Ut-battelle, Inc.

Planet Technologies

Thomas Eugene Potok

Robert Matthew Patton

Mark Thomas Elmore

Joel Wesley Reed

Jim N Treadwell

Nagiza Faridovna Samatova

Nenad Ivezic

Thomas P Karnowski

Steven R Young

Derek C Rose

Sreenivas R Sukumar

J Travis Johnston

Justin M Beaver

Barbara G Beckerman

Chad Allen Steed

Rob Gillen

Yu Jiao

Robert E Gillen

Christopher T Symons

Carlos C Rojas

Seung-Hwan Lim

A system and method design and optimize neural networks. The system and method include a data store that stores a plurality of gene vectors that represent diverse and distinct neural networks and an evaluation queue stored with the plurality of gene vectors. Secondary nodes construct, train, and evaluate the neural network and automatically render a plurality of fitness values asynchronously. A primary node executes a gene amplification on a select plurality of gene vectors, a crossing-over of the amplified gene vectors, and a mutation of the crossing-over gene vectors automatically and asynchronously, which are then transmitted to the evaluation queue. The process continuously repeats itself by processing the gene vectors inserted into the evaluation queue until a fitness level is reached, a network's accuracy level plateaus, a processing time period expires, or when some stopping condition or performance metric is met or exceeded.

Optimizing neural networks

A system and method for document recommendation combines linguistic and term frequency approaches in order to improve overall performance and recommendation. Hypernyms are added to the source document set so that related documents can be recommended from the source document set. There are a variety of different parameters to configure that change operation of the document recommendation system.

System and method of content based recommendation using hypernym expansion

Disclosed are several examples of systems that can read millions of news feeds per day about topics (e.g., your customers, competitors, markets, and partners), and provide a small set of the most relevant items to read to keep current with the overwhelming amount of information currently available. Topics of interest can be chosen by the user of the system for use as seeds. The seeds can be vectorized and compared with the target documents to determine their similarity. The similarities can be sorted from highest to lowest so that the most similar seed and target documents are at the top of the list. This output can be produced in XML format so that an RSS Reader can format the XML. This allows for easy Internet access to these recommendations.

Method and system to discover and recommend interesting documents

A computer implemented method detects intrusions using a computer by analyzing network traffic. The method includes a semi-supervised learning module connected to a network node. The learning module uses labeled and unlabeled data to train a semi-supervised machine learning sensor. The method records events that include a feature set made up of unauthorized intrusions and benign computer requests. The method identifies at least some of the benign computer requests that occur during the recording of the events while treating the remainder of the data as unlabeled. The method trains the semi-supervised learning module at the network node in-situ, such that the semi-supervised learning modules may identify malicious traffic without relying on specific rules, signatures, or anomaly detection.

In-situ trainable intrusion detection system

Disclosed is a method and system for discovering documents using a computer and providing a small set of the most relevant documents to the attention of a human observer. Using the method, the computer obtains a seed document from the user and generates a seed document vector using term frequency-inverse corpus frequency weighting. A keyword index for a plurality of source documents can be compared with the weighted terms of the seed document vector. The comparison is then filtered to reduce the number of documents, which define an initial subset of the source documents. Initial subset vectors are generated and compared to the seed document vector to obtain a similarity value for each comparison. Based on the similarity value, the method then recommends one or more of the source documents.

Method and system of filtering and recommending documents

A cloud computing platform includes first device having a graph or tree structure with a node which receives data. The data is processed by the node or communicated to a child node for processing. A first node in the graph or tree structure determines the reconfiguration of a portion of the graph or tree structure on a second device. The reconfiguration may include moving a second node and some or all of its descendant nodes. The second and descendant nodes may be copied to the second device.

