched data strings between the target data instance and the related data instance.

Abstract: As an example, a system includes a hierarchical tag database that includes a plurality of tags arranged according to a hierarchical data schema for a number of tag hierarchies, and tags within each level of a respective tag hierarchy have a different specificity relative to each other level of the respective tag hierarchy. The data can also include profile tag data for a given user. The system can also include a processor configured to access the media and execute the instructions to expand each of the stored profile tags for the given user to provide an expanded profile tag data set for the given user, in which the expanded profile tag data set includes each of the assigned profile tags and each one or more higher-level tag in each associated tag hierarchy to which each of the assigned profile tags belongs.

Abstract: A computing device receives a user input to add a data visualization to a first visualization card in a first scene of an interactive presentation. In response to the user input, when the first scene is an initial scene to be added, the device computes, for each metric of a plurality of metrics, a respective parameter that measures the variability of values of the respective metric, and identifies a first subset of metrics based on the computed parameters. The device generates a respective data visualization for each metric. The device displays a plurality of the generated data visualizations. The device receives user selection of a first data visualization of the plurality of data visualizations, corresponding to a first metric of the subset of metrics. In response to the user selection, the device populates the first visualization card with the first data visualization.

Abstract: A computing device displays, in a graphical user interface, a canvas region that includes a first scene. The first scene includes a first visualization card having a first data visualization and a first text card, adjacent to the first visualization card. The device receives, via the first text card, (i) text input from a user and (ii) user selection of a first user interface element for linking the first text card to the first visualization card. In accordance with the receiving, the device determines whether the text input includes a first expression having a first time span that intersects with a temporal domain of the first data visualization. In accordance with a determination that the text input includes the first expression, and in response to a first user interaction with a first region of the first text card that includes the first expression, the device visually emphasizes a first portion of the first data visualization, corresponding to the first time span.

Abstract: A central database system provides predictive web navigation using machine learning and clustering to guide a user to a web page. After tracking a number of web pages viewed by various users on one or more web domains and the orders in which these web pages are viewed, the central database system can train a model to predict which web page a user is likely to view next. If the user is lost while navigating, the central database system may guide the user to the predicted web page. In one example of guidance, the central database system presents a web element with a hyperlink to the predicted web page. For example, the central database system can modify a web page that the lost user is presently viewing to include an iframe with a recommendation to travel to a different web page.

Abstract: An invoice or other document is automatically classified, for example to identify a vendor or other entity associated with the document, based on geometric characteristics of the document. Geometric information extracted from incoming images of documents is used to reconstruct a unique template, thereby providing a mechanism for automatically classifying newly received documents, for example to determine which vendor is their source. The template for a target document is expressed as a feature vector, which can then be compared with feature vectors previously generated for reference documents, to provide a basis for determining similarity between the target document and various reference documents.

Abstract: The present disclosure is directed towards systems and methods for finding documents that are similar to a reference text. The inventive systems and methods examine a set of collected documents to determine the facts present in those documents by, for example, extracting triplets and expanding them. A user's input reference text is similarly examined to extract and expand triplets therein and the facts identified with respect to the reference text are used as a basis to find documents having similar facts. The present disclosure is also related to systems and methods for mining facts from documents relating to a primary source such as a piece of legislation and using the mined facts to improve the results of subsequent searches.

Abstract: A method for transferring electronic evidence is provided. The law enforcement agencies can make efficient use of social media and other forms of public communications to make a public appeal for information on crimes and other investigations wherein the public appeals allow members of the public to easily submit information and/or media files from smartphones and other computers in a way that allows the submission to be linked to the public appeal (e.g. the specific case file or the attributes of the case file) so that the submission data can be found and accessed by law enforcement investigators.

Abstract: In accordance with some embodiments, processes and interfaces provide for enhancing search results of a group research project. For example, members of a group may be provided with information regarding other group member's search activities and/or be restricted from viewing certain search results (e.g., search results that are most popular with the public for a given search term, that are most popular with the group for the given search term or for the project, and/or search results that are restricted by a group manager).

Abstract: A technique for providing search results may include determining a first entity type, a second entity type, and a relationship type based on a compositional query. The technique may also include identifying nodes of a knowledge graph corresponding to entity references of the first entity type and entity references of the second entity type. The technique may also include determining from the knowledge graph an attribute value corresponding to the relationship type for each entity reference of the first entity type and for each entity reference of the second entity type. The technique may also include comparing the attribute value of each entity reference of the first entity type with the attribute value of each entity reference of the second entity type. The technique may also include determining one or more resultant entity references from the entity references of the first entity type based on the comparing.

Abstract: Disclosed in embodiments of the present application is a method for displaying information in an application. A provider creates flowable information to be displayed by means of a data standard provided by a sharing platform, and provides same to the sharing platform. When one party (a display party) is to present information of another party (a provider party), the display party sends a display request to the sharing platform, the display request comprising an identifier of information to be displayed. Upon receiving the display request, the sharing platform acquires the information to be displayed according to the identifier in the display request, so that the display party can display the information to be displayed.

Abstract: To provide a presentation system for effectively displaying a keyword for effecting the selection of the next slide during a presentation. Provided is a display device comprising: a voice recognition unit 53; a conversation-derived term extraction unit 55; a search keyword storage unit 57; a search keyword extraction unit 59; a material storage unit 61; a relevant page information extraction unit 63; a selection term extraction unit 65; and a selection term display unit 71.

Abstract: The embodiments are directed to systems and methods for generating groups of news stories. A system for generating groups of news stories may include a processor and a memory. The memory may be coupled to the processor. The memory may include instructions configured to cause the processor to determine an article based on a topic. The memory may also include instructions configured to cause the processor to generate an article rank for the article based on an age of the article, a base rate of the article, and a bonus of the article. Additionally, the memory may include instructions configured to cause the processor to display the article based on the article rank being above a target rank.

