3663 results about "Cluster analysis" patented technology

Systems and methods for discovery and classification of denial of service attacks in a distributed computing system may employ local agents on nodes thereof to detect resource-related events. An information later agent may determine if events indicate attacks, perform clustering analysis to determine if they represent known or unknown attack patterns, classify the attacks, and initiate appropriate responses to prevent and / or mitigate the attack, including sending warnings and / or modifying resource pool(s). The information layer agent may consult a knowledge base comprising information associated with known attack patterns, including state-action mappings. An attack tree model and an overlay network (over which detection and / or response messages may be sent) may be constructed for the distributed system. They may be dynamically modified in response to changes in system configuration, state, and / or workload. Reinforcement learning may be applied to the tuning of attack detection and classification techniques and to the identification of appropriate responses.

The present invention provides a communication and analysis system that can manage data operations with a vehicle centric system with a planned route path. A vehicle that is in a communications link with a network can also manage its activities based on real time, historical and predictive knowledge, without having this all knowledge processing on board. Such data processing would include a cluster analysis of geo-spatial, internal functions and operator specific requirements. The rule based system would also incorporate the use patterns of a specific vehicle, or a specific user. A vehicle operator or vehicle multiple operators could share or upload information that would assist with efficient data processing and display, including fuel conservation and time management. Cluster weighting patterns can be assigned based on activities as efficient operation, safe travel and navigation.

A server-side system that detects and classifies malware and other types of undesirable processes and events operating on network connected devices through the analysis of information collected from said network connected devices. The system receives information over a network connection and collects information that is identified as being anomalous. The collected information is analyzed by system process that can group data based on optimally suited cluster analysis methods. Upon clustering the information, the system can correlate an anomalous event to device status, interaction, and various elements that constitute environmental data in order to identify a pattern of behavior associated with a known or unknown strain of malware. The system further interprets the clustered information to extrapolate propagation characteristics of the strain of malware and determine a potential response action.

Described invention and its embodiments, in part, facilitate discovery of ‘Most Recent Common Ancestors’ in the family trees between a massive plurality of individuals who have been predicted to be related according to amount of deoxyribonucleic acids (DNA) shared as determined from a plurality of 3rd party genome sequencing and matching systems. This facilitation is enabled through a holistic set of distributed software Agents running, in part, a plurality of cooperating Machine Learning systems, such as smart evolutionary algorithms, custom classification algorithms, cluster analysis and geo-temporal proximity analysis, which in part, enable and rely on a system of Knowledge Management applied to manually input and data-mined evidences and hierarchical clusters, quality metrics, fuzzy logic constraints and Bayesian network inspired inference sharing spanning across and between all data available on personal family trees or system created virtual trees, and employing all available data regarding the genome-matching results of Users associated to those trees, and all available historical data influencing the subjects in the trees, which are represented in a form of Competitive Learning network. Derivative results of this system include, in part, automated clustering and association of phenotypes to genotypes, automated recreation of ancestor partial genomes from accumulated DNA from triangulations and the traits correlated to that DNA, and a system of cognitive computing based on distributed neural networks with mobile Agents mediating activation according to connection weights.

Determining a semantic relationship is disclosed. Source content is received. Cluster analysis is performed at least in part by using at least a portion of the source content. At least a portion of a result of the cluster analysis is used to determine the semantic relationship between two or more content elements comprising the source content.

The invention relates to a multiple-template matching identity recognition method based on an ECG (Electrocardiogram) under an electrocardiogram abnormality state, and belongs to the technical field of biological characteristic identity recognition. The ECG data of a user to be recognized is compared with the data of a registered user in a template library to obtain an identity recognition result. The key technology of the method comprises the following steps: carrying out electrocardiosignal preprocessing for eliminating noise interference; carrying out electrocardiosignal decomposition to separate an electrocardiogram waveform of each period; carrying out standardized processing for independently achieving standardization on time and amplitude scales; carrying out characteristic extraction: in the step, characteristics are extracted by wavelet transform, and clustering analysis is carried out by an ISODATA (Iterative Self-organizing Data Analysis Techniques Algorithm) so as to construct an ECC template library; and carrying out correlation analysis: in the step, correlation between ECG test data and each template is calculated, an optimal matching template is selected, and finally, an identity recognition result is obtained. The multiple-template matching identity recognition method provided by the invention utilizes the intrinsic electrocardiosignal of a human body to recognize an identity, and the ECG data under the abnormality state is considered.

