Abstract: A robot movement programming method and apparatus for effectively creating a program for operating a robot with respect to workpieces of identical or similar shape. Sample programs for specifying sequence patterns suitable for the robot operations to be carried out are prepared, and also conversion formulas for obtaining teaching point position data are prepared as functions of dimensional parameters. The sample programs and the conversion formulas associated therewith are registered in a pattern library. The operator selects from the pattern library a sample program representing the sequence pattern suited for the robot operation to be carried out and designates the sample program as a base for creating a robot movement program after editing it, if necessary. Then, the operator inputs various parameter values, watching a graphic image of the sequence pattern specified by the selected sample program.

Abstract: A chromosome is decoded by a decoding unit, and converted into parameters of a problem model calculation unit. In the problem model calculation unit, a controller executes a local search method unit, a GA search unit, or a stochastic search unit while suitably selecting any of them, so that a solution is generated. If a constraint violation is detected by a constraint violation determination unit during a solution generation process, an added part (a part which causes a constraint violation) is removed from a current solution by the constraint violation processing unit, and the solution generation process is continued.

Abstract: A method and apparatus are provided of adapting a dictionary into which a keyword effective for judging whether or not information provided from either one of an information recording medium or an information communication network is recorded preferentially. The information includes information data and one or more keywords attached to the information data. The method is implemented by providing a user with the information data, enabling the user to input a signal indicating that the provided information data is necessary or unnecessary; calculating a prediction value for predicting a user's necessity degree of each keyword attached to the provided information data on the basis of the signal inputted in the inputting step, and registering preferentially the keyword effective for the dictionary by using the prediction value.

Abstract: A multi-kernel neural network computing architecture configured to learn correlations among feature values 34, 38 as the network monitors and imputes measured input values 30 and also predicts future output values 46. This computing architecture includes a multi-kernel neural network array 14 with the capability to learn and predict in real time. The CIP 10 also includes a manager 16 and an input-output transducer 12 that may be used for input-output refinement. These components allow the computing capacity of the multi-kernel array 14 to be reassigned in response to measured performance or other factors. The output feature values 46 computed by the multi-kernel array 14 and processed by an output processor 44 of the transducer 12 are supplied to a response unit 18 that may be configured to perform a variety of monitoring, forecasting, and control operations in response to the computed output values.

Abstract: An occupant of a cubicle or office can activate transparent partitions to partially or completely enclose a cubicle and/or office to minimize noise, optimize privacy, lighting, air flow and temperature. This transparent partition is coated with a thin layer material connected to electrodes. Once the occupant decides that changing working conditions (e.g., minimize noise, optimize privacy, adjust lighting, air flow and/or temperature) is needed, the occupant can activate the thin transparency control layer by appropriate electronic control. Movable parts of the partitions may also be adjusted. The transparent partitions may be made darker or opaque. Movable partitions may be used to provide optimized privacy, noise level, lighting, air flow and temperature control. Conversely, the partitions may be made transparent and movable sections retracted to permit greater worker-to-worker interaction.

Abstract: A knowledge modeling and delivery system combines pre-configured, parameterized models of human resource knowledge with organization-specific and employee-specific data. It delivers personalized answers to employee and manager questions about benefit plans, payroll, human resource (HR) policies, procedures, and practices. The pre-configured, parameterized models can be created and maintained outside the organization, the organizational data are created and maintained by the organization, and personal data are accessed or read from existing human resource management systems (HRMS's). This threefold partitioning yields a system that can be quickly customized to an individual organization, yields detailed and accurate information for individuals within the organization, and can be maintained in a cost-effective manner.

Abstract: A telecommunications fault location and diagnostic system employs a remote test unit (RTU) to collect system parameter data. The RTU is operatively coupled to a trained neural network, which receives the system parameter data from the RTU. The neural network is trained using pre-screened historical fault data, which is stored in a database. Once trained, the neural network classifies the RTU data into one of a predetermined number of fault probabilities.

Abstract: A problem solver for generating an action sequence executed from an initial state to a goal state in a state space defined by a particular problem is disclosed, that comprises a cognitive distance learning unit learning a cognitive distance that represents a cost acted on the environment of the state space, the cost being spent in an action sequence executed from a first state in the state space to a second state that is different from the first state, and a next action deciding unit deciding a next action contained in the action sequence that has to be executed in a particular state to attain the goal state in the state space based on the cognitive distance learnt by the cognitive distance learning unit.

Abstract: A method, apparatus, and article of manufacture for providing a description of logic used in determining an outcome based on automatic processing of data is disclosed. The method comprises the steps of hierarchically applying a series of decision criteria to the data to arrive at the outcome, while recording a rule determined from application of each decision criteria to the data, and retrieving the recorded rules. The article of manufacture comprises a data storage device tangibly embodying instructions to perform the method steps outlined above, and the apparatus for performing the method steps.

Abstract: To identify the best choice from a set of alternatives, team members select the criteria to be used in identifying the best choice. Each criterion describes at least one feature of the alternatives. Each team member then, for each alternative, defines a target value for the feature of the criteria and sets a level of knowledge about the feature of the criteria. Using the target values and the team member's knowledge of each feature, each team member estimates how well the alternatives meet the target values for the features. The system then calculates an expected utility for each alternative based on the team member's estimations and a probability that each alternative is the best choice. Finally, based on the expected utility and probability that each alternative is best, a best choice is identified.

