152 results about "Similarity data" patented technology

An electronic status notification message for an output job request, which identifies a document for processing at an output device, is received from a submitter. A copy of the document is received, and the output job request is forwarded to the output device. Selected document services are applied to the copy of the document to define an archive of the document. The archive of the document is stored on a document repository. Similarity data is produced by evaluating the archive of the document with other documents in the document repository. Job status information is received concerning the output job request forwarded to the output device. The electronic status notification message is prepared using static and dynamic data concerning the job status information of the output job request and the similarity data of the document archived in the document repository. The electronic status notification message is transmitted to the submitter.

A system and method for enhancing spam avoidance efficiency and brand protection by automatically identifying a phishing website without human intervention. The system receives a stream of suspect Internet urls for potential phishing websites and uses a comparison strategy to determine whether the potential phishing website has already be labeled as a bonefid phishing website. A comparison system is utilized in which similarity data is calculated on various elements of the potential phishing website and then compared to similarity data of known phishing websites and known brands to determine whether the site needs human intervention. Various types of categorization structures and notification strategies are utilized in the system, including the adjustment of threshold comparison values in response to the identification of a potential phishing site displaying a brand of interest.

System, method and computer program for facilitating media playlist generation based at least in part on media library inventory information provided by a plurality of program participants. Data is transmitted from a program participant's client device indicative of media inventory in a media library of the program participant. Media item similarity ratings are received at the client device that have been compiled based on cumulative data collected from a plurality of program participants, including identification data of individual media items contained in media libraries of the program participants regardless of each individual media item's source. Similarity ratings compilation includes processing the cumulative data to determine an incidence of co-occurrence of pairs of individual media items in different program participants' media libraries and making an assignment of a similarity rating based on the determined incidence of co-occurrence.

The invention discloses a differential compression method based on data de-duplication. The differential compression method includes steps of partitioning files in data flow to obtain multiple data blocks; computing data block fingerprint of each data block for searching duplicate data; grouping all the data blocks to establish data block groups and double link lists thereof; searching the fingerprint of each data block in each data block group for realizing data de-duplication so as to determine whether the data block is duplicated or not; searching similar data locally to the data block group which is subjected to the data de-duplication process according to the duplicated data information in the double link lists of the data block groups, namely, determining the non-duplicated data blocks adjacent to the duplicated data blocks as potential similar data blocks; verifying the similarity of the similar data blocks by differential compression; and finally complementarily searching similarity data to the data block groups according to the similarity. The differential compression method based on data de-duplication has the advantages of rapidness in similar data searching, low computing and indexing overhead and high data compression efficiency.

System, method and computer program that encompasses one or more client devices cooperating with a program host's on-line data processing center to generate media playlists. Data is transmitted from a program participant's client device to the processing center and is representative of media inventory in a media library on the client device and includes identification data of contained individual media items. Data is further received from a plurality of program participants representative of each respective program participant's media inventory, regardless of the source of the members of the inventory. Co-occurrence of pairs of individual media items in different program participants' media libraries is determined and a similarity rating is assigned between the pairs of individual media items based on the determined incidence of co-occurrence in the different program participants' media libraries. From this, a media playlist is generated.

The invention relates to a face model training module, relative method, and a face real-time identify system, and method, wherein in the identification, first using face sample image, to supply one face model supporting vector machine to each user; via collecting the video image input by camera, searching and checking the face of image, and tracking and identifying the image; then automatically marking the organ character point of face, to pre-treat checked face; calculating the Gabor character of face image after pretreatment; selecting low-dimension character vector from high-dimension Gabor character; inputting the selected low-dimension character vector into face model, to process face recognition, to feedback similarity data of each face model; based on said similarity data, outputting final face identifying result. The invention can improve the right rate of face recognition identification.