The invention discloses a clustering analysis and decision tree algorithm-based truck loading work time prediction model. A clustering analysis and decision tree mixed algorithm is introduced, factors influencing inventory control are abstracted out, related historical data serves as a training sample, and finally the truck loading work time can be effectively predicted by using a trained decision tree data model; and the historical data of truck loading is deeply mined by utilizing a data mining technology based on a demand, and an available, easy-to-use and high-accuracy data model is generated. The clustering analysis and the decision tree algorithm are combined and complement each other, so that the accuracy of the data model is improved; an optimization policy is adopted for an original decision tree algorithm under the condition of establishing a simple and accurate data model, so that the calculation amount is reduced and the algorithm efficiency is improved; and through the data model, a relatively accurate time interval of cargo loading can be predicted and used for better manual decision-making.

A portfolio of entertainment project such that the risk and return available to investors is attractive compared to other investments. Risk and return for a portfolio of entertainment projects is predicted based on historical performance of past “similar” projects. In one implementation, characteristics that are predictive of a project's revenue are determined by performing a cluster analysis of historical revenues from past projects. Projects in the portfolio are classified into various segments based on these predictive characteristics. Projects are selected to contruct a portfolio. The risk and return for the portfolio is calculated according to a risk-return model that is based on historical risk and revenues for past projects in the same segment and further based on historical covariance of revenue for past projects in different segments.

A method for analyzing types of water sources based on natural geographical feature, the method includes: collecting and processing remote sensing image data of target area, and obtaining maximum and minimum value of an annual vegetation index; subtracting the minimum value from the maximum value to obtain maximum variation range of annual vegetation index; extracting topography factors from a digital elevation model in target area; obtaining a natural vegetation area in target area; carrying out a normalization processing for the maximum variation range and the topography factors in this natural regions, and obtaining landform zones and situation of plant growth of different zones in the natural vegetation area by spatial cluster analysis in ArcGIS; obtaining a precipitation of landform zones in the growing season and the distances between the landform zones and the water sources, and obtaining the zones for the types of water sources based on natural geographical features.

The invention discloses a real-time monitoring method for the energy consumption of a public building based on data mining, belonging to the technical field of building energy saving. The method disclosed by the invention comprises the following steps: S1. establishing a building energy consumption mode judgment tree; S2. collecting building energy consumption data in real time; and S3. judging whether the current building energy consumption data are energy consumption abnormal points or not, carrying out the mode matching on the current building energy consumption data and the building energy consumption mode judgment tree and judging whether the current building energy consumption data are isolated points or not. In the method disclosed by the invention, the specific energy consumption mode of the building is identified by carrying out the cluster analysis on the historical energy consumption data; the building energy consumption mode judgment tree is obtained by classifying the data; the mode matching is carried out on the energy consumption data which are dynamically collected in the real-time monitoring course for the energy consumption of the building; and the isolated point analysis is carried out on the energy consumption data and the historical data which have the same mode, thereby judging whether the current building energy consumption data are abnormal or not. The method disclosed by the invention has the characteristics of good real-time characteristic, generality and robustness.

The invention discloses a 3D terrain imaging system of the interferometric synthetic aperture radar (InSAR) and an elevation mapping method thereof, which mainly solve the problems that the existing InSAR has bad imaging pragmaticality and can not implement 3D elevation mapping on the fast-changing terrain and the transilient terrain. The system comprises three sub-aperture antennas, a radar transmitter, a radar receiver and an imaging data processor; the imaging signal processor comprises a SAR image processing unit and an InSAR image processing unit. The invention receives radar echo through the three sub-apertures, then conducts SAR imaging process on the radar echo respectively received by the three sub-apertures, and then conducts InSAR imaging process on the obtained SAR complex pattern, wherein the InSAR imaging process comprises image registration, phase filtering and phase unfolding based on cluster analysis. The processed InSAR phase unfolded image is processed with an elevation inversion to recover a three dimensional digital elevation map. The invention has the advantages of wide adaptability to mapped terrains, and high imaging effectiveness, therefore, the invention can be used in the mapping of the 3D terrain.

The invention provides a video recommendation system and a method thereof. The video recommendation system and the method thereof comprises: acquiring multi-source data including personal information, social network information and video classification information of users by a information acquisition module; presetting the multi-source data by a data pre-processing module; storing the multi-source data after preprocessing by a user database which is established by a data storage module; enabling an analytical module of the characteristics of users to acquire feature and mood information of the users according to microblog data sent by the users and enabling an analytical module of the social network information to analyze friend groups of the users to acquire friend circles; enabling a video recommendation module to select favorable videos of the users according to the feature, mood and friend circles; enabling a front-end display module to display the selected videos for the users.

