Xerox Corporation

Conduent Business Services, LLC

Andres Quiroz Hernandez

Yasmine Charif

Shanmuga-Nathan Gnanasambandam

Julien Jean Lucien Bourdaillet

Gueyoung Jung

David R Vandervort

Jie Lin

Zhiguo Li

Saurabh Singh Kataria

Shi Zhao

Julien Bourdaillet

Hyun Joo Kim

Haengju Lee

Jian Ni

William Voll

Gary Morey

David Cacciola

A system and method for characterizing textual data by generating a first data abstraction based on a set of textual data. The first data abstraction can be presented to a user, and the user can provide instructions to make changes to the first data abstraction to generate a second data abstraction. The textual data can be extracted and characterized from the set of textual data using the second data abstraction.

Methods and systems for automated data characterization and extraction

The present invention generally relates to systems and methods for visual process analysis. The disclosed techniques can include: obtaining a theoretical and an empirical process model, generating a theoretical process layout corresponding to the theoretical process model, where the theoretical process layout is generated using a layout algorithm, generating an empirical process layout corresponding to the empirical process model, where the empirical process layout is generated using the layout algorithm, superposing the empirical process layout onto the theoretical process layout, such that a superposition layout is generated, annotating the superposition layout based on ugliness indicators, such that an annotated superposition layout is generated, and causing the annotated superposition layout to be displayed.

Automatically enhanced visual process repair using process superposition and ugliness indicators

The present invention generally relates to systems and methods for executing scripts (a sequence of declarative operations) on large data sets. Some implementations store descriptions of previously-executed operations and associated input and output data sets. When executing similar operations on the same, a subset of, a superset of, or any fragment of data subsequently, some implementations detect duplication of operations and access previously-stored output data sets in order to re-use data and reduce the amount of execution, thus avoiding time-consuming duplicative computations.

Systems and methods for managing duplication of operations

Systems and methods of data analytics, which in various embodiments enable business analysts to apply certain machine learning and analytics algorithms in a self-service manner by binding them to generic business questions that they can be used to answer in particular domains. The general approach may be to define the application of an algorithm to solve specific problems (questions) for particular combinations of a business domain and a data category. At design time, the algorithm may be linked to canonical data within a data category and programmed to run with this canonical data set. At runtime, given a dataset and its category, and a business domain, a user may choose from the corresponding questions and the system may run the algorithm bound to that question.

Systems and methods of data analytics

Methods and systems for scalable extraction of episode rules using incremental episode tree construction in a multi-application event space comprise compiling events from multiple, different domain logs into in a universal log file, rolling domain-dependent and domain-independent windows through the universal log file to identify distinct event-pattern episodes, adding episodes to an episode tree data structure, pruning less frequent episodes from the episode tree, analyzing the episode tree to identify frequent episode rules, and applying the frequent episode rules to future interactions with users.

Methods and systems for identifying frequently occurring intradomain episodes and interdomain episodes in multiple service portals using average user session length

A method and system for processing informational items originating from a plurality of information sources into a derived document for topical analysis thereof. Informational items are collated from a one of the sources in accordance with a predetermined plurality of relevant attributes and a key property value of common to select ones of the relevant attributes. Informational items are then grouped from the plurality of sources associated with the key common property value to form a document, wherein the informational items therein are marked on the informational source thereof. The document is then analyzed for topical identification.

Multi-source contextual information item grouping for document analysis

A process definition is partitioned for execution in a system architecture that enables the communication and meta-orchestration of multiple distributed engines. The partitioning method creates separate scripts for each group (execution engine, computer, distributed computer, etc.) where each script has the same representation as the original control flow, but keeps local services and replaces remote services with data flow messages and synchronization points. This method ensures that the resulting process has the same result as the original process executed with a single engine. Additional advantages include: the number of partitions of the process is minimized to equal to the number of distributed engines; the communication between engines is minimized to only data flow messages; there is no dependency on a specific process representation such as BPMN; and reduced implementation complexity.

Method for automated generation of minimal partitioning of a process specification for supporting its distributed execution

A system and method for predicting the class of future customer calls to a call center. Saved call data is analyzed using a robust tokenizer of a computerized device. The tokenizer transforms a sequence of characters in a call summary field of the saved call data into a sequence of tokens. Tokenized call data is produced. Multiple maximum entropy (MaxEnt) models are created based on the tokenized call data, using the computerized device. The MaxEnt models produce a probability distribution of all classes for a next call to a call center. A conditional random field (CRF) classifier is trained with the MaxEnt models and information from the saved call data, using the computerized device. The CRF classifier uses chronologically ordered sequences of prior calls to the call center and predicts a class for a new call to the call center based on the saved call data. A call class prediction is produced for the new call received from a returning customer based on the CRF classifier and the MaxEnt model.

Predicting the class of future customer calls in a call center

A method and system for designing file replication schemes in file sharing systems consider node storage constraints and node up/down statistics, file storage costs, and file transfer costs among the nodes, user request rates for the files, and user specified file availability requirements. Based on these considerations, a systematic method for designing file replication schemes can be implemented. The method first determines the number of copies of the files to be stored in the system to achieve the desired goal (e.g., to satisfy file availability requirements, or to maximize the system hit rate), and then selects the nodes at which to store the file copies to minimize the total expected cost. The file replication scheme for a peer-to-peer file sharing system in a distributed and adaptive manner can scale to a large number of nodes and files and can handle changes in the user request pattern over time.

Growing community of inventors

Rochester, NY, United States of America

Andres Quiroz Hernandez