6205 results about "Weight value" patented technology

An artificial neural network calculating device for sparse connection comprises a mapping unit used for converting input data into the storage mode that input nerve cells and weight values correspond one by one, a storage unit used for storing data and instructions, and an operation unit used for executing corresponding operation on the data according to the instructions. The operation unit mainly executes three steps of operation, wherein in the first step, the input nerve cells and weight value data are multiplied; in the second step, addition tree operation is executed, the weighted output nerve cells processed in the first step are added level by level through an addition tree, or the output nerve cells are added with offset to obtain offset-added output nerve cells; in the third step, activation function operation is executed, and the final output nerve cells are obtained. By means of the device, the problems that the operation performance of a CPU and a GPU is insufficient, and the expenditure of front end coding is large are solved, support to a multi-layer artificial neural network operation algorithm is effectively improved, and the problem that memory bandwidth becomes a bottleneck of multi-layer artificial neural network operation and the performance of a training algorithm of the multi-layer artificial neural network operation is solved.

Host computer systems dynamically engage in independent transactions with servers of a server cluster to request performance of a network service, preferably a policy-based transfer processing of data. The host computer systems operate from an identification of the servers in the cluster to autonomously select servers for transactions qualified on server performance information gathered in prior transactions. Server performance information may include load and weight values that reflect the performance status of the selected server and a server localized policy evaluation of service request attribute information provided in conjunction with the service request. The load selection of specific servers for individual transactions is balanced implicitly through the cooperation of the host computer systems and servers of the server cluster.

An artificial neural network device that utilizes one or more non-volatile memory arrays as the synapses. The synapses are configured to receive inputs and to generate therefrom outputs. Neurons are configured to receive the outputs. The synapses include a plurality of memory cells, wherein each of the memory cells includes spaced apart source and drain regions formed in a semiconductor substrate with a channel region extending there between, a floating gate disposed over and insulated from a first portion of the channel region and a non-floating gate disposed over and insulated from a second portion of the channel region. Each of the plurality of memory cells is configured to store a weight value corresponding to a number of electrons on the floating gate. The plurality of memory cells are configured to multiply the inputs by the stored weight values to generate the outputs.

A semantic search system using a semantic ranking scheme including: an ontology analyzer analyzing ontology data related to a search target to determine a weight value of each property according to a weighing method for property; a semantic path extractor extracting all the semantic paths between resources and query keywords and determining a weight value of each extracted semantic path according to the semantic path weight value determination scheme by using the weight value of each property; a relevant resource searcher traversing an instance graph of ontology based on a semantic path having a pre-set length and weight value of more than an expectation level to search resources that have a semantic relationship with the query keywords and are declared as a type presented in the query; and a semantic relevance ranker selecting a top-k results having the highest rank from among the candidate results extracted by the relevant resource researcher by using a relevance scoring function.

Method and apparatus for collecting and storing information about individuals in a social network. A member account is created by verifying a user's identity and collecting user identity data from the user. The user sends an invitation to one or more members or non-members. The one or more members or non-members can accept or decline the user's invitation. If the one or more non-members accept or decline, they must create a member account. The one or more members post paid comments or free comments about the user and one or more user weight values are calculated based on one or more of the following: a number of members in a user's group and their one or more user weight values, a total value of posted paid comments about the user and volume of posted comments about the user. The user can post a reply to the posted comment by the one or more members.

One embodiment of the present invention provides a system for facilitating use of a collation element that supports a large number of characters. The system operates by receiving the collation element and reading a primary weight value from a primary weight field within the collation element. If the primary weight value falls within a reserved set of values, the system reads an additional portion of the primary weight value from both a secondary weight field and a tertiary weight field within the collation element. On the other hand, if the primary weight value is not within the reserved set of values, the system reads a secondary weight value from the secondary weight field, and also reads a tertiary weight value from the tertiary weight field.

A method and system for mapping a number of characters in a string, wherein the string comprises a combination of characters representing indexed expressions and a combination of characters representing non-indexed expressions. One embodiment produces a weight array that can be utilized to compare a first and second string having indexed and non-indexed expressions. In one embodiment, a method generates a set of special weights for characters that represent indexed and non-indexed expressions. The method then associates a weight value of an indexed expression with the specific group of characters representing a specific non-indexed expression, and generates a weight array by retrieving a plurality of special weights associated with the specific group of characters representing the specific non-indexed expression and the associated weight value of the indexed expression.