Abstract: A computer-implementable method of selecting which engine of a plurality of inference engines to use to predict the categories into which individuals fall, such as buyer/non-buyer, and produce forecast reports based on the predictions is disclosed. Training (known) sample data that categorizes individuals based on the individual's profile is sequentially applied to multiple inference engines to determine which engine is best based on a desired objective. Then, a classifier associated with the selected engine is used to analyze unknown sample data, create category predictions and produce forecast reports based on the predictions.

Abstract: A rule present between a plurality of data values is extracted through data mining to output the rule and the information showing the reliability of the rule. A set-division evaluation-value S-ratio and a t-value showing the clearness of division when dividing a set constituted of a plurality of data values into two subsets are defined as the information showing the reliability of a rule. S-ratio=((S1+S2)/2)/S0 wherein, S0 is sum of squares of purpose variables of set before divided, S1, S2 are sums of squares of purpose variables of each subset after divided.

Abstract: An algorithm for improving the probability of identifying the repair that will correct a fault aboard a machine, such as a locomotive. The invention utilizes historical fault log and repair data and further calculates the number of times a particular fault occurs in a given number of days and also the number of times a particular fault occurs on each day. Averages are calculated for these results and when the number of fault occurrences exceed some or more of those averages, then these faults are deemed statistically significant for subsequent processing.

Abstract: A system for helping a chemist to identify pharmacophoric mechanisms, based on a set of input data representing many chemical compounds. Given an input data set defining for each compound a feature characteristic and an activity characteristic, a computer agglomeratively clusters representations of the molecules based on their feature characteristics. The result of this process is a multi-domain pyramid structure, made up of a number of nodes each representing one or more molecules. For each node, the computer identifies a representative feature set (such as a largest substructure common among the molecules in the node) and a representative activity level (such as an average of the activity levels of the molecules in the node). The computer then provides as output to a chemist a description of all or part of the pyramid. This process thus converts a large set of raw data into an understandable and commercially useful form, which can assist the chemist in developing beneficial new pharmaceuticals.

Abstract: A method of constructing data classifiers in cases where the number of features is much larger than the number of training examples and it is critical to avoid overtraining, is particularly useful in recognizing computer objects containing or likely to contain a computer virus. The method provides a novel way of tolerating imperfectly classifiable data, of learning in time polynomial in the size of the input data, and of avoiding overtraining. Training is performed by constructing and solving a linear program (LP) or integer program (IP).

Abstract: A decision operation tree (“DO Tree”) according to the present invention is a logical tree representation of decision making rules and actions instituted according to those rules. In preferred embodiments of the invention, one or more DO Trees are tailored to serve a user's rating and billing system needs. According to the present invention, Do Trees are comprised of a plurality of nodes and branches, which branches connect the nodes together in a descending relationship. Any node may have defined for it one or more actions (whose implementation can be prioritized or ordered) which are triggered whenever the node is reached. Those nodes having branches stemming (“descending”) from them have two parameters, a Decision Attribute and a Branching Type, which defines the node's relationship with its descending branches. Upon occurrence of the predefined event, the method includes proceeds from node to node, and optionally performs one or more actions in each node.

Abstract: A method, system and program product 100 usable by domain developers having any experience level in creating domain models. A representation of domain model knowledge is derived from a domain specification. The domain specification includes multiple potential domain objects, e.g., tables of APIs functional arguments, and each of the potential domain objects include one or more attributes. Potential domain objects are selected one at a time 102 from the specification and offered to the developer. The developer decides 104 whether or not to include the potential domain object in the domain model. If the developer decides to include the potential domain object 106, then the system provides a default name 108, i.e., the table name or argument name, and allows the developer to rename the selected domain object 110. Then, after having selected the object, potential attributes 112, e.g., table columns 1122, are selected from the object and offered to the developer 116.

Abstract: An integrated circuit comprising a logic processor and a fuzzy logic coprocessor is disclosed which processes a plurality of analog inputs. The logic processor and fuzzy logic processor are combined in the form of a single integrated circuit. The integrated circuit accepts a plurality of analog inputs which are digitized and provided as output to a display peripheral or are used to control an actuator peripheral such as a control unit for a valve. The integrated circuit includes means for loading or exchanging informational elements with other units of an installation.

Abstract: A method and apparatus for simulating human intelligence and natural language dialog capability is disclosed. The present invention contains a cognitive model of human intelligence (20), a mathematical model of information abstraction, synthetic dialog interaction (202), a method of language-independent computer learning through training (201), interaction and document reading (203) and a method of efficient computer implementation (200) of all preceding parts. The cognitive model (20) is the theoretical basis of the entire invention, describes the way humans learn and interact in general terms, provides a mathematical basis for natural language (40) learning and interaction and establishes a basis for detailed computer implementation of the theory.

Abstract: An automated interface program designed to interact and communicate with users, is disclosed that defines a list of categories activatable by the program; identifies a set of categories activated by user input; selects another set of categories from the activated categories based upon a metric and thereafter executes actions associated with the selected categories. The metric is a computed value based on conditions located within each category. The metric gives an estimate of the appropriateness of the particular response included in an activated category. This estimate is currently based on the current input and the current state of the automated interface program at the time user input is entered.