833results about How to "Efficient mining" patented technology

A big data real-time storage, processing and querying system includes a distributed support module, a data acquisition module, a message-oriented middleware module, a data cleaning module, a data processing module, a data deep mining module, a data management module, a data querying module, and a system scheduling module; all the modules run under the coordination of the system scheduling module, and scheduling and circulation of data flows among all the modules can be achieved Processing flows among all the modules can be coordinated, and the date real-time performance is improved; effective improvement is performed for high throughput, mass data, fast response messages and data high availability; and then the comprehensive ability of a big data real-time data service system can be improved.

The invention discloses a region-level and pixel-level fusion saliency detection method based on convolutional neural networks (CNN). The research object of the method is a static image of which the content can be arbitrary, and the research goal of the method is to find a target striking the eyes of a person from images and assign different saliency values to the target. An adaptive region generation technology is mainly proposed, and two CNN structures are designed and used for pixel-level saliency prediction and saliency fusion respectively. The two CNN models are used for training a network model with the image as input and the real result of the image as a supervisory signal and finally outputting a saliency map of the same size as the input image. By means of the method, region-level saliency estimation and pixel-level saliency prediction can be effectively carried out to obtain two saliency maps, and finally the two saliency maps and the original image are fused through the CNNs for saliency fusion to obtain the final saliency map.

A method to extract new keywords based on word frequency and multiple grammars is provided, which belongs to the technology field of a natural language processing, and is characterized by extracting the potential models of part of speech of the multiple grammars of the keywords by researching characteristic part of speech of the keywords and adopting computer to assist excavation and taking the models as the basis of the keywords to extract arithmetic. When extracting the new keywords, firstly excavating the multiple phrases in text in accordance with the potential models of part of speech and extract candidate word set of the keywords, and then excavating potential keywords not loading from titles and add the potential keywords to the candidate keyword set. The application brings forward an improved single text word frequency / inverse text frequency value (tf / idf) format, introduces target-oriented characteristics, grades the candidate keywords, obtains the order of the candidate keywords and gives the keywords of news document after optimizing the results. Compared with the traditional keyword extraction method based on single text word frequency / inverse text frequency value (tf / idf), the method has higher recall rate under the condition of the same precision.

The invention discloses an exploitation method for a wax precipitation oil reservoir. The method comprises the step of adding a chemical agent and thermal fluid into an oil layer to make crude oil in the oil layer smoothly seep into an oil well and then rise to the ground surface. The method has the advantages of effectively solving the problem of wax precipitation of the crude oil in the oil layer and exploiting the wax precipitation oil reservoir, greatly reducing the exploitation cost of the wax precipitation oil reservoir, prolonging the production period and increasing the yield of the oil well and enhancing the final recovery efficiency of the oil reservoir, along with easy implementation, safety and reliability and obvious economic benefit.

The invention provides an upward and downward staged rock drilling, sublevel bottom cutting and ore removal subsequent backfilling mining method. The method includes the following steps that panels are divided in the ore direction, a stope preparation system is arranged, upward fan-shaped medium-depth blast holes are drilled from one side to the other side of a chamber, the blast holes are filled with explosives for detonation, and an undercut space is formed after detonation; cut grooves are formed through a VCR method, and as for ore above the undercut space, downward vertical deep holes are drilled in a drilling chamber at the top of the chamber, and the holes are filled with explosives for detonation with the cut grooves as a compensation space; the chamber in each panel is subjected to two-step backstoping, after ore removal of a first-step chamber is finished, the first-step chamber is filled with cementing materials, after the first-step chamber is filled and cured, a second-step chamber is subjected to backstoping, the backstoping steps of the second-step chamber are the same as those of the first-step chamber, and the second-step chamber is filled with cemented filling bodies and non-cemented filling bodies; and the steps are repeated in this way till backstoping of all chambers are completed. The upward and downward staged rock drilling, sublevel bottom cutting and ore removal subsequent backfilling mining method has the advantages that the rock drilling efficiency of the upward fan-shaped medium-depth blast holes is improved, the waste hole rate is lowered, efficient and safe bottom ore removal is achieved, and the mechanization level is high.

