53 results about "Combinatorial search" patented technology

A method of providing a search engine for use on global computer networks which identifies and merges categories of information that reflect, influence and imitate intelligent choice by concurrently searching one or more of eight factors of consumer trust: books, experts, news and articles, associations, celebrities and pro choice, awards, web information and blogs and people's choice. The results from the search of these selected consumer trust factors are then combined to generate a final report.

Markdown optimization may be performed using a computer system. Markdown objectives may include margin maximization or inventor minimization. Markdown schedule optimization uses an efficient, iterative, targeted combinatorial search. First, permissible discount steps are identified, then for each permissible discount step, a subspace of permissible step dates is identified. Elements from the date step subspace and price reduction subspace are combined and evaluated against the markdown objectives. If an improved markdown schedule is located, a new subspace of permissible step dates is defined to search the subregion of the solution space where the improved markdown optimization schedule was identified. Scheduling of inventory allocation from a distribution center is also disclosed.

Methods, systems and program storage devices for generating search engine search results. Online files relevant to a requested search query are located by a search engine, while simultaneously previously accessed offline files residing on a client computer are located using the same search query. The search engine retrieves the offline files from the client computer and combines such offline files with the locate online files in a single search result overview. The listing of files within the combined search results are ranked according to relevancy to the search query using a search engine scoring algorithm, whereby the offline files within the combined search results are preferably brought to a user's attention to indicate that such files reside on the client computer.

A method of and system for ranking users by reputation enables better searching for a service provider. Service providers are also forced to conform to reputation requirements since they are published. The reputation requirements include reviews, earnings, duration on a site, recent visits and other components that are able to establish a user's reputation. The components are also weighted so that more important factors count more towards a user's reputation. By combining a search score and a reputation score, a total score is determined which is used to generate a list of results ordered based on the total score.

Methods, systems and program storage devices for generating search engine search results. Online files relevant to a requested search query are located by a search engine, while simultaneously previously accessed offline files residing on a client computer are located using the same search query. The search engine retrieves the offline files from the client computer and combines such offline files with the locate online files in a single search result overview. The listing of files within the combined search results are ranked according to relevancy to the search query using a search engine scoring algorithm, whereby the offline files within the combined search results are preferably brought to a user's attention to indicate that such files reside on the client computer.

A method and system for searching a wide area network that enables users to find the information they seek more quickly and more easily than prior art search engines are disclosed. The method employs various innovative processes that can be included separately or in combination in different embodiments of the invention to improve searching on a wide area network. These processes include a process for making use of a user's prior search queries for query definition; a process for similarly making use of prior search queries by other users; a process for expanding and clarifying the meaning of queries; a process for including paid and other preferred results in an overall query result; a process for using user evaluation data from previous search query results to identify results that should be included or removed from a current result; a process for executing the selected search query, or set of queries, on one or more search engines, collecting the search results and combining the search results; a process for removing subordinate pages from a search result; a process for testing Web pages that appear on a search result, and removing or flagging pages which are unresponsive or slow to respond; a process for ranking search results by applying a set of ranking factors; a pre-caching process enabling faster viewing of search result; a process for providing links to a set of related Webrooms; a process for providing a set of Web maps for Web pages identified in the search results; and a process for integrating user evaluation of the content and usefulness of the Web pages provided by a search and other user data for system use in enhancing search effectiveness.

The invention discloses a system for intelligently searching and treating information and relative method. Wherein, said system comprises a data intelligent treating subsystem, a processing database, an issue and manage module, a search database, and an intelligent search server subsystem, while the issue and manage module comprises a data issue and synchronize module, a data develop manage module; the inventive system can treat the data to divide the data into 12 kinds. Said method comprises: inputting search condition; pretreatment; dividing the search request into simple direct search, advanced integrated search, classified browse search, full-text search, and intelligent logic search, while the first three methods directly use the relationship search engine to search; the full-text search uses full search engine; the intelligent logic search uses logic relationship to recombine the inquire condition to search with the relationship search engine; after obtaining result, feeding back the search result.