The invention relates to a design method for an integrated energy system with source-load-storage coordination and interaction. A lot of source-load output scenes are generated via Latin hypercube sampling, and a few of specific probability of planning scenes in which original scene characteristics are reserved are obtained via cluster analysis combined with a scene reduction method. An installedcost model and an operation cost model are established for each piece of equipment, an upper-level optimization model is optimized to minimize the sum of the annual investment conversion cost and theannual operation cost in different probability scenes, a lower-level optimization model is optimized to minimize the annual operation cost in different probability scenes a double-layer planning modelof the integrated energy system with source-load-storage coordination and interaction is established, and different reliable and safe operation constraints are met. The upper-level and lower-level optimization models are solved by an element model global optimization algorithm and a second-order conic optimization method respectively, solving is carried out by interactive iteration till convergence, and the installed capacity and optimal operation scheme of each piece of equipment of the integrated energy system are obtained by optimization. The method can be applied to capacity planning andoptimized design of the multiple energy complementary integrated energy system.

The invention discloses a method for solving a multiple-depot logistics transportation vehicle routing problem. The method comprises steps as follows: inputting multiple-depot problem basic parameters based on real-time traffic information, establishing a multiple-depot logistics transportation scheduling mathematic model based on the real-time traffic information, adopting a clustering analysis method, introducing the particle swarm optimization algorithm to adjust and optimize ant colony algorithm pheromones, optimizing ant colony algorithm heuristic factors with the particle swarm optimization algorithm, solving an optimal distribution route, and establishing a mathematic model according to the multiple-depot logistics transportation vehicle routing problem based on the real-time traffic information; taking distances between clients and parking lots as main factors, performing area division on the clients and the parking lots with the clustering analysis method, and converting a multiple-depot problem into a single-depot problem; introducing the particle swarm optimization algorithm to improve the ant colony algorithm to solve the model. The method has the better global and local optimization capacity and has higher efficiency and stability when solving the multiple-depot problem.

The invention discloses a prediction method of driving risk based on a hidden Markov model. The method comprises the steps of (1) classifying driving risk states through a cluster analysis method based on vehicle operating characteristics, (2) estimating the influences of a driver behavior and surrounding traffic environment characteristics on a transition probability between driving risk states through multiple logistical models for different driving risk states, (3) with a risk state as a hidden state, with an actual observed vehicle movement variable as a state output value, with multiple logistic model parameters as parameter initial values of a state transition probability matrix, establishing a hidden Markov chain model that reflects the evolution rule of driving states, and (4) obtaining the vehicle operating characteristics in real time and predicting a future risk state based on the hidden Markov chain model. According to the method, the hidden Markov model which can reflect the above characteristic real-time change and has a variable state transition probability is established, the accuracy and prediction accuracy of a driving risk model are improved, and the real-time requirements of anti-collision warning can be satisfied.

The invention discloses a method and a system for life mode analysis based on mobile device sensor data. The method comprises the steps: initial data are collected from each sensor of a mobile device; data preprocessing is conducted on the initial data according to data characteristics and energy consumption characteristics of each sensor and a data sequence is obtained; a resident point sequence is obtained according to a resident point detection mode; cluster analysis is conducted on the resident point sequence so that a site historical sequence is obtained; interest point search is conducted on each piece of data in the site historical sequence and site history is marked; and whether identity of a user is known or not is judged, if the answer is negative, the identity of the user is deduced according to interest point marks. The method and the system for the life mode analysis based on the mobile device sensor data reduce data mining and analyzing cost, improve using flexibility, have transportability and are convenient to use and capable of improving experience of the user.

The invention is applicable to the field of computer technology and provides a generation method and device for a decision network model of vehicle automatic driving. The method comprises the steps that a sample triad corresponding to each test moment is generated according to vehicle state information collected at each test moment, a preset vehicle movement set and a preset return function, all the sample triads are stored as sample data in a pre-established experience database, and all the sample data is subjected to clustering analysis; training samples are uniformly collected from each cluster obtained after clustering analysis of the experience database according to a preset sampling scale value, and a return accumulated value of each training sample is calculated; and according to all the training samples, the return accumulated value of each training sample and a preset deep learning algorithm, training is performed to obtain the decision network model of vehicle automatic driving. Therefore, the training efficiency of the decision network model and the generalization ability of the decision network model are effectively improved.

A method for categorizing body shape is provided comprising the steps of providing a data set of body shape-defining measurements of a portion of the body of interest from a plurality of subjects' bodies, wherein the measurements define a silhouette and profile (front and side) perspectives of the portion of the body of interest; conducting a principal component (PC) analysis of the data set of measurements to calculate and generate PC scores; conducting cluster analysis using the PC scores as independent variables to produce cluster analysis results; and establishing one or more body shape categories from the cluster analysis results, thereby categorizing body shapes of the plurality of subjects. A shape prototyping system is also provided for designing a custom fit garment for an individual subject, the system being based on the method for categorizing body shape.