The invention discloses a house pillar changeable panel upward demix filling mining method. An ore body is divided into an ore house and an ore pillar arranged at intervals along the strike direction; a plurality of house pillars form a panel; a rock trackless preparation system is adopted and comprises a ramp, a rock slice drift, a demix ventilation connection and a rock concentrated ore removal winze; the mechanical upward demix filling mining method is adopted for mining, which mines a slice and fills the slice until the position required by the lower slice blasting compensation is reached. The backstopping order of the house pillars in the panel is the ore house first and then the ore pillar; all ore houses are subjected to upward backstopping at the same time; after a backstopping unit of the ore house is finished, the roof-contacted filling is carried out, and the mining of the ore house is temporarily stopped and the mining of the ore pillar is started; after two backstopping units of the ore pillar are finished, the roof-contacted filling is carried out, and the mining of the ore house is resumed; the abovementioned steps are repeated until the backstopping of the ore block is completed. The invention has the remarkable characteristics of improving the backstopping safety, reducing the resource dilution rate, lowering the support cost, increasing the enterprise profit, and the like.

The invention discloses an exploitation method of a thick oil type oil deposit, which comprises the following steps of: adding a chemical agent and hot fluid into an oil layer, enabling crude oil in the oil layer to be successfully seeped to an oil well and then enabling the crude oil to be raised to the ground from the inside of the oil well. The invention has the following advantages that: the invention can effectively solve the problem of phase precipitation of the asphaltene of the crude oil in the oil layer, can effectively exploit the thick oil type oil deposit, lowers the exploitation cost of the thick oil type oil deposit greatly, prolongs the production period of the oil well, increases the yield of the oil well, enhances the final recovery ratio of the oil deposit and has easy execution, high safety and reliability and obvious economic benefit.

The invention discloses a short text clustering and hotspot theme extraction method based on TF-IDF characteristics. The method includes the following steps of firstly, conducting Chinese word segmentation on short text samples, and screening out high-frequency vocabularies; secondly, automatically conducting TF-IDF characteristic extraction and generation on each short text sample on the basis of the screened-out high-frequency vocabularies, and establishing a whole sample characteristic vector spatial model; thirdly, reducing spatial dimensions of the samples through singular value decomposition (SVD); fourthly, clustering the short text samples through the combination of the cosine law and the k-means method, and finding potential hotspot themes in each cluster through a visual analysis means. By means of the method, the characteristic selection problem, the sample control dimension reduction problem and the clustering problem of short texts can be well solved; meanwhile, visual analysis on the clustering result can be achieved by means of the visual technology; finally, extraction and analysis are conducted on hotspot themes.

A multi-level anomaly detection method based on exponential smoothing, sliding window distribution statistics and an integrated learning model comprises the following steps of a statistic detection stage, an integrated learning training stage and an integrated learning classification stage, wherein in the statistic detection stage, a, a key feature set is determined according to the application scene; b, for discrete characteristics, a model is built through a sliding window distribution histogram, and a model is built through exponential smoothing for continuous characteristics; c, the observation features of all key features are input periodically; d, the process is ended. In the integrated learning training stage, a, a training data set is formed by marked normal and abnormal examples; b, a random forest classification model is trained. The method provides a general framework for anomaly detection problems comprising time sequence characteristics and complex behavior patterns and is suitable for online permanent detection, the random forest model is used in the integrated learning stage to achieve the advantages of parallelization and high generalization ability, and the method can be applied to multiple scenes like business violation detection in the telecom industry, credit card fraud detection in the financial industry and network attack detection.

The invention discloses a method for determining a fracturing index of a shale formation. The method comprises the following steps of determining a plurality of fracturing influence factors used for representing reservoir quality and fracturing engineering quality of the shale formation according to logging information; respectively determining weight coefficients of the various fracturing influence factors by using an analytic hierarchy process; and performing weighted summation on the various fracturing influence factors to obtain the fracturing index. By the method, the fracturing index is determined according to brittleness, stress states, fracture distribution, porosity, permeability and fracturing effects of the shale formation and rock physical properties of the formation, and the pressing ability of the shale formation can be described quantitatively.