Systems and methods are provided for identifying instances of media programming available within a media distribution system. An exemplary method involves searching a listing of media programming available from a first content source using search criteria indicative of a desired media program to identify an instance of the desired media program available from the first content source, obtaining a target program identifier associated with the desired media program from that instance, identifying another instance of the desired media program in a second listing of media programming available from another content source using the target program identifier, and providing a combined search result that indicates availability of the desired media program from the content sources.

A system and method that that interactively provides search previews to users from one or more data sources is provided. In addition, the user interactively combines searches using a drag and drop interface and receive previews of the combined searches. A palate of graphical objects (icons) are displayed to the user. Each of the graphical objects corresponds to a different database search. The user selects one or more of the graphical objects and drags them to a data source representation, such as a data source window. In response to the user's action, a preview request is created and sent to the data source. The user's computer then receives the preview results from the data source and the preview results are displayed to the user.

A combinatorial search method implemented in a computer control system utilizes on-line state-space planning of operations for multi-step production processes. The planner considers various possible combinations of actions, searching for one that correctly transforms the initial state of the object into the specified desired final state. Each combination of actions the planner considers is called a search node, with each node containing a plan representing a series of actions of various machines on a single object and also containing the predicted state of the object with those actions applied either forward or backward. The method includes determining which of the search nodes to extend further at each search iteration and if the object state in the chosen search node conforms to the desired state of the object, or whether actions should be added to the node's plan. Actions that are applicable to the chosen node's object state are selected, transformations are applied to the attributes, and the resulting plan is returned to the system.

Iterative repair problems are generally solved using a combinatorial search method such as simulated annealing are addressed with a FPGA-based coarse-grain pipelined architecture to accelerate a simulated annealing based iterative repair-type event scheduling application. Over 99% of the work done by any simulated annealing algorithm is the repeated execution of three high-level steps: (1) generating, (2) evaluating, and (3) determining the acceptability of a new problem solution. A pipelined processor is designed to take advantage of these steps.

Disclosed herein are optimal or near optimal algorithms implemented as software tools that will allow the user / machine to identify optimal media out of the job set automatically. There are at least two main process steps to the exemplary embodiment. The first step involves the off-line characterization of the printing / image rendering system that will be used to extract the principal basis vectors from the experimental data. The second step involves the run-time mode, in which the pre-characterized basis vectors are used in conjunction with an optimal algorithm that will identify the media for re-calibration / re-characterization from the group of media set that the customer is interested to print. The optimal algorithm performs the combinatorial search every time the customer wishes to select the media. Once the optimal media is selected, the customer or the machine can print color patches (which are also optimal set) and execute the rest of the processing to reconstruct the best color management LUTs.

A search technique having enhanced ease-of-use is provided. A search server analyzes web contents, extracts index information (e.g., time, geographical name, personal name, and business type), and registers the index information in an index information database. A time information search unit performs search using time as a key, a geographical name information search unit performs search using a geographical name as a key, a personal name information search unit performs search using a personal name as a key, a business type information search unit performs search using a business type as a key, and a topic information search unit performs search using a topic as a key. In order to perform narrowing down search while using these search engines integrally, a federated search unit provides a user interface to specify combined search conditions for the search engines.

The invention provides a combinatorial search method, client and server. The combinatorial search method includes acquiring search term combinations selected from multistage classification pages, and sending the search term combinations to the server; receiving search results returned by the server according to the search term combinations. According to an embodiment, the search term combinations selected by a user from the multistage classification pages are acquired and are sent to the server, the server is allowed to search according to the search term combinations, repeated search is avoided, and searching efficiency is improved; on the condition that the user has no specific information of searched objects, searching results meeting the requirements of the user can be acquired, and user experience is improved.