A system and method for online real-time multi-object tracking is disclosed. A particular embodiment can be configured to: receive image frame data from at least one camera associated with an autonomous vehicle; generate similarity data corresponding to a similarity between object data in a previous image frame compared with object detection results from a current image frame; use the similarity data to generate data association results corresponding to a best matching between the object data in the previous image frame and the object detection results from the current image frame; cause state transitions in finite state machines for each object according to the data association results; and provide as an output object tracking output data corresponding to the states of the finite state machines for each object.

The invention relates to an output power prediction method based on similarity data selection for a photovoltaic plant, and belongs to the technical field of photovoltaic power generation. The method comprises the following steps: step 1, collecting irradiation intensity values, temperature values and actual photovoltaic output power values of historical days, as well as irradiation intensity values and temperature values of predicted days in weather forecast; step 2, determining weights w1 (i) corresponding to irradiation intensity of all whole points from 6 am to 18 pm every day, and determining weights w2 (i) corresponding to temperature of all whole points from 6 am to 18 pm every day; step 3, performing selection on similar days; step 4, determining weight of power in each similar day during prediction according to the degree of correlation between the similar days and the predicated days; step 5, obtaining a power predication value required in the process that the photovoltaic output is performed in the predicated days through calculation, and performing evaluation on a predicated result. The method can well excavate the correlation between the predicated days and history data, is easy to implement and improves predicated accuracy of the photovoltaic output power.

A moving image search device includes: a moving image database (11) for storage of sets of moving image data; a scene dividing unit (21) which divides a visual signal of the sets of moving image data into shots and outputs, as a scene, continuous shots having a small characteristic value set difference of an audio signal to the shots; a video signal similarity calculation unit (23) which calculates, for each of scenes obtained by the division by the scene dividing unit (11), video signal similarities to the other scenes according to a characteristic value set of the visual signal and a characteristic value set of the audio signal, and thus generates video signal similarity data (12); a video signal similarity search unit (26) which searches the scenes according to the video signal similarity data (12) to find a scene having a smaller similarity to the each scene than a certain threshold (12); and a video signal similarity display unit (29) which acquires and displays coordinates corresponding to the similarity for each of the scenes searched out by the video signal similarity search unit (26).

Data elements from data sources and having a data value set are linked by using hash functions to determine a dimensionally reduced instance signature for each data element based on all data values associated with that data element to yield a plurality of dimensionally reduced instance signatures of equivalent fixed size such that similarities among the data values in the data value sets across all data elements is maintained among the plurality of instance signatures. Candidate pairs of data elements to link are identified using the plurality of instance signatures in locality sensitive hash functions, and a similarity index is generated for each candidate pair using a pre-determined measure of similarity. Candidate pairs of data elements having a similarity index above a given threshold are linked.

An electronic status notification message for an output job request, which identifies a document for processing at an output device, is received from a submitter. A copy of the document is received, and the output job request is forwarded to the output device. Selected document services are applied to the copy of the document to define an archive of the document. The archive of the document is stored on a document repository. Similarity data is produced by evaluating the archive of the document with other documents in the document repository. Job status information is received concerning the output job request forwarded to the output device. The electronic status notification message is prepared using static and dynamic data concerning the job status information of the output job request and the similarity data of the document archived in the document repository. The electronic status notification message is transmitted to the submitter.

A system and method for enhancing spam avoidance efficiency by automatically identifying a phishing website without human intervention. The system receives a stream of suspect Internet urls for potential phishing websites and uses a comparison strategy to determine whether the potential phishing website has already be labeled as a bonefid phishing website. A comparison system is utilized in which similarity data is calculated on various elements of the potential phishing website and then compared to similarity data of known phishing websites. Various types of similarity measure methodologies are potentially incorporated and a similarity threshold value can be varied in order to respond to phishing threats.

System, method and computer program that encompasses one or more client devices cooperating with a program host's on-line data processing center for making purchase recommendations of media items to a program participant. Accordingly, there is a transmission from a program host to a program participant of identification of at least one recommended-for-purchase media item selected based on the recommended-for-purchase media item being rated as similar to a designated media item presently contained in an inventory of a media library of the program participant. The rating of the recommended-for-purchase media item as being similar to the designated media item presently contained in the inventory of the media library of the program participant is representative of a frequency at which the recommended-for-purchase media item co-occurs with the designated media item in media libraries of others.

