Abstract: A system and a method for managing various Customer Productivity Applications utilize a web-enabled interactive database to organize, store and retrieve information for the benefit of the customer community to effectively manage the business. The system captures all customers related information and provides on-line, up-to-date information to the users upon request. The system is configured to manage various Customer Productivity Applications to increase the customers' productivity and minimizes cost of ownership by analyzing the customers' fleet, perform industry comparisons, and run iterations to develop fleet optimal performance requirements.

Abstract: The centerpiece of this invention is clusters. Clusters include taste signatures, which represent the tastes of a cluster. Users are enabled to find clusters that match their tastes, and artists are enabled to find the clusters where the users are who are likely be interested in their work. Clusters thus serve as hubs of activity for particular tastes; in most embodiments ratings of items of interest to those tastes can be viewed, and various embodiments may include various means for inter-user communication so that communities of people with similar tastes are formed.

Abstract: A computer-implemented method for managing a plurality of customer relationship plans may include a step of collecting profile check values from a database of customer data, the collected profile check values including customer data corresponding to profile checks for each of the plurality of customer relationship plans. The collected profile check values may then be applied to predefined selection criteria in one or more of the plurality of customer relationship plans. The selection criteria determines whether a customer relationship plan of the plurality of customer relationship plans applies to a customer. One or more customer relationship plans may then be assigned or de-assigned to each of the selected customers, depending upon whether the applied profile check values satisfy or fail to satisfy the predefined selection criteria, respectively.

Abstract: A system and method is provided for improved reporting of supplier performance. A system receives and processes data to enable reporting of on-time performance relative to a plurality of order start points and a plurality of order end points. Buyers and suppliers report on-time performance for system processing using the delivery start points and end points collected in the normal course of their business operations. System users may obtain supplier on-time performance reports indicating the on-time performance of suppliers relative to the plurality of start point/end point pairs reflected in the collected data. A system and method is also provided for improved reporting of reject performance. A system stores collected information predictive of whether an order reject was supplier caused or buyer caused. The predicted cause of order rejects is reflected in supplier reject performance as reported to system users.

Abstract: A collection request inputted at a consignor destination system is recorded on an electronic tag attached to freight and sent to a collection management system so as to be received. During collection of freight, information concerning collection is recorded on the electronic tag by means of a terminal unit and the collection information is sent to the collection management system. Anin-base station freight information management system informs a freight information overall management system of the collection information and information concerning freight passing through the base station. The freight information overall management system acquires a transport history of the freight on the basis of the information sent from the in-base station freight information management system, calculates a delivery schedule of the freight and informs it to each in-base station freight information management system.

Abstract: A method and system for ascribing a reputation of an entity as a rater of other entities is provided. A first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, are provided. The second ratings are combined to produce a first combined rating. The first rating is compared to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating. A resulting rater reputation is generated based at least in part on the first rating predictability.

Abstract: An artificial neural network performs error correction on an input signal vector. The input signal vector is process in a forward direction through synapses in each of a plurality of neurons for providing an output signal from each of the neurons. The output signals from the neurons are monitored until the one having the greatest activity level is identified. A reverse flow signal having a predetermined magnitude is processed in the reverse direction through the neuron having the greatest activity level for updating the input signal vector. Alternately, the output signals of competing neurons may be applied through synapses weighted to favor the neuron having the greatest output signal activity. Thus, the neuron with synapses most closely matched to the elements of the input signal vector overpowers the remaining neurons and wins the competition.

Abstract: A self-organizing adaptive replicated (SOAR) for creating a replicate of human expert behavior. The SOAR can be embedded invisibly within multiple types of systems to observe, adapt and grow to emulate a user's interactive behavior and performance level. The system yields near equivalent responses to near equivalent stimuli in real time. The SOAR is based on a three layer perceptron type architecture which guarantees arbitrary M to N mapping of continuous valued spaces. The architecture uses a competitive, additive, and layer independent learning rule which insures excellent rapid learning. A self-organizing, adaptive algorithm permits the SOAR to adapt to the true classification space. The SOAR has applications in areas such a speech recognition, target detection, pattern recognition of multi-feature data, electro-mechanical subsystem control and resource allocation and optimization.

Abstract: A digital encoding process for transmitting and/or storing acoustical sigs and, in particular, music signals, in which scanned values of the acoustical signal are transformed by means of a transformation or a filter bank into a sequence of second scanned values, which reproduce the spectral composition of the acoustical signal, and the sequence of second scanned values is quantized in accordance with the requirements with varying precision and is partially or entirely encoded by an optimum encoder, and in which a corresponding decoding and inverse transformation takes place during the reproduction. An encoder is utilized in a manner in which the occurrence probability of the quantized spectral coefficient is correlated to the length of the code in such a way that the more frequently the spectral coefficient occurs, the shorter the code word.

Abstract: Method and device having a neural network for classifying data, and verification device for signatures.The device includes a neural network with an input layer 3, an internal layer 4, and an output layer 5. This network is designed to classify data vectors to classes, the synaptic weights in the network being determined through programming on the basis of specimens whose classes are known. Each class is defined during programming as corresponding to a set of neurons of which each represents a domain which contains a fixed number of specimens. The network includes a number of neurons and synaptic weights which have been determined as a function of the classes thus defined.