A combinatorial search method implemented in a computer control system utilizes on-line state-space planning of operations for multi-step production processes. The planner considers various possible combinations of actions, searching for one that correctly transforms the initial state of the object into the specified desired final state. Each combination of actions the planner considers is called a search node, with each node containing a plan representing a series of actions of various machines on a single object and also containing the predicted state of the object with those actions applied either forward or backward. The method includes determining which of the search nodes to extend further at each search iteration and if the object state in the chosen search node conforms to the desired state of the object, or whether actions should be added to the node's plan. Actions that are applicable to the chosen node's object state are selected, transformations are applied to the attributes, and the resulting plan is returned to the system.

Disclosed herein are optimal or near optimal algorithms implemented as software tools that will allow the user / machine to identify optimal media out of the job set automatically. There are at least two main process steps to the exemplary embodiment. The first step involves the off-line characterization of the printing / image rendering system that will be used to extract the principal basis vectors from the experimental data. The second step involves the run-time mode, in which the pre-characterized basis vectors are used in conjunction with an optimal algorithm that will identify the media for re-calibration / re-characterization from the group of media set that the customer is interested to print. The optimal algorithm performs the combinatorial search every time the customer wishes to select the media. Once the optimal media is selected, the customer or the machine can print color patches (which are also optimal set) and execute the rest of the processing to reconstruct the best color management LUTs.

The present invention relates to a method and network element for classification of bit strings such as data packets on a per-flow basis, wherein a plurality of classification fields are searched for in respective search trees based on a matching procedure in which an index value is obtained in a leaf node of a search tree for each classification field. Index values obtained in said searching step are used to derive a policy to be applied to said data packet, wherein the number of index values is reduced by combining intermediate results of the searching step or the deriving step. Thereby, computational efficiency can be improved and memory requirements reduced.

A method, device, and software for presenting search results obtained from a plurality of databases, based on an end-user specified query, is disclosed. In an embodiment, the search results are combined from results from a first index and results from a second index. The first index comprises a plurality of index entries modifiable by an administrator, and the second index comprises a plurality of index entries that are not modifiable by the administrator. In the combined search results, any search result from the second index for which an associated key field is identical to the associated key field of a matching search result in the first set of search results is discarded in favor of the matching search result in the first set of search results.

A customized, specialty-oriented database and index. of a subject matter area and methods for constructing and using such a database are provided. Selection and indexing of articles is done by experts in the specialty with which the database is concerned. A thesaurus allows indexing and search in accordance with terminology familiar to different anticipated groups of users (e.g. doctors, patients, nurses, technicians, and the like) by translating search terms to a standard and limited list of topics; a limited number of which are assigned to each document / article and allowing the topics most likely to be relevant to be found. The search can be conducted based on the most relevant topic found or sequential searches can be performed based on topics most relevant to individual search terms or combined search terms and the search results combined on the basis of common topic classifications to leverage the benefits of document classification.

The invention discloses a method for automatically designing physical database based on SQL load digging, comprising two flows of digging load information, database mode information and statistic information and combined-searching candidate and optimum physical characteristics. It makes frame and algorithm completely independent of physical characteristic library and introduces automatic design into different relation database products or adds physical characteristics and optimization solutions to the existing relation database products, not changing integral frames and internal algorithms but only modifying the physical characteristic libraries.

A method of providing a search engine for use on global computer networks which identifies and merges categories of information that reflect, influence and imitate intelligent choice by concurrently searching one or more of eight factors of consumer trust: books, experts, news and articles, associations, celebrities and pro choice, awards, web information and blogs and people's choice. The results from the search of these selected consumer trust factors are then combined to generate a final report.

The invention provides an information searching method and device, the information searching method comprises following steps: obtaining the associated searching word of the searching word in the first stage according to the searching word in the first stage, displaying the associated searching word, the associated searching word comprises the searching word, which has different language compared with the searching word in the first stage; obtaining the searching word in the second stage, the searching word in the second stage is the searching word selected from the associated searching word; and obtaining the corresponding searching result according to the searching word in the second stage, displaying the searching result. The cross-language domain search can be realized by the method and the combinatorial search is achieved.