The invention discloses an alarm correlation analysis method in a communication network, targeting topology research in a tree-hierarchy structure network. The method comprises the following steps: in accordance with the space-time relativity of malfunctions occurring to network nodes, defining space-time relativity of upper layer network nodes in the tree-hierarchy structure network; based on the space-time relativity of the upper layer network nodes, clustering upper layer nodes in the tree-hierarch network; based on clustering result, dividing the total alarm database into a plurality of sub alarm databases; based on the attributes of alarm items, such as alarm occurrence frequency, alarm priority and alarm failure type, determining weight of each alarm item; utilizing the weighted Apriori correlation rules algorithm, conducting correlation rules mining on each sub alarm database. The method of the invention aims at addressing the problems of alarm relativity analysis of tree-hierarchy structure, and can effectively mine alarm correlation rules of interest among a large amount of alarm information.

The invention discloses a tool wear condition prediction method of a numerical control machine tool based on a parallel deep neural network. A dynamometer, an acceleration sensor and an acoustic sensor are installed on a workbench and a fixture of the numerical control machine tool; a milling experiment is conducted, the cutting force and vibration and acoustic signals of a milling process are collected so as to obtain multisensor data, and the wear capacity of a tool is collected; pretreatment is performed so as to obtain training data and to-be-tested data; a parallel deep neural network model is established; the treated training data and the label of the wear capacity of the tool are input into an offline training model in the parallel deep neural network model; and the to-be-tested multisensor data are introduced into the trained model so as to predict the wear capacity of the tool in real time and on line. According to the method, the implied characteristics during tool processingof the numerical control machine tool are fully mined, and the wear capacity of the tool can be predicted in real time. The method has the advantage of wide applicability and can be widely applied tovarious numerical control machine tools.

The invention provides a Chinese short text subjective question automatic scoring method using an LSTM neural network. The method comprises the steps that firstly, an answer text is segmented, and thetext is converted into a word sequence; secondly, a vectorization expression of each word in the answer text is obtained, and an answer text mapping matrix is constructed; thirdly, the LSTM neural network is used for carrying out operation on the answer text mapping matrix, output of all or a part of hidden layers is obtained to obtain a semantic feature matrix of the answer text; fourthly, down-sampling is conducted on the semantic feature matrix by utilizing a pooling algorithm to obtain a semantic feature vector of the answer text; fifthly, the semantic feature vector of the answer text isgiven to a classifier, and the category of the answer text is predicted; sixthly, the many-to-one relationship between the category where the answer text belongs and the score is considered, and thescore of the answer text is determined according to preset mapping between the category and the score. According to the method, answer text semantic information can be effectively mined without depending on subjective question standard answers, and Chinese short text subjective question automatic scoring is achieved.

The invention relates to a new coalbed gas thermal production process which is characterized in that full hot coke is arranged on the lower part of a coalbed gas production well or a gasification well, a gasification agent is pressed in so as to ignite a hole bottom coalbed; and the heat released from the combustion of part of the coalbed can heat up the adjacent coalbed, and improve the analysis capability, permeability coefficient and osmotic pressure of the coalbed, thereby improving the gas yield of the coalbed.

The invention provides a method and system for recommending friend information in the social network. The method comprises the following steps that firstly, after recommending request information of a user is received, a social network information set is established according to operation behavior information and attribute information of users in the social network; secondly, a friend community set including the user and other K users is found in the social network information set; thirdly, behavior orientation information of the other K users in the friend community set is analyzed and recommended to the user. According to the method and system for recommending the friend information in the social network, the information of obvious friends is analyzed in a recommended result, the preference and influence of hidden friends of the user are also taken into consideration, and compared with an existing scheme, the method and system have the advantage that the produced recommended result is further improved on the aspect of precision.

The invention provides a webpage clustering method based on node property label propagation, which comprises the following steps: structuring a topological structure diagram model according to the link relationship of the webpage; building property feature vector for each node in the diagram model; initializing node labels in the network; iteratively updating the node labels according to the topological structure and the node property; and defining the condition of stopping updating the node labels. With the method, effective and high-quality webpage clustering can be realized within the nearly linear time only by extracting the topological relationship of the internet and the property information marking webpage characteristics without knowing priori knowledge such as the quantity and the scale of webpage groups and the like or predefining parameters such as the critical value and the like. The algorithm is simple in concept, easy to understand and realize and lower in time complexity, and can generate high-quality network clustering result, so as to apply to large-scale internet.

