118 results about "Sequence search" patented technology

A speech recognition system includes an input device to receive voice sounds, one or more processors, and one or more storage devices storing parameters and program modules including instructions which cause the one or more processors to perform operations. The operations include extracting an acoustic feature sequence from audio waveform data converted from the voice sounds, encoding the acoustic feature sequence into a hidden vector sequence using an encoder network having encoder network parameters, predicting first output label sequence probabilities by feeding the hidden vector sequence to a decoder network having decoder network parameters, predicting second output level sequence probabilities by a hybrid network using character-base language models (LMs) and word-level LMs; and searching, using a label sequence search module, for an output label sequence having a highest sequence probability by combining the first and second output label sequence probabilities provided from the decoder network and the hybrid network.

The invention provides a method, a device and a search engine for sequencing searching keywords, belonging to the field of data mining. The method comprises the following steps: collecting the searching log information and the attribute information of users, wherein the searching log information includes user marks and the searching keywords, and the attribute information includes the user marks and user attributes; establishing the congruent relationship between the user attributes and the searching keywords according to the searching log information and the attribute information; computing an attribute weighted value of each searching keyword by a predetermined searching keyword weighted formula according to the congruent relationship between the user attributes and the searching keywords; and sequencing the searching keywords according to the attribute weighted values of the searching keywords. The embodiment of the invention can accurately and fast mine data information which not only has higher searching heat but also represents searching habits of various user groups, thereby obtaining the information capable of really reflecting the searching habits of network users so as to provide data support for further knowing and meeting requirements of the users.

The invention discloses a method and device for sequencing search results, and main contents are as follows: in the embodiment of the invention, not only can the degree of association between the search results and a retrieval word be taken as a sequencing basis, but also amendment conditions are pre-set for each search result, and the degree of association between the search results and the retrieval word and amendment parameters of the search results are combined for sequencing the research results, thereby enabling the final sequencing to be capable of meeting search demands of a user to the maximum extent, improving the efficiency of inquiring the required result from the search results of the user, and reducing system resources occupied by the search process.

A bidding-sequence advertisement system based on search engine as per time slot comprises information issuing-module for inputting information content to be issued by publisher, storage databank for storing information and exhibiting price, checking module for verifying the truth of inputted information and for storing qualified information in said databank, index module for searching information in time slot according to key word, sequencing module for sequencing searched result according to high-low of price and result page generating module for generating final result page according to webpage format template.

The invention provides a search method and a search engine. The search method comprises the following steps of receiving search words of a user, and acquiring the current environmental parameter of the environment where the user is positioned; acquiring a plurality of search results according to the search words; sequencing the plurality of search results according to the current environmental parameter; providing the sequenced search results for the user. The search method provided by the invention has the advantages that the sequencing for the search results is optimized according to the environmental parameter, so that the user can acquire needed results quickly, the time and the operation of the user for acquiring the search result are reduced, the requirements of the user on personalization and diversification are met and the search experience of the user is improved.

The embodiment of the invention provides a hot event searching method and device. The method comprises the steps of determining query words with the search volume greater than a first preset threshold value in a preset time period from the current time; determining the search trend index of the query words; determining the query words with the search trend index greater than a second preset threshold value as real-time hot words; storing the determined real-time hot words in a hot word database; when receiving search terms sent by users, sequencing search results on the basis of the hot word database; and sending the sequenced search results to the users. According to the method provided by the embodiment of the invention, the search results are sequenced on the basis of the hot word database, the sequencing of the search results can be controlled by means of the sequencing mode when the search terms sent by the users directs at hot events and the results, related to the hot events, in the search results are forward positioned, so that the problem that the hot event search feedback of search engines is delayed in the prior art is avoided.

An audio analysis system for contents including: an audio flow obtaining module, an audio flow segment module, an audio flow type judgment module, a phone flow analysis module and a key word search module based on spelling sequences, in which, the audio flow obtaining module gets an audio flow from outside and sends it to the segment module to segment it to enable each section has a single acoustic character, and the audio flow with a single character is sent to the audio flow judgment module for analysis to get its character kind, the segment determined to be non-phonetic is cast away and the phone audio segments are sent to the phone analysis module to get a spelling pattern and then the key word search module based on the spelling sequence searches the key words to get the position in the audio flow.

The invention discloses a radar signal sorting method based on a PRI histogram. The method comprises the following steps: establishing the PRI histogram of a radar signal pulse flow; sitting differentjitter ranges for a fixed signal and a dither signal; firstly extracting the fixed signal, and extracting the dither signal, specifically comprising: enabling a box mid-value of each PRI overlappingbox in the PRI histogram to compare with a discrimination threshold: if a value exceeding the discrimination threshold is not existent, zero-clearing the box mid-values of all PRI overlapping boxes inthe PRI histogram; otherwise, using a PRI value corresponding to a peak value exceeding the discrimination threshold as the possible PRI value for performing sequence search; if a sequence is successfully searched, eliminating a detected pulse from the radar signal pulse flow, otherwise, zero-clearing the box mid-values of all PRI overlapping boxes in the PRI histogram, and performing the next order of pulse extraction. The method is capable of improving sorting capacity of the dither signal, and has the better sorting effect to pulse data in a complicated environment.