The invention provides a data deduplication method based on internet small computer system interface (iSCSI), belongs to the technical field of computer information storage, and is suitable for an iSCSI-protocol-based internet protocol (IP) network remote mirroring system. In the invention, bandwidth simplification and synchronous time reduction can be realized through deleting repeated data of an iSCSI write data block on the premise of not changing the structure of the conventional IP network remote mirroring system. The data deduplication comprises two stages: in the first stage, a coarse-grained similarity data chunk detection technology is adopted, and a content-defined chunking (CDC) algorithm and a bloom filter algorithm are combined to search similarity chunks in a full range, so that the data deduplication can be more flexible and more accurate; and in the second stage, an improved fine-grained similarity data chunk detection technology is adopted, and a fixed-sized partition and sliding window method is combined, so that the deduplication is performed on chunks rather than files, and transparency of deduplication to users is realized.

The invention discloses a method for the rough draft shape matching based on the Modular product graph and the maximum clique. The method comprises the following specific steps that shape analysis and feature extraction are conducted on graphic data, so that essential data expressing images are obtained; based on shape feature expressions of the labeled graph and the weight graph, the extracted data are labeled or coded, so that topological graphs are obtained; the topological graphs are compared through a similarity calculation model, the labeled or coded data are completely matched or part of the data are matched, similarity data are obtained through calculation, and whether the graphs are similar to one another is judged. According to the method, matching of graph models with different numbers of nodes can be processed and partial matching can be conducted as well. Compared with the matching method based on vectors, the method for the rough draft shape matching based on the Modular product graph and the maximum clique is higher in accuracy, higher in adaptability and higher in matching capacity.

The present invention provides a music similarity processing method which comprises the following steps: after inputting first characteristic information of multimedia information of first music or the multimedia information of first music, extracting first characteristic information from the multimedia information of first piece of music; decomposing the first characteristic information to a plurality of information segments which start from a random starting point and have a certain length; after inputting the second characteristic information of multimedia information of second piece of music or inputting the multimedia information of second piece of music, extracting the second characteristic information from the multimedia information of second piece of music; calculating the information similarity data of a random segment of a plurality of information segments and a random segment in the second characteristic information; selecting a maximum value of similarity from the similarity data; and determining whether the maximum value of similarity exceeds the preset threshold. If the preset threshold is exceeded, high similarity between the first piece of music and the second piece of music is determined, and otherwise low similarity between the first piece of music and the second piece of music is determined.

A method and apparatus for reducing artifacts during image processing is described. In some embodiments, the method includes examining an input image comprising at least one source region, for each target region in an output image, identifying a portion of the at least one source region based on similarity data, defining an omnidirectional cut boundary and determining intensity values for the each target region and at least one pixel between the each target region and the omnidirectional cut boundary.

The present invention discloses a discrete particle swarm optimization based local community detection collaborative filtering recommendation method, mainly in order to solve the problem of low recommendation accuracy due to that the prior art has sparseness when obtaining similarity data among users. The method comprises the following steps: obtaining scoring information of users to the recommendation item, and indirectly generating a relationship network among the users by using the scoring data of the users to the to-be-recommended item; calculating similarity among the users, carrying out local community detection on the user relationship network through the similarity so as to obtain the user community with the densest local, and expanding the user community to obtain the local user community; dividing the user relationship network into a plurality of user communities, selecting k users with the largest similarity in the user communities to form a neighbor user group; and according to the neighbor user group, predicting the score of the items that are not evaluated by the target users, and recommending the item with the largest predictive score to the users. According to the method disclosed by the present invention, a better recommendation result can be obtained, and the method can be applied to recommend items that the user is interested in to the user.