The invention discloses a personalized recommendation method based on a knowledge graph. The method has the advantages that the link relations among conceptual entities in the knowledge graph are utilized to measure the semantic association between every two optional nodes, a network representation learning method is used to obtain the representation vectors of the nodes in a network structure, and items are precisely recommended for users by calculating node similarity; the item recommendation based on knowledge graph feature learning is used for efficiently mining the entity features in theknowledge graph, user and recommendation object features can be well modelled, and multi-dimensional features are sufficiently utilized.

The invention discloses a semantic segmentation method and system for an RGB-D image. The semantic segmentation method comprises the steps: extracting RGB coding features and depth coding features ofan RGB-D image in multiple stages; inputting the RGB coding features and the depth coding features of each stage in the plurality of stages into an attention model to obtain each multi-mode fusion feature corresponding to each stage; extracting context semantic information of the multi-modal fusion features in the fifth stage by using a long short-term memory network; splicing the multi-modal fusion features and the context semantic information in the fifth stage to obtain context semantic features; and performing up-sampling on the context semantic features, and fusing the context semantic features with the multi-modal fusion features of the corresponding stage by using a jump connection mode to obtain a semantic segmentation map and a semantic segmentation model. By extracting RGB codingfeatures and depth coding features of the RGB-D image in multiple stages, the semantic segmentation method effectively utilizes color information and depth information of the RGB-D image, and effectively mines context semantic information of the image by using a long short-term memory network, so that the semantic segmentation accuracy of the RGB-D image is improved.

The invention discloses a drug repositioning method based on multi-information fusion and a random walk model. According to the method, disease-target-drug heterogeneous network is constructed through integrating existing disease data, drug data, target data, disease-drug associated data, disease-gene associated data and drug-target associated data; the basic random walk model is extended to the constructed heterogeneous network; and candidate therapeutic drugs are recommended for diseases through effectively utilizing global network information. The method disclosed by the invention is simple and effective; and compared with other methods and proved by tests on a standard data set, the method has good prediction performance in the aspect of drug repositioning.

The invention discloses a news topic timeline abstract generating method based on a breakthrough point, and the method is used for automatically and efficiently discovering significant instants and important events happening in the development process of a target news topic, thereby greatly assisting readers in understanding the evolution process of a news topic. The method comprises the following steps: (1) according to topic keywords input by users, downloading all news articles (searched by using search terms) from related news websites, and then carrying out pretreatment on the news articles; (2) establishing a topic-activity hidden Markov model for the activity variation trend of a target topic in each time slice, and deleting the time slices in which the target topic is not active; (3) carrying out modeling on a topic shift sequence in each time slice by using a topic-shift hidden Markov model; (4) extracting sentences (relevant to an important event happening on that day) as an abstract of the breakthrough point; and (5) outputting the timeline abstract of the target topic.

The invention relates to the technical field of social networks, in particular to a micro-blog hot word and hot topic mining system and method. The method includes the following steps that content data released in a micro-blog are preprocessed to acquire a candidate hot word sequence; according to the frequency of occurrence and suddenness of candidate hot words in a candidate hot word set at the current moment and in a given historical time window, the vitality of each candidate hot word is worked out, and a hot word set is formed by screening the candidate hot words; according to the hot word set formed by screening the candidate hot words, hot word correlation is worked out, and a hot word co-occurrence network is constructed; according to the hot word co-occurrence network, the hot word set is partitioned through the hot word clustering algorithm based on multi-label propagation to acquire a hot topic set. By means of the micro-blog hot word and hot topic mining system and method, efficient micro-blog hot word and hot topic mining is achieved, and mining precision and processing efficiency are improved.