The invention provides a visualized media unit-based cloud computing search output architecture method. A typical visualized media unit has six structures of a contracted drawing, an original chain, a cloud backup interface, a cloud storage calling interface, a cloud interface and a visualized media information area; visualized media unit structures can be freely combined; search output expression consists of N visualized media units; the cloud interface is expressed in an event function mode; one visualized media unit can contain a plurality of cloud interfaces; all structures are agilely controlled through a background; the visualized media information area is a direct reading view area for deploying a fast news information report, is also agilely controlled by the background, and can perform real-time releasing, modification and supplementation. By the visualized media unit-based cloud computing search output architecture method, a novel immediate-search-immediate-read media search mode is established, and the method meets the requirement of rapid pace and real-time report; and the cloud interface starts a novel cloud computing search service mode, so that the search is rich in expressive force and business models are added.

A system and method for identifying a selection of index candidates for a relational database. The system includes a Predicate Analyzer for examining database queries in workload and recording column and predicate information; a Search Engine for performing a combinatorial search of index candidates; an Index Emulator for emulating the existence of a set of candidate indexes by making necessary “fake” entries into the relational database management system data dictionary; and a Query Optimizer for estimating the cost of a given query within a given index configuration. In accordance with the method, data indicative of a workload comprising one or more queries is received, and predicates are then analysed and organized into column groups for each basic predicate type. Statistics are collected on each of these groups. Non-predicate portions of each query are scanned to obtain additional covering columns. A combinational search is performed on the column groups to identify covering and non-covering candidates. These are then used in conjunction with collected statistics and baseline costs are calculated to perform a cost bases analysis of the candidates and generate index candidate recommendations.

The invention discloses a train beat type running optimization method based on the multi-path combinatorial search. The method comprises the steps that the combination of all train travel time of the minimum beat unit is taken as the optimization target, an optimization model of the train beat type running time table is built, and a weighted diagraph of describing the train running is built; then based on the diagraph, the combination of the directed arc weights passed by all the trains in the diagraph of the minimum beat unit is used as the optimization target, the train beat type running model is rebuilt; on the basis of defining a main path, a secondary path, a path interval, a diagraph node label collection and an initialization node label collection, the main path and the running interval with the minimum total costs from the beginning point to each node is searched through the constant generation of new labels and the update of the label values of the existing labels, until the check and the update of the node labels of the train end station are completed; finally the arrival and departure times and the running time intervals of each train station are determined according to the searched main path and the secondary path collection.

The invention belongs to the field of communication radiation source individual recognition, and discloses a semi-supervised communication radiation source individual recognition method based on an improved TSVM. The method is based on a traditional semi-supervised support vector machine TSVM method. Priori information of various distribution proportions in a training sample and a threshold adjustment factor are utilized, a previous prediction result is optimized through a preset threshold, iteration is carried out according to the confidence coefficient of the prediction result, and a traditional TSVM iteration method based on local combination search is improved. According to the method, in individual recognition of the communication radiation source, when the number of training sampleswith labels is small, compared with a TSVM method, the number of iterations of a classifier can be effectively reduced while the recognition accuracy is guaranteed, and the operand of a model is reduced.

The invention discloses an article searching method and device based on hand drawing, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of receiving a hand-drawn image, extracting a hand-drawn image in the hand-drawn image, and identifying feature information in the hand-drawn image; comparing the hand-drawn image with each article image in anarticle image library to extract article images of which the similarity exceeds a preset similarity threshold, obtaining article information corresponding to each extracted article image, and arranging the article information to generate a first article list; converting the feature information into a feature text label, performing article search based on the feature text label, and combining thesearched article information to obtain a second article list; and combining the article information in the first article list and the second article list to obtain a third article list, and displayingthe third article list. According to the embodiment, when the user cannot provide accurate keywords and cannot obtain corresponding article pictures, a more flexible search mode can be provided for the user, and the search experience of the user is improved.

A method for creating a custom work list includes executing a query based on search criteria. Results of the query comprising at least one of a business object and a task conforming to the search criteria. The search criteria and the returned results are combined to create said custom work list.