A method is provided for analysing the self-similarity of an audio file. The method involves obtaining the audio spectrum envelope data of an audio file to be analysed; performing a clustering operation on the spectrum envelope data to produce a clustered set of data; for a first portion of the clustered data, performing a string matching operation on at least one other portion of the clustered data; and based on the results of the string matching operation, determining the at least one other portion of the clustered data most similar to said first portion of the clustered data. There is also provided a method of repairing an audio stream received over a network using similarity data to replace damaged or missing portions of data with similar “good” portions of data.

The invention discloses an entity alignment method and system for digital education resources, and the method comprises the steps: carrying out the subject partitioning of an existing education resource library through a subject topic tree, and building a partition index system in each education resource block; extracting knowledge point tags of the to-be-aligned digital education resources through an SKE algorithm, and screening out an entity alignment candidate set through an index; calculating knowledge point tag similarity, attribute value similarity and text resource similarity of the entity pairs in the candidate set; and finally, inputting the entity pair similarity data obtained by calculation as a characteristic value into the constructed decision tree, and judging an alignment result. When the method is adopted to align the entities of the digital education resources, the complexity of entity alignment calculation can be greatly reduced, and the alignment efficiency is improved; and the text similarity calculation precision of the digital education resources is improved, so that the entity alignment accuracy is improved.

The invention provides a three-dimensional image face recognition method and a three-dimensional image face recognition system. The three-dimensional image face recognition method specifically comprises the steps that a standard reference face model and a registration library of a face database under a front posture are established in advance; aligning the depth data with a depth image of a standard reference face model, and obtaining attitude parameters; realizing texture image alignment according to the attitude parameters; respectively carrying out feature extraction on the aligned depth image and texture image, carrying out calculation through a depth classifier to obtain depth similarity Sdepth, and selecting a corresponding texture classifier according to the deflection attitude angle to calculate texture similarity Stexture; and utilizing the finally aligned face image and the obtained similarity data to carry out final face recognition by utilizing RGB-D. The 3D texture matching is carried out by using a pre-stored standard 3D face library to predict shielded facial textures, so that the recognition performance under the conditions of large posture change and serious shielding can be greatly improved.

The invention discloses a similarity data clustering method for dam safety monitoring data. The method comprises the following steps of separating a single measuring point sequence trend term from thehigh-frequency noise by utilizing an EMD algorithm, detecting the time sequence change points by adopting an inflection point detection method of a cumulative sum control graph, and splitting to obtain all subsequence sets; adopting a DTW distance measurement method for calculating the distance problem of the subsequence, and calculating the distance minimum value between the two pieces of subsequence data dynamically; and clustering the mined sub-time sequences by using hierarchical clustering, and dynamically analyzing the time sequence clustering distribution condition under different clustering numbers through the obtained tree-shaped clustering graph. According to the method, the similarity of the monitoring data is reasonably analyzed, the correlation of the same monitoring point inthe time sequence can be mined, and meanwhile the correlation between the safety monitoring data can be quantified. And the monitoring data subjected to similarity analysis processing can accuratelyreflect the change trend of the dam in the time dimension, and the subsequent monitoring data mining difficulty can be effectively reduced in combination with the change trend rule.

The invention discloses a method for predicting the relations between microorganisms and diseases. According to known microorganism-disease relation data and disease similarity data based on symptoms,a microorganism similarity network and a disease similarity network are constructed, and then through network consistency projection, the potential relations between microorganisms and diseases are predicted. Compared with the prior art, the method is characterized in that by integrating the disease similarity data based on symptoms, any parameter is not required in the whole calculation processes; the method is suitable for a condition of less microorganism-disease relation data; and the relations between microorganisms and diseases can be quite precisely predicted. The provided method is aglobal calculation method, so relations between multiple pairs of microorganisms and diseases can be predicted massively; and problems of blindness and high cost of a bio-experiment method can be effectively solved.

