Abstract: A sequence of measured quantities is determined for a chemical reactor and a data generator generates a curve in normal coordinates from a respective predetermined normal form that describes a type of critical state. Each normal form is imaged onto the sequence of measured quantities by neural networks whereby a respective neural network is allocated to a data generator. The imaging is optimized by applying parameter optimization methods. The neural network that converges best through use of parameter optimization method describes the critical state that lies closest to the actual state of the chemical reactor.

Abstract: An event composing and monitoring system that allows high-level events to be created from combinations of low-level events. An event specification tool, contained in the system, allows for rapidly developing a general event processor that creates high-level events from combinations of user actions. An event system, in combination with an inference system, monitors and infers, for various purposes, about several classes of events including: current program context; state of key data structures; user input sequences, including actions with a mouse-controlled cursor while interacting with a graphical user interface; words typed in free-text help queries; visual user information, such as gaze and gesture information; and user speech information.

Abstract: A system is disclosed that provides a goal based learning system utilizing a rule based expert training system to provide a cognitive educational experience. The system provides the user with a simulated environment that presents a business opportunity to understand and solve optimally. Mistakes are noted and remedial educational material presented dynamically to build the necessary skills that a user requires for success in the business endeavor. The system utilizes an artificial intelligence engine driving individualized and dynamic feedback with synchronized video and graphics used to simulate real-world environment and interactions. Multiple "correct" answers are integrated into the learning system to allow individualized learning experiences in which navigation through the system is at a pace controlled by the learner.

Abstract: A method of predicting a desired property, such as thickening time, of a cement slurry comprising measuring or determining other properties of a slurry composition and obtaining a predicted value of the desired property by applying values representative of the other properties to a model formed by determining the other properties for a series of cement compositions and correlating these with measured values of the desired property. In one embodiment the method comprises measuring and determining the other properties and inputting values corresponding to these properties to a neural network device configured to output a value representative of the desired property, the neural network device being a) configured to utilize each of said values as input values, b) provided with at least one hidden layer of at least one node, and c) trained with a dataset comprising series of values of said properties and a value corresponding to the desired property when measured for a slurry having the measured properties.

Abstract: A semantic attractor memory uses an evolving neural network architecture and learning rules derived from the study of human language acquisition and change to store, process and retrieve information. The architecture is based on multiple layer channels, with random connections from one layer to the next. One or more layers are devoted to processing input information. At least one processing layer is provided. One or more layers are devoted to processing outputs and feedback is provided from the outputs back to the processing layer or layers. Inputs from parallel channels are also provided to the one or more processing layers. With the exception of the feedback loop and central processing layers, the network is feedforward unless it is employed in a hybrid back-propagation configuration. The learning rules are based on non-stationary statistical processes, such as the Polya process or the processes leading to Bose-Einstein statistics, again derived from considerations of human language acquisition.

Abstract: A learning method for performing a method of inductively learning rules from machine data, and a method for analyzing an inference process to learn rules useful for increasing efficiency by a common program are provided. Data desired to be classified is converted into a graph and is then input. The course of process by a computer is also converted into a graph and is then input. Then, patterns appearing in the input graph are extracted and patterns having a good evaluation result are output. The patterns are converted into rules for classification and rules for high-speed operation in accordance with a kind of the input graph, and are output.

Abstract: A method implemented on a network of computers for allowing a plurality of users to collaborate in decision making to solve a problem, includes the steps of processing, at a central server, a group of proposals for solving the problem, submitting the group of proposals to the plurality of users at remote clients, and thereafter receiving, at the central server from the plurality of users, selections of proposals from the group of proposals. Determining, at the central server, a narrowed group of proposals in response to the selections of proposals, communicating the narrowed group of proposals to the plurality of users at the remote clients, each proposal in the narrowed group of proposals including sets of statements, and thereafter receiving, at the central server from one of the plurality of users, suggestions for modifying proposals in the narrowed group of proposals, are also included.

Abstract: A problem solving unit obtains a solution in a symbol process in response to a given problem. A neural network learning control unit makes a neural network unit perform a learning process on a solution output from the problem solving unit. After completing the learning process in response to the given problem, the neural network unit provides an output as a hint on solving the problem to the problem solving unit.

Abstract: A delta learning system takes as an initial fault hierarchy (KB.sub.0) and a set of annotated session transcripts, and is given a specified set of revision operators where each operator within a group maps a fault hierarchy (KB) to a slightly different or revised fault hierarchy (.theta..sub.i (KB)). The revised fault hierarchy (.theta..sub.i (KB)) is called a neighbor of the fault hierarchy (KB), and a set of all neighbors (N(KB)) is considered the fault hierarchy neighborhood. The system uses the revision operators to hill climb from the initial fault hierarchy (KB.sub.0), through successive hierarchies (KB.sub.1 . . . KB.sub.m), with successively higher empirical accuracies over the annotated session transcripts. The final hierarchy (KB.sub.m), is a local optimum in the space defined by the revision operators. At each stage, to go from a fault hierarchy (KB.sub.i) to its neighbor (KN.sub.i+1), the accuracy of the fault hierarchy (KB.sub.