The invention discloses a method and a device for realizing an association rule mining algorithm supporting a distributed computation. An HDFS (Hadoop Distributed File System) programming model is used to carry out two-stage analysis of a map function stage and a reduce function stage on the association rule mining algorithm, and the analysis steps comprises the following steps: step 1, a job scheduler is configured; step 2, a data set is read by a prior probability mapping module, and the data of the data set are converted by a map function into a value pair; step 3, the value pair processed in the step 2 is read by the prior probability reduction module, an ordering rule Top N containing an i item set is randomly generated by a reduce function, and the prior probability distribution value of a confidence coefficient is calculated at the same time; step 4, the same data set is read by a rule mapping module, and the data row of the data set is converted by the map function into the value pair; and step 5, the value pair processed in the step 4 and the prior probability distribution value in the step 3 are read by a rule reduction module, and the predication accuracy value of the ordering rule Top N is calculated by the reduce function. The method and the device for realizing the association rule mining algorithm supporting the distributed computation are mainly applied to the PA (Pridictive Apriori)-distribution type computing technology.

The invention provides a coal mining method of fully-mechanized caving mining with large mining height in a super-high seam, a coal caving process, reasonable type of a support, production capacity of equipment, comprehensive control measures of coal wall spalling of working face, a technical route for improving the recovery rate of the fully-mechanized caving face with the large mining height, and feasibility evaluation indexes (thickness of the seam, top coal cavability, coal wall stability and roof pressure strength) of the fully-mechanized caving mining with the large mining height, and further forms a whole set of fully-mechanized caving mining process technology with the large mining height. The coal mining method is that the coal cutting height is 4.2-4.8m; the length of the working face is 200-280m; and the web of a coal mining machine is 0.8-1m. The process technology can increase the coal cutting quantity by 43%, shorten the coal caving time by 9%, relatively reduce the coal caving quantity by 9%, improve the resource recovery rate by 5% and reduce the mining-caving ratio to below 1:3 by increasing the coal cutting height, and can further realize the relatively balanced production of mining and caving of the working face.

The invention discloses an urban route travel time forecasting method based on pattern matching. The method comprises the steps that firstly, historical traffic data are analyzed, and traffic patterns are dug from the historical data and include road segment chain patterns and intersection patterns, wherein the traffic data adopted in the method are floating vehicle data; secondly, a traffic management department or an expert establishes a default road segment chain traffic pattern and a default intersection traffic pattern; the departure point, the terminal point and the departure time of an input vehicle are provided, a corresponding traffic pattern is matched according to the current traffic condition and weighting is conducted to work out the travel time of all candidate routes. By the adoption of the method, the urban road segment chain traffic patterns and the urban intersection traffic patterns can be effectively and comprehensively considered, the travel time of the urban routes can be effectively predicted, the adopted method and technology are simple and practicable, operating conditions can be met easily, and the method can guide vehicles to select the travel routes reasonably, plays an important role in relieving urban traffic congestion and is easy to apply and popularize in large and medium-sized cities.

The invention discloses an intelligent fuzzy testing method and system based on vulnerability learning. The intelligent fuzzy testing system comprises a data preprocessing module, a bug prediction module and a bug-oriented fuzzy testing module, wherein the data preprocessing module is used for performing reverse analysis on a to-be-tested binary program to obtain a control flow chart of the to-be-tested binary program and performing feature extraction on each basic block in the control flow chart to obtain a feature vector of each basic block; the bug prediction module is used for predicting the probability of bugs exiting in each function in the program according to the control flow chart of the to-be-tested binary program; and the bug-oriented fuzzy testing module is used for testing theto-be-tested binary program and calculating input fitness scores by combining a certain input execution path, the probability of bugs existing in a function in the execution path and an execution result; input of a high-fitness score is used as a seed to perform heritable variation to generate next-generation input, and the to-be-tested binary program is tested cyclically till testing is over. Through the intelligent fuzzy testing system, the bugs in the binary program can be mined more efficiently.

The invention discloses a deception voice detection method based on a deep neural network. The method comprises the following steps: A, training and building a deception voice detection model based onthe deep neural network according to voice data, with the known authenticity, of a user, wherein the deception voice detection model has network parameters; and B, performing classification judgmenton test voices to be tested in the trained deception voice detection model with the network parameters so as to judge whether the test voices are real voices or deceptive voices. The method has the following advantages: novel unknown voice synthesis, voice conversion, record playback and other deceptive attacks are supported.