Cloud computing method for dynamically scaling a process across physical machine boundaries

Medical reports are converted to document vectors in computing apparatus and sampled by applying a maximum variation sampling function including a fitness function to the document vectors to reduce a number of medical records being processed and to increase the diversity of the medical records being processed. Linguistic phrases are extracted from the medical records and converted to s-grams. A Haar wavelet function is applied to the s-grams over the preselected time interval; and the coefficient results of the Haar wavelet function are examined for patterns representing the likelihood of health abnormalities. This confirms certain s-grams as precursors of the health abnormality and a parameter can be calculated in relation to the occurrence of such a health abnormality.

Method and system for determining precursors of health abnormalities from processing medical records

The method and system of the invention involves processing each new document () coming into the system into a document vector (), and creating a document vector with reduced dimensionality () for comparison with the data model () without recomputing the data model (). These operations are carried out by a first computer () while a second computer () updates the data model (), which can be comprised of an initial large group of documents () and is premised on the computing an initial data model () to provide a reference point for determining document vectors from documents processed from the data stream ().

Dynamic reduction of dimensions of a document vector in a document search and retrieval system

A computer method and system for storing, retrieving and displaying information has a multiplexing agent () that calculates a new document vector () for a new document () to be added to the system and transmits the new document vector () to master cluster agents () and cluster agents () for evaluation. These agents () perform the evaluation and return values upstream to the multiplexing agent () based on the similarity of the document to documents stored under their control. The multiplexing agent () then sends the document () and the document vector () to the master cluster agent (), which then forwards it to a cluster agent () or creates a new cluster agent () to manage the document (). The system also searches for stored documents according to a search query having at least one term and identifying the documents found in the search, and displays the documents in a clustering display () of similarity so as to indicate similarity of the documents to each other.

Agent-based method for distributed clustering of textual information

A computer method of gathering and summarizing large amounts of information comprises collecting information from a plurality of information sources () according to respective maps () of the information sources (), converting the collected information from a storage format to XML-language documents () and storing the XML-language documents in a storage medium, searching for documents () according to a search query () having at least one term and identifying the documents () found in the search, and displaying the documents as nodes () of a tree structure () having links () and nodes () so as to indicate similarity of the documents to each other.

Method for gathering and summarizing internet information

A method of analyzing computer intrusion detection information that looks beyond known attacks and abnormal access patterns to the critical information that an intruder may want to access. Unique target identifiers and type of work performed by the networked targets is added to audit log records. Analysis using vector space modeling, dissimilarity matrix comparison, and clustering of the event records is then performed.

Method for detecting sophisticated cyber attacks

System for gathering and summarizing internet information

A method for distributed agent based non-expert simulation of manufacturing process behavior on a single-processor computer comprises the steps of: object modeling a manufacturing technique having a plurality of processes; associating a distributed agent with each the process; and, programming each the agent to respond to discrete events corresponding to the manufacturing technique, wherein each discrete event triggers a programmed response. The method can further comprise the step of transmitting the discrete events to each agent in a message loop. In addition, the programming step comprises the step of conditioning each agent to respond to a discrete event selected from the group consisting of a clock tick message, a resources received message, and a request for output production message.

Method for distributed agent-based non-expert simulation of manufacturing process behavior

A method for automatically evaluating a manufacturing technique comprises the steps of: receiving from a user manufacturing process step parameters characterizing a manufacturing process; accepting from the user a selection for an analysis of a particular lean manufacturing technique; automatically compiling process step data for each process step in the manufacturing process; automatically calculating process metrics from a summation of the compiled process step data for each process step; and, presenting the automatically calculated process metrics to the user. A method for evaluating a transition from a batch manufacturing technique to a lean manufacturing technique can comprise the steps of: collecting manufacturing process step characterization parameters; selecting a lean manufacturing technique for analysis; communicating the selected lean manufacturing technique and the manufacturing process step characterization parameters to an automatic manufacturing technique evaluation engine having a mathematical model for generating manufacturing technique evaluation data; and, using the lean manufacturing technique evaluation data to determine whether to transition from an existing manufacturing technique to the selected lean manufacturing technique.

Growing community of inventors

Oak Ridge, TN, United States of America

Thomas Eugene Potok