Abstract: The present invention discloses a method and system for developing a time horizon based financial model for investing towards attaining at least one financial objective. Each financial objective is characterized by an event having a financial amount and a future time when the financial amount is expected to be needed. In the present invention, a time horizon is specified in accordance with the financial objectives. In addition, the time horizon is divided into a plurality of segments. An array of future expenses for each of the financial objectives is then provided. The array indicates recurring costs of a specified amount, frequency of the costs, and duration of the costs. A fraction of the future expenses accruing in each time horizon segment is then identified. An investment of a specified amount of money is then modeled during each of the time horizon segments for satisfying the future expenses. The amount of money invested may change as the time horizon changes from segment to segment.

Abstract: Each of a plurality of units 2 has the learning section 9 which updates the parameters PRM and performs learning. The learning means 9 modifies the parameters PRM based on the error signal OS' propagated from one unit to another. To build a system consisting of such units 2, the user uses the GUI or some other method to select the units 2 as well as the signal paths for connecting the units via the user interface, displays the system configuration status, and adjusts the functions of units 2. Therefore, an easy-to-build data processing system and a system-build system and method used to build such a system can be provided.

Abstract: A computer-based method for assisting a user in making decisions in the process of completing a task is shown and described. The method of the system includes providing a set of questions for completing the task, asking the user a first question from the set, and providing a choice of answers to the first question. The system also provides advice from its knowledge base for deciding which of the answers to the first question to select. The choice of answer, however, remains under the control of the user, who can exercise judgment based on the user's expertise and the advice provided by the decision management system. The method continues by asking following questions, with the following questions, answers and advice influenced by the user's previous answers. Throughout the decision-making process, the user retains control of the answers if so desired, with the system providing advice from its knowledge base.

Abstract: A system for assisting a user in making decisions on a task through the asking of questions about the task and providing answer choices to the questions for the user. The system includes a set of question procedures which include a reference to another procedure and also answer procedures which include references to other procedures. Questions are dynamically generated by executing a question procedure which, in turn, calls a referenced procedure to generate a result based on one or more previous answers by the user. The referenced procedure passes the result to the question procedure for use in generating the question. The same process is used for dynamically generating answer choices for the questions. The referenced procedures may include, among other things, fetches of previous answers, conditional statements that choose among possible results based on a previous answer, or references to still other procedures. With this system, the questions and answers are kept current as the user moves through them.

Abstract: A polishing apparatus comprises a polishing member for polishing a workpiece and a manipulator for manipulating the workpiece and having a base disposed at a coordinate origin of a base coordinate system. The polishing member and workpiece define therebetween a contact surface having a contact surface coordinate system relative to an initial coordinate system defined when the polishing member and the workpiece initially contact one another at the contact surface but before a force is exerted on the contact surface to polish the workpiece. A first control device successively updates a position and attitude of the contact surface coordinator system with respect to the initial coordinate system on the basis of a deviation between preselected values of force and moment acting on the workpiece and detected values of force and moment acting at the contact surface between the polishing member and the workpiece when a force is exerted on the contact surface to polish the workpiece.

Abstract: In an apparatus and a method for detecting and assessing a spatially discrete dot pattern disposed in a multidimensional coordinate system, each dot in the pattern assumes at least two differentiatable status values. A measuring device records the coordinate values and status values of each dot of the multidimensional spatial dot pattern. A memory stores data that correspond to the recorded coordinate values and status values of each dot of the multidimensional spatial dot pattern. A computer into which the stored data are entered and in which a coordinate counter for each coordinate value of a coordinate axis is determined from the stored data, is associated with the memory. The value of the coordinate counter is formed from the number of detected dots of the coordinates that have a predetermined status value. A neural network is associated with the computer.

Abstract: An automatic data flow architecture builder is disclosed that is operable to take raw data which is stored in a raw data buffer (60) and transform it for storage in a transformed data buffer (78). A plurality of transform blocks (68), (70) and (72) are provided for this transform operation, these disposed in a predetermined data flow. When the user inputs a new transform to be disposed within the data flow of the transforms, rules are applied via a rule-base processing system (74) to apply a set of predetermined rules in a rule database (76) to the transform. These rules determine where the transform is to be inserted. This provides an automatic construction operation of a data flow architecture.

Abstract: Non-uniform filters are designed with a method which utilizes an orthogonal subbanding technique. A tiling representation of a signal is established. The signal is localized for both time and frequency using a mixed tiling representation of the signal. A prototype is then established using the joint time frequency representation of the signal. Lastly, non-uniform filters are developed utilizing the prototype.

Abstract: An inherently, and massively parallel data processing system and methodology is disclosed for evolving relative awareness, and related response states in a computerized environment. Any number of disparate data types can be configured with the goal of evolving complex data patterns involving these data forms across relative time and space. New data processing principles and a methodology are provided based on a new default logic that emulates the logic of the force of gravity as applied to proximity in time, space and form. Data form standards are provided that require time, place, and form factors at every convergent processing step. This system is capable of evolving complex awareness states as represented by ever larger associations of the original data samples without the need for knowledge bases, conditional rules or the like. Patterns are discerned in an emergent manner.

Abstract: Memory and storage method in an electronic controller operating with fuzzy logic procedures for membership functions (FA) of logical variables (M) defined in a so-called discourse universe (U) discretized at a finite number of points (m) which provide memorization of triangular or trapezoid membership functions (FA). Memory words have a first portion for codification of the vertex of the membership function (FA), a second portion for a codification corresponding to the slope of one side of the membership function (FA), and a third portion for a codification corresponding to the slope of the other side of the function.

Abstract: A multi-mode oriented polynomial transformation based defuzzification strategy uses a Kalman filter in a parameter leaning procedure defuzzification application where two or more distinct possibility peaks exist in its membership function distribution.