A talent service two-way recommendation method based on Internet big data comprises a data access module, a data matching calculation cluster and a data distribution module, wherein a job hunting andrecruitment behavior database and a resume position business database are arranged in the data access module; an active user cache, a data text analysis result cache, a job seeker matching arithmeticunit and a position matching arithmetic unit are arranged in the data matching calculation cluster; and a data distribution module is provided with a result database. The method comprises the specificimplementation steps of 1 data understanding; 2 data cleaning; 3 selecting and evaluating features; 4 calculating and storing; 5recommending;6 increasing and updating the data. According to the method, the behaviors of the user are captured, the talent and position information is collected in real time, the real-time accurate matching of mass resumes and position data is achieved, the positions more conforming to intentions of job seekers can be recommended to the user, and the service efficiency is greatly improved.

The invention provides a method and system for extracting natural gas hydrate through thermal jet flow. The method includes the steps that a combustion reaction device is lowered to a downhole target position of a natural gas hydrate reservoir, oxygen and fuel are injected into the downhole target position, the combustion reaction device is powered on, ignition is performed, combustion products are jet out of the combustion reaction device to form high-speed jet flow, the jet flow and heat generated during combustion act on the natural gas hydrate reservoir so that natural gas hydrate can be decomposed, and decomposed natural gas returns upwards to the ground; the amount of injected fuel is gradually reduced till injection stops, meanwhile part of the natural gas returning upwards to the ground and water are injected and mixed with oxygen for continuous combustion, and natural gas hydrate can be furthermore decomposed. The system at least comprises the combustion reaction device, a continuous oil tube inner tube, a continuous oil tube outer tube, a cable, a production sleeve, a water supply device, a fuel supplying device, an oxygen supplying device, a natural gas storing, collecting and supplying device and a pump set. The method and system for extracting natural gas hydrate through thermal jet flow have the advantages that extracting cost is low, extracting efficiency is high and the system structure is simple.

The invention discloses a multimode exploitation simulation device and method for different kinds of natural gas hydrate reservoirs. A production well is connected with a high pressure reactor througha rotary tube with an angle capable of being adjusted at will, and the high pressure reactor and the production well are both provided with visual windows. According to the multimode exploitation simulation device disclosed by the invention, production wells with different angles, including a vertical shaft, a horizontal shaft and an inclined shaft, can be obtained by adjusting the rotary tube. Multiple kinds of natural gas hydrate reservoirs can be simulated and synthesized in the high pressure reactor; a pressure reducing method, a heat injecting method, an inhibitor injecting method and acombined method can be simulated to exploit natural gas hydrate; the generation, decomposition and secondary generation situations of the hydrate and the gas-water flowing situation can be monitored in the whole experiment process and the water yield and the gas yield can be obtained. According to the multimode exploitation simulation device, a hydrate sample in the high pressure reactor is synthesized, natural gas hydrate exploitation simulation experiments can be systematically performed under different hydrate reservoirs, different production wells and different exploiting methods, and efficiency and safety evaluation can be performed for different hydrate exploiting modes.

The invention discloses a meta-path-based link prediction method for aligned heterogeneous social networks, mainly solves the problem of link prediction (mainly including friend relationship prediction and position prediction) in the aligned heterogeneous social networks, and relates to related technologies of the aligned heterogeneous social networks, feature selection, link prediction and the like. The method mainly comprises six steps of (a) modeling the aligned heterogeneous social networks; (b) automatically extracting a meta-path; (c) defining a meta-path-based eigenvalue calculation method; (d) dividing data sets according to timestamps; (e) proposing a two-stage gradual forward greedy feature selection algorithm for performing feature selection; and (f) training a decision tree classifier based on a feature selection result to perform the link prediction. Based on the method, the aligned heterogeneous social networks consisting of Foursquare and Twitter are subjected to the link prediction; and the method has practical application values for entity recommendation, accurate marketing, criminal gang discovery and the like in the social networks.