Abstract: Systems, methods, and computer-readable storage media for aggregating media (and commentary on that media) into a topology. To do so, the system receives first content (and associated metadata) as well as second content (and associated metadata). The system then generates a topology based on a relationship between the first content and the second content, where the topology has a number of dimensions based on the metadata of the different pieces of content. The system then compares the topology and the metadata to previously stored topologies and/or metadata and, based on that comparison, executes a machine learning algorithm. The output of that machine learning algorithm includes predicted future changes to the topology, which the system uses to reduce the number of dimensions within the topology.

Abstract: The present invention relates to an apparatus, method and system for making web pages accessible to users with some type of hearing disability using sign language associated with a geolocated region from where a user accesses a URL site. The method comprises the steps of: receiving an input language in the form of text or words to be transformed into an output language in visual form; transmitting the input language to a web server coupled to a knowledge database comprising a plurality of dictionaries according to the language of the URL site and the sign language related to said language to interpret the input language to the output language; and reproducing the output language in the form of a video clip. The system of the present invention comprises at least one electronic device in communication with a web server coupled to a knowledge database to interpret words/text into sign language.

Abstract: In a cloud browser system, an image forming apparatus stores user data previously acquired from an image generation server and transmits a uniform resource locator (URL) of a Web page and the stored user data to the image generation server. The image generation server receives the URL and the user data, accesses the Web page based on the URL and the user data, and generates a rendering image by interpreting the Web page. The image forming apparatus receives the rendering image corresponding to a result acquired by accessing the Web page by using the user data from the image generation server.

Abstract: A method of operating a microcontroller to perform a Fast Fourier Transform, the method including receiving, by the microcontroller, N samples from a signal; and performing, by the microcontroller, a first butterfly operation of the Fast Fourier Transform before all of the N samples have been received from the signal, based on the performing of the first butterfly operation, the microcontroller performs the Fast Fourier Transform at a higher performance to power efficiency than a Fast Fourier Transform operation that begins after all of the N samples are received.

Abstract: Disclosed is a conjugate gradient acceleration apparatus using band matrix compression in depth fusion technology including a band matrix conversion unit configured to convert an adjacency matrix for correcting depth data acquired from data of an image sensor through deep learning based on depth information acquired from a depth sensor into a band matrix using rows of the adjacency matrix as addresses of query points and columns of the adjacency matrix as the nearest neighbors at the query points, a band matrix compression unit configured to mark an index on each band in order to compress the band matrix and to compress data, a memory unit configured to store tile data of the band matrix, and a band matrix calculation unit configured to perform computation of the band matrix and a transposed band matrix or computation of a symmetric band matrix with respect to the band matrix and a vector.

Abstract: Systems, methods, and apparatuses relating to a matrix operations accelerator are described.

Abstract: A hardware accelerator is configured to perform matrix multiplication and/or additional operations to optimize keypoint matching. A sum of squared error (SSE) calculation may be determined by utilizing the hardware accelerator to perform matrix multiplication to obtain a cost matrix for two sets of keypoint descriptors from two images. The hardware accelerator may determine a best cost calculation for each keypoint in each direction, which is utilized to perform keypoint matching.

Abstract: Disclosed embodiments relate to accelerating multiplication of sparse matrices. In one example, a processor is to fetch and decode an instruction having fields to specify locations of first, second, and third matrices, and an opcode indicating the processor is to multiply and accumulate matching non-zero (NZ) elements of the first and second matrices with corresponding elements of the third matrix, and executing the decoded instruction as per the opcode to generate NZ bitmasks for the first and second matrices, broadcast up to two NZ elements at a time from each row of the first matrix and each column of the second matrix to a processing engine (PE) grid, each PE to multiply and accumulate matching NZ elements of the first and second matrices with corresponding elements of the third matrix. Each PE further to store an NZ element for use in a subsequent multiplications.

Abstract: Methods and apparatus for performing diversity matrix operations within a memory fabric. Various embodiments of the present disclosure are directed to converting a memory array into a matrix fabric for spatial diversity-related matrix transformations and performing matrix operations therein. Exemplary embodiments described herein perform MIMO-related matrix transformations (e.g., precoding, beamforming, or data recovery matrix operations) within a memory device that includes a matrix fabric and matrix multiplication unit (MMU). In one variant, the matrix fabric uses a “crossbar” construction of resistive elements. Each resistive element stores a level of impedance that represents the corresponding matrix coefficient value. The crossbar connectivity can be driven with an electrical signal representing the input vector as an analog voltage. The resulting signals can be converted from analog voltages to a digital values by an MMU to yield a matrix-vector product.

Abstract: A goal of this invention is to minimize the cost of shifting a circadian state or entraining a state to a target cycle by identifying a preferred zeitgeber stimulus. Another goal of this invention is to provide a method to determine a person's circadian state trajectory given a stimulus time series.

Abstract: Various embodiments of the present invention disclose techniques for implementing an automatic data processing scheme for evaluating robust data sets to optimize procedure efficiency. A method may include determining a weighting-based input data object parameter for a robust data set; determining a variance-based input data object cohort parameter for a subset of the robust data set; generating a predictive variance metric for a data object of the subset based at least in part on the variance-based input data object cohort parameter and a first predictive metric attribute for the data object; generating a predictive weighting metric for the data object based at least in part on the weighting-based input data object parameter and a second predictive metric attribute for the data object; and generating a predictive metric data object for evaluating the robust data set based at least in part on the predictive variance metric and the predictive weighting metric.