A method and apparatus is provided for extracting key terms from a data set, the method includes identifying a first set of one or more word groups of one or more word that occur more than once in the data set, and removing from this first set a second set of word groups that are sub-strings of longer word groups in the first set. The remaining word groups are key terms. Each word group is weighted according to its frequency of occurrence within the data set. The weighting of any word group may be increased by the frequency of any sub-string of words occurring in the second set and then dividing each weighting by the number of words in the word group. This weighting process operates to determine the order of occurrence of the word groups. Prefixes and suffixes are also removed from each word in the data set. This produces a neutral form of each word so that the weighting values are prefix and suffix independent.

A system and method for employing a number of producer modules to produce and deliver recommendations to a requester. Each of the recommendations having associated therewith a confidence level. Each of the producer modules having associated therewith a weighting value. The confidence levels in each of the recommendations produced by the producers being modified based on weighting value associated with the producer that produced the produced recommendation. The weighing values associated with each of the producer modules being modified based on information from the requester related to the recommendations delivered to the requester.

A system that builds a network using a document collection wherein the documents are collected and represented as a plurality of nodes in a network matrix. The documents that are to be analyzed are bound to the network (corpus) at a discrete node corresponding to the document. The documents are then analyzed to determine term frequency within each document and the overall term frequency of the same term throughout the entire document grouping. This creates a weighting value that determines the relevancy of each document as compared to the entire network of documents. Finally, weighting values are normalized with relative weighting values so that the sum of the weights of all edges connected to a given node equals 1. User queries then proceed through the network from node to node using the algorithm of the present invention to locate documents relevant to the search.

Systems and methods are provided for multi-entity ontology weighting. In one implementation, a computer-implemented method is provided for identifying one or more target entities. The method receives an input data set. The input data set includes a plurality of tokens that reflect the one or more target entities. The method determines which of the plurality of tokens are labels, identifies a plurality of candidate entities that are related to each of the labels, assigns a weight to each of the plurality of candidate entities based on a frequency of each of the plurality of candidate entities, and ranks, for each of the labels, the plurality of candidate entities according to the assigned weights. The method then outputs, for each of the labels, the one of the plurality of candidate entities for that label that has been assigned a highest weight value.

The invention discloses a method for automatic correction and tiled display of plug-and-play large screen projections. The method comprises the steps as follows: adaptively generating a checker pattern with a certain resolution ratio, projecting by projectors in sequence, capturing by cameras, carrying out feature detection and identification on complicated projection surfaces or checkers under the illumination condition with a multi-feature detection method based on color and geometry, and creating a Bezier curve function to represent a corresponding relation of points between projector images and camera images; and obtaining effective display areas for screen projection with a quick approach method, determining a corresponding relation of the projection contents of the projectors and the display areas, carrying out geometrical distortion on the images to be projected for geometrical correction, and calculating weight values of pixels in projection overlapping areas with a distance-based nonlinear weight value distribution method for fusion of edge brightness. According to the invention, a plurality of projection images of irregular surfaces can be aligned and seamlessly spliced, and the whole method is simple and easy to use, higher in autonomy and better in seamless splicing performance.

A motion object detection method using an adaptive background model and a computer-readable storage medium are provided. In the motion object detection method, a background model establishing step is firstly performed to establish a background model to provide a plurality of background brightness reference values. Then, a foreground object detecting step is performed to use the background model to detect foreground objects. In the background model establishing step, a plurality of brightness weight values are firstly provided in accordance with the brightness of background pixels, wherein each of the brightness weight values is determined in accordance with the relative background pixel. Thereafter, the background brightness reference values are calculated based on the brightness of the background pixels and the brightness weight values. In addition, a computer can perform the motion object detection method after reading the computer-readable storage medium.

One embodiment of the present invention sets forth a technique for selecting a content distribution network (CDN) comprising at least one content server, from a plurality of CDNs, and a playing digital content file from the CDN on a content player. Selecting the CDN is based on a rank order of CDNs, an assigned weight value for each CDN, and a bandwidth measured between the content player and each CDN. Advantageously, a given content player may select a CDN based on prevailing network and CDN loading conditions, thereby increasing overall robustness and reliability when downloading digital content file from a CDN.

The invention relates to a method and a system thereof for dispatching application requests in a virtual calculation platform.The method comprises the following steps: (1) collecting the total resource quantity, the resource utilization rate and the connection number of the service requests of each virtual machine; (2) according to the current weight value, the total resource quantity, the resource utilization rate and the connection number of the service requests of the virtual machines, updating the weight value of the virtual machines; and (3) dispatching the service requests by adopting apolling dispatching method based on the weight value. The invention can realize the matching of resources and loads in the system of the virtual machines and enhance the utilization efficiency of theentire virtual calculation platform.