Abstract: A defuzzifier circuit which is used to obtain a manipulated variable for controlling an actuator by using fuzzy information appearing over a plurality of lines. The defuzzifier circuit can share the membership function with a plurality of converting mechanisms and enables a weight to be set externally.

Abstract: A fuzzy inference engine (10) performs fuzzy logic operations with a high degree of accuracy in a minimal amount of time. The fuzzy inference engine (10) includes a fuzzification module (12) which decodes an input signal to access a memory location (18, 20). When accessed, the memory location provides a unary value to a rule evaluation module (26). The rule evaluation module (26) subsequently processes the unary value to indicate a relative strength of a fuzzy inference rule. Because the fuzzy logic operations use unary numbers, rather than more traditional binary numbers, the only time required to perform fuzzy logic operations is equal to only a time required for the signal to propagate through the logic gates (28, 30, 32, 34, 36, and 38) forming rule evaluation module (26).

Abstract: In automatic speech recognition, confusion easily arises between phonetically similar words (for example, the German words "zwei" and "drei") in the case of previous recognition systems. Confusion of words which differ only in a single phoneme (for example, German phonemes "dem" and "den") occurs particularly easily with these recognition systems. In order to solve this problem, a method for recognizing patterns in time-variant measurement signals is specified which permits an improved discrimination between such signals by reclassifying in pairs. This method combines the Viterbi decoding algorithm with the method of hidden Markov models, the discrimination-relevant features being examined separately in a second step after the main classification. In this case, different components of feature vectors are weighted differently, it being the case that by contrast with known approaches these weightings are performed in a theoretically based way.

Abstract: A neural processor, comprising neural calculation apparatus (30, NQ, RQ) which extracts a root Q of a quantity X, said root constituting either a norm of a data or a distance between data. The calculation apparatus calculates (30) by iteration a series of contributions .DELTA.Q.sub.i which are used (NQ, RQ) to update a partial root QP which becomes the root Q at the end of calculation. The calculation can be performed on an arbitrary arithmetic base which determines the number of neurons utilized and also the accuracy of calculation. It is possible to execute the calculation of a partial remainder (NR, RR). Several programming modes are presented.

Abstract: A neural processor, comprising neural calculation apparatus (30, NQ, RQ) which extracts a root Q of a quantity X, said root constituting either a norm of a data or a distance between data. The calculation apparatus calculates (30) by iteration a series of contributions .DELTA.Q.sub.i which are used (NQ, RQ) to update a partial root QP which becomes the root Q at the end of calculation. The calculation can be performed on an arbitrary arithmetic base which determines the number of neurons utilized and also the accuracy of calculation. It is also possible to execute the calculation of a partial remainder (NR, RR). Several programming modes are presented.

Abstract: A neural network learning system is applied to extensive use in applications such as pattern and character recognizing operations, various controls, etc. The neural network learning system operates on, for example, a plurality of neural networks each having a different number of intermediate layer units to efficiently perform a learning process at a high speed with a reduced amount of hardware. A neural network system having a plurality of hierarchical neural networks each having an input layer, one or more intermediate layers and output layers is formed from a common input layer shared among two or more neural networks, or the common input layer and one or more intermediate layers and a learning controller for controlling a learning process performed by a plurality of neural networks.

Abstract: This invention is a robot control system based on a high level language implementing a spatial operator algebra. There are two high level languages included within the system. At the highest level, applications programs can be written in a robot-oriented applications language including broad operators such as MOVE and GRASP. The robot-oriented applications language statements are translated into statements in the spatial operator algebra language. Programming can also take place using the spatial operator algebra language. The statements in the spatial operator algebra language from either source are then translated into machine language statements for execution by a digital control computer. The system also includes the capability of executing the control code sequences in a simulation mode before actual execution to assure proper action at execution time. The robot's environment is checked as part of the process and dynamic reconfiguration is also possible.

Abstract: In a fuzzy control apparatus in which a rule is capable of being changed during operation, a fuzzy reasoning output which prevails immediately prior to a rule change is stored when a rule change is made, and a fuzzy-control output is generated by combining, after the rule change, the stored fuzzy reasoning output which prevailed immediately prior to the rule change and a fuzzy reasoning output which prevails following the rule change. The ratio of the combination is altered in such a manner that, with the passage of time from the moment at which the rule change is made, there is a gradual decrease in a percentage of the fuzzy reasoning output which prevailed immediately prior to the rule change, and a gradual increase in a percentage of the fuzzy reasoning output which prevails following the rule change, in the fuzzy-control output.

Abstract: A fuzzy inference apparatus comprises a rule storing circuit for storing then-part positions respectively corresponding to first if-part membership functions and second if-part membership functions. The first and second if-part membership functions have shapes of triangles each having a base corresponding to positions of vertices of triangles of two adjacent membership functions.

Abstract: The invention involves a fuzzy inference development system which prepares, automatically, an inference engine module file and a knowledge (i.e., rules and membership functions) data file which are optimized for the system (the CPU or equivalent) which the user intends to use and for the knowledge established by the user. The user inputs the required specifications and knowledge data via a user interface. The characteristic parameters (number of rules, number of types of input and output variables, etc.) are extracted from the knowledge data. A number of inference engine modules are stored, prior to the use of the device, in an inference engine library. The inference engine module, which is both optimal for the required specifications which have been input and for the parameters which have been extracted, is selected from the modules in the library by means of a fuzzy inference performed in an inference engine selection unit.