A computer implemented method is for generating a media playlist including a plurality of tracks to be played on a listener's mobile or stationary client device with Internet radio capabilities, the client device intended to be connected to the Internet. The method comprises: obtaining, as a listener's input into the listener's client device, a playlist definition; selecting, from a plurality of tracks, tracks meeting the playlist definition to form the playlist, wherein the playlist is formed by playlist entries that include track identifications referring to selected ones of the plurality of tracks; tracks present in a remote master media inventory, tracks present in an Internet-based cloud memory environment, and tracks present in a local media content inventory of the listener's client device form the plurality of tracks; selecting tracks that meet the playlist definition includes comparing the playlist definition with entries for tracks in a metadata encyclopedia which includes metadata derived from a master metadata encyclopedia referring to the tracks present in the remote master media inventory and the tracks present in the cloud memory environment, and the local media content inventory kept in the local listener's client device; the metadata encyclopedia is kept locally in the client device; and each entry in the local metadata encyclopedia refers to a respective track and includes at least one track descriptor and at least one similarity data descriptor; and providing the playlist to the listener's client device for obtaining the tracks indicated on the playlist for playing the tracks in the playlist in an order defined in the playlist.

Methods and arrangements for facilitating media playlist generation for a program participant based at least in part on media library inventory information provided by a number of program participants. The system or program in which the individuals are participating is an on-line media store. Those individuals that decide to be program participants are interested in organizing, maintaining and playing their music, based at least in part, on data derived from a population of other participants in the program that have similar or the same music in their libraries. To be a program participant, the individual music holder must send, and the on-line music store receive, data representative of that program participant's media inventory. This data typically contains identification data of the individual media items presently contained in that participant's media library regardless of the individual media item's source. The system or program determines an incidence of co-occurrence of pairs of individual media items in different program participants' media libraries. Based on this determination, a similarity rating is assigned between the pairs of individual media items based on the determined incidence of co-occurrence in the different program participants' media libraries.

Relationships between digital text volumes in a corpus are identified by comparing a plurality of digital text volumes in the corpus to identify clusters of similar volumes. For a given cluster, the pages of pairs of volumes within the cluster are compared to determine page similarity data, the page similarity data indicating degrees of similarity between the compared pages. The relationships between pairs of volumes in the clusters are classified based on the similarity data.

A fast recognition algorithm of similarity data in a big data set comprises the steps of correcting the length of a data file, calculating the distance between sampling data blocks, calculating the positions of the sampling data blocks, extracting the data blocks, extracting one data block at the head of the data file and one data block at the tail of the data file, calculating the characteristic value of the extracted data blocks and judging the similarity of data through set operation. According to the fast recognition algorithm, the space-time overhead does not increase along with increase of the size of the data file, the length of the data file is corrected through a position influence factor, recognition failure of the similarity data due to position offset of the sampling data blocks can be avoided, and the effectiveness and the accuracy of recognition of the similarity data are effectively guaranteed through the information retrieval method.

System and methods for delivering personalized offers are provided. In particular, based on an identification of a user from a plurality of users associated with a portal, items and inducements for an offer can be generated and delivered to the user. This involves identifying other user associated with the item and the portal and projecting a degree of interest for the user in the item based on similarity data between the plurality of users. The similarity data can be utilized and generated using a collaborative filter engine. Once items and corresponding inducements are identified, a communication can be provided for the user that facilitates access to information regarding the item and the associated inducement.

The invention discloses a multi-source heterogeneous data identity recognition method based on an attention mechanism, relates to the field of equipment, and adopts the technical scheme that a plurality of face cameras are mounted on a pedestrian advancing route, face images are snapshot, and the snapshot time is recorded at the same time; the face images are recognized through the built face recognition system, and similarity data is returened; an IDs similarity vector is obtained; offline training is carried out on the model through a cross entropy loss function by utilizing a large amount of label data; and identity recognition is realized. The method has the advantages that the face images are captured through the face camera capture points, the problem that the recognition accuracy isnot high due to the fact that the method is limited by a non-matching scene of capturing the face images through a single camera is solved, and the method is more friendly and more rapid and effective; external influence factors of a human face picture snapshot scene are simulated by utilizing external weather data, time intercept points and surrounding building data, and the human face similarity of a real scene can be restored as much as possible.