The invention provides a recommendation method and search engine for search candidate words. The recommendation method comprising steps that a search engine server receives input information input by a user and obtains prefix information of the input information; the prefix information serves as an index to obtain a plurality of search candidate words and weight of each search candidate word; whether at least two search candidate words of the search candidate words belong to the same subject is judged; if at least two search candidate words belong to the same subject, the weight of one of the at least two search candidate words is kept unchanged, downgrading processing is conducted on the weight of others of the at least two search candidate words; sequencing is conducted according to the weight of the search candidate words, and the sequenced search candidate words are provided for the user. According to the recommendation method, diversity and accuracy of the search candidate words are improved, search requirements of the user can be met, an algorithm is simple and easy to implement, and user experience is improved.

The invention relates to the field of internet information treatment and discloses a method and a device for sequencing search results. The method comprises the following steps of: acquiring corresponding search results and preset field attributions corresponding to the search results according to search keywords input by a user; calculating the field attribution of the search keywords according to the field attribution of each search result; respectively regulating the correlation of each search result with the search keywords and the authority of each search result according to the field attribution of each search result and the field attribution of the search keywords; and sequencing the search results according to the correlation and authority of each search result after regulation. Therefore, the sequencing accuracy of the search results are improved at the aspects of the correlation and the authority, the search quality is effectively improved, and the system performance is promoted.

The invention relates to a motion sequence search method based on an alignment clustering analysis. According to an existing motion sequence search method, when segmenting is taken into consideration, the motion time sequence of each frame is not taken into consideration, and segmentation points cannot be accurately positioned. According to the motion sequence search method, the human body motion sequence is recorded and stored; a server end processes motion sequences to obtain the characteristics of the motion sequences, and the characteristics of all the motion sequences are combined to generate a characteristic database; the server end calculates the characteristics of the motion sequences provided by a client side; the server end matches the characteristics extracted by the motion sequences provided by the client side with the characteristics in the characteristic database, the distances between the characteristics are respectively calculated, and the motion sequences in the database are ranked and output according to the distance to serve as a retrieved result to be returned to the client side. According to the motion sequence search method based on the alignment clustering analysis, the motion characteristics are extracted on that basis and applied to the motion search, and compared with a true segmentation result, the segmentation method has efficiency.

A method for optimizing data transfer through parallelized interlacing of data based upon sorted characteristics to minimize latencies inherent in the system. Operating a plurality of threads in parallel on disparate file sizes ordered by an additional thread. Efficient backup of heterogeneous non-volatile mass store to a network attached server. Scalable distribution of backup processes for computing hashes and eliminating duplication. Increased granularity of file pieces to match blocking of file I/O with network transmission. Efficiently packing each network transmission block using sequence search criteria. Avoidance of sending undesired pieces. Segmentation of each file and object into a hierarchy of pieces in a plurality of types.

The invention provides a referential DAN sequence compressing method and system. The method comprises the step of matching repeated patterns, namely, creating an inverted full-text subsequence structure based on an input reference sequence, and performing the longest matching subsequence searching on the input sequence to be compressed, wherein the matching information is applied to the compressed encoding step, and the unmatched symbols are applied to the step of predictive encoding of non-repeated symbols; the step of compressed encoding step, namely, performing compressed encoding on the matching sequence length and location information, wherein the encoding information is applied to decompressing; the step of predictive encoding of non-repeated symbols, namely, receiving unmatched symbols in the step of matching the repeated patterns, and predicting the symbol occurrence probability and encoding through a mixed context model. According to the method, the properties of efficient searching of index data and efficient compressing of the mixed context model to single symbol are fully combined, so that when being compared with that of other referential DNA sequence methods, the compressing rate is high within the acceptable compressing time consumption; the method is high in practicability.

The invention provides a method for sequencing search results. The method comprises the following steps: responding the operation of a user on the search result which comprises a specific search result and a common search result, and extracting characteristic indexes; judging whether a corresponding specific satisfaction formula and a common satisfaction formula exist or not, if not, establishing the specific satisfaction formula and the common satisfaction formula, and if so, calling the specific satisfaction formula and the common satisfaction formula; according to the characteristic indexes, utilizing the specific satisfaction formula and the common satisfaction formula to calculate user satisfaction values of the specific search result and the common search result; and according to the user satisfaction values, sequencing the search result. Correspondingly, the invention also provides a system using the method. According to the invention, the search result can be sequenced reasonably.

The invention discloses a search method, which comprises the following steps of: sequencing search results according to a user operation situation and user influence, namely determining search object importance levels of the search results according to behavior influence factors and user influence factors, which correspond to the search results; and sequencing the search results according to the search object importance levels from high to low. The invention correspondingly discloses a search system. According to the search method and the search system, the search precision can be improved, and the homogenization problem is avoided, so that the user experience is enhanced.