A system for rejecting wind noise at a plurality of sensors includes input logic, a processor and output logic. The input logic receives a signal from each of the plurality of sensors. The processor assigns a weight value to each of the received signals. The output logic derives a wind noise rejected output signal based on a function of the assigned weight values and the received signals.

A method and device for determining a parameter (RTO) associated with a retransmission time-out feature in a data unit transmitter implementing a protocol (TCP) that provides an acknowledgment feature and a retransmission time-out feature, where according to one aspect decreasing values of the round trip time (RTT) are treated differently with respect to their influence on a mean deviation parameter (RTTVAR) than increasing values of the round trip time, where according to another aspect weight values (g, h, w) used in the determination of the parameter (RTO) are chosen such that at least one of them is time variable, and where according to a further aspect the parameter (RTO) is also calculated on the basis of a value (SR) indicative of the number of spurious time-outs in the communication.

A system and method for generating weight values based on maximum data rate for weighting elements included within signal weighting and combining arrangements used in various multi-antenna transmitter and receiver structures is disclosed herein. Weighting values for a given signal combining arrangement are set so as to maximize an output data rate of the applicable multi-antenna system in the presence of adaptive bit loading of the subcarriers of a transmitted signal. The disclosed techniques may be employed to maximize a data rate of a multi-antenna communication system by using adaptive bit loading and RF and baseband weighting schemes. In this case a search is conducted over various combinations of RF and baseband weights in order to find the weight combination which, when adaptive bit loading is also employed, maximizes the data rate.

The invention provides an application recommendation method and an application recommendation system. The application recommendation method includes the following steps: access behavior data of multiple reference users to preset multiple applications are obtained, the multiple reference users are subjected to grouping according to the access behavior data, and each of the multiple applications has a corresponding application class; a group which a target user belongs to is determined according to access behavior data of the target user to the multiple applications, access weighted values of the target user to different application classes are subjected to statistics, and the application classes with the access weighted values which are in a preset range are regarded as interest application classes; and the applications which belong to the interest application classes are extracted from the multiple applications which are accessed by all the reference users and are in the group which the target user belongs to, and then the applications are recommended to the target user. The application recommendation method and the application recommendation system can occupy fewer system resources when the applications are recommended.

A system and method for generating a list of suggested scheduled television programs wherein television schedule guide data which comprises characteristics of scheduled programs and optionally relevancy of those characteristics is obtained from broadcast information. User profiles comprising characteristics of television programs previously viewed, the cumulative amount of time programs having those characteristics has been previously viewed, and optionally the relevancy of the previously viewed characteristics is stored. The user is allowed to manually adjust the weights assigned to characteristics so as to override the system assigned weight values. The system calculates the suggested program list from the schedules it receives from broadcast or other means, using the characteristic weights in the user profiles. The user interface displays percentage values currently assigned to characteristics in that user's profile, but smoothes the percentage values by calculating them based on a ratio of characteristic weights to the greater of a preset value such as 500 or the actual weight of the highest weighted characteristic in that user's profile.

The invention relates to a centralized heating whole-network heat balance control method, in particular to a control system which is charged on area, which can eliminate the heat imbalance of the whole network and balance the heating effect. The control method adopts different control strategies for a heat source and two links of a primary side in the heating system so as to meet the requirement of users. For the heat source, the online prediction of the heat load is realized according to external temperature variation, and a running instruction is supplied for the heat source; a uniform control strategy is adopted for a primary network, automatic adjustment of a primary valve of a heat exchanging station of the heat network can be realized, a weighted value can be manually given to a special heat exchanging station, and the heat imbalance degree is used for analyzing and evaluating a whole-network heat balance control effect; and the control method can effectively solve the problems of the heating system such as imbalance heat and energy waste, a great amount of manpower and material resource can be saved in the adjustment process of the heat network, and the balance and consistency of the heat supply effect can be guaranteed.

A system, method, and medium for providing configurable item configuration data by receiving a plurality of configuration data sets from a plurality of sources, all which pertain to the same configurable item (CI), then using weighting rules and preferences to assign weight values for each option, feature, and component in each report according to discrepancies and commonalities between the data sets, creating a new data set for the configurable item by selecting options, features, and components having the greatest weight values among all of the data sets, assigning confidence factors to each option, feature, and component by weight comparisons between the plurality of data sets; updating a configuration management database entry for the configurable item accordingly. The confidence factors are useful for inhibiting regulated processes, such as software release processes and change control processes, and for generating human-readable configuration reports indicating confidence levels of the reports and their contents.