A process provides for searching through a written text in response to a spoken question comprising a plurality of words. The first step in the process is to transcribe the written text into a first sequence of phonetic units. Then, a spoken question is segmented into a second sequence of phonetic units. The search is conducted through the written text for an occurrence of the spoken question. The search comprises aligning the first and second sequences of phonetic units.

The invention provides a method for sequencing search results. The method comprises the following steps: firstly, generating a search strategy submatrix for each searching item, wherein each row in the search strategy submatrix corresponds to a search strategy layer; the priority weights of different search strategy layers increase or decrease progressively in sequence; and the priorities of the search strategies on the same search strategy layer are same; secondly, selecting at least a search strategy from each search strategy layer of the search strategy submatrix to assemble a query set, matching the query sets with all searched files according to the search strategies and multiplying the file scores by the priority weights corresponding to the search strategies to obtain result set objects; and selecting the result set objects and assembling the selected result sets to form a complete result set. The method has the following beneficial effect: the sequencing accuracy of the search result sets is ensured by the following steps: generating the search strategy matrix by using various search strategies, carrying out data domain division on the search result sets and then assembling the search result sets.

The present invention relates to a computer-implemented method, system and computer readable medium for providing a scalable bio-informatics sequence search on cloud. The method comprises the steps of partitioning a genome data into a plurality of datasets and storing the plurality of data sets in a database. Receiving at least one sequence search request input and searching for a genome sequence in the database corresponding to the search request input and scaling of the sequence search based on the sequence search request input.

A method for sequencing search engine of multimedia file includes setting relevant rules in advance, obtaining download-chaining information of multimedia file from Internet by reptile, analyzing the sequencing rule to be atomic rule by detector and carrying out detecting and grading on said download-chaining information according to analyzed-out atomic rule then carrying out sequencing on download-chaining of multimedia file by indexer according result of said grading.

The invention discloses a history segmented sequence search and time sequence rarefaction-based wind power forecast method. The method comprises the following steps of: normalizing wind power historydata and determining window width optimum values of searched segmented time sequences according to power time sequence fluctuation features and basic statistical characteristics; calculating matchingdegrees between all the history segmented time sequence and a segmented time sequence at the current moment by taking the newest segmented time sequence at the current moment as a standard and synthesizing a correlation index and a similarity distance index; sorting the matching degrees according to a big-to-small sequence, and determining a number of optimum history segmented sequences accordingto an average matching degree aggregation principle; aiming the current segmented time sequence at each moment, determining a number of corresponding optimum history segmented sequences and a number of corresponding optimum average history segmented sequences; aiming at all the moments for training the time sequences, establishing a time sequence rarefaction-based power forecast model; and solvingthe model by adoption of a multiplier alternating direction method to obtain parameters of the model, wherein the parameters are used for future power forecast.

The invention discloses a method for searching and sequencing keywords of extensible markup language (XML) documents based on semantic correlation, which comprises the following steps of: sequentially analyzing the XML documents; calculating the semantic correlation of a subject node and an attribute node, and the semantic correlation of the attribute node and the keywords; optimizing search time; performing word stemming on the input query keyword; taking subject node information and correlation information which correspond to the keyword out of an inverted index; searching a subject closest to the keyword; sequencing search results according to correlation from high to low; searching a subject second closest to the keyword; and returning an information segment to a user according to a Dewey code in a result. Aiming at the unique structural semantic characteristic of XML data, SRank correlation search model and method is provided, so that the accuracy of the research results can be improved.

The present invention provides a method and an apparatus for updating an index and sequencing search results based on the updated index in a terminal. The method comprises: retrieving whether there is any modification in a file; if there is any modification in the file, performing an increment index the modified file to generate new index file, wherein the increment index includes a number of times that the modified file is selected historically; merging the new index file into the original index file; obtaining key words input by the user; querying the search results related to the key words, sequencing the search results according to the relevance between the search results and the key words and the number of times that the modified file is selected historically and displaying the sequenced search results to the user. By the present invention, the user experience of the mobile terminal is improved.

Provided is a system for searching for patent documents by using a DNA fragment number. The system includes a DNA fragment sequence database storing information about DNA fragment sequences; a patent sequence database storing information about patent DNA fragment sequences of patent documents; a patent database storing information about patent documents; an interface receiving a DNA fragment number from a client; a sequence extracting unit extracting a DNA fragment sequence corresponding to the DNA fragment number from the DNA fragment sequence database; a matching sequence search unit searching for patent DNA fragment sequences matching the extracted DNA fragment sequence in the patent sequence database; a patent number extracting unit searching for patent numbers of patent documents including the patent DNA fragment sequences from the patent sequence database; and a patent search unit searching for patent documents from the patent database by inputting the extracted patent numbers.