A system and method for selecting a training data set from a set of multidimensional geophysical input data samples for training a model to predict target data. The input data may be data sets produced by a pulsed neutron logging tool at multiple depth points in a cases well. Target data may be responses of an open hole logging tool. The input data is divided into clusters. Actual target data from the training well is linked to the clusters. The linked clusters are analyzed for variance, etc. and fuzzy inference is used to select a portion of each cluster to include in a training set. The reduced set is used to train a model, such as an artificial neural network. The trained model may then be used to produce synthetic open hole logs in response to inputs of cased hole log data.

An adaptive network fault location and analysis method based on logs is principally composed of a log preprocessing method based on priority, a log event cluster analysis method based on time-series and a multidimensional log statistical analysis method. Magnanimity of log information is classified to store according to priority, and then key event types namely 'policy' are filtered out according to log event cluster analysis method based on time-series, subsequently the key log information is displayed on the interface by using multidimensional log statistical analysis method, and an alarm judging mechanism is triggered to alarm in real time. The method has favorable expandability and accuracy, and is easy to butt-joint with related application interfaces of operators.

A method for identifying free spaces (parking not permitted) and / or permitted parking areas, vehicles transmitting pieces of information about possible parking spaces (PPS) to a central computer facility (CCF). Positions of PPS are detected with vehicle surroundings sensors, and the detected PPS are evaluated based on the data collected, a categorization being performed for recording the PPS, with positions, in a CCF database and evaluating the data using a cluster analysis. When the analysis is performed, PPS are assigned to a street portion, a function is assigned to the street portion, which is given by the quotient of the frequency of PPS detections in a certain position along the street portion and the number of vehicle passages through the street portion and a weighting factor from the evaluation. A free space is inferred when the function value is greater than a predefined second limiting value and / or a parking area is inferred when the function value is within a predefined range. Also described is a device for assisting a driver, a central computer facility, and a related computer program.

The invention discloses a network hot event detection method based on text classification and clustering analysis. The method solves the problem that the efficiency and accuracy rate of the existing network hot event detection method based on clustering analysis need to be improved. The method comprises the steps that feature words are respectively selected for various classes of files through feature extraction and feature selection by utilizing a training corpus; each training text and test text are represented as vectors in all of the feature spaces by utilizing a vector space model method, and the weight of each dimension of the vectors is determined by utilizing a TF-IDF (term frequency-inverse document frequency) method, and then each test text is classified; the classified test texts in different classes are respectively subjected to clustering analysis, so the hot cluster of each class is obtained, the feature word representing the hot event is obtained through further analysis, and then the word property and other aspects of each feature word are analyzed; the description of each hot event is generated by utilizing relevant language knowledge and necessary linguistic organization. With the network hot event detection method based on text classification and clustering analysis, the detection efficiency and accuracy rate of hot events can be effectively improved.

The invention discloses a method for identifying flame by video, a method for monitoring fire, and a system thereof. The method for monitoring fire comprises the following steps: acquiring video image, establishing a background model, and detecting prospect target in the video image; transmitting the image of the prospect target to a server end; judging which target points of the prospect target are color points of flame according to the color model of the flame established by training image; carrying out cluster analysis on the color points of the flame of the prospect target to acquire a flame area of the prospect target, and matching the prospect targets of the front and rear frame images; judging whether flame occurs in the flame area; when the matching relation exceeds the set threshold value; carrying out verification of a alarming verification mechanism when judging that flame occurs in the flame area; and when satisfying the warning condition, performing warning. The inventionhas wider range of application without limiting distance of scene, and achieves intelligentized processing.

In a user-activity identification technique, a user's actions are monitored while the user is using a computer. While these user actions are associated with user activities, the user activities are initially unspecified, so the tracked user actions constitute unsupervised data. Then, the tracked user actions are aggregated into subsets (for example, using clustering analysis), and user-activity classifications for the subsets (such as activity labels) are provided by the user, so the subsets constitute supervised data. Subsequently, when additional user actions (which are associated with one or more initially unspecified current user activities) are tracked, they can be associated with one or more of the classified subsets. For example, information about the additional user actions can be mapped in real time (or near real time) to one or more of the subsets using a supervised learning technique. In this way, the one or more current user activities can be identified.

Systems and methods for discovery and classification of denial of service attacks in a distributed computing system may employ local agents on nodes thereof to detect resource-related events. An information later agent may determine if events indicate attacks, perform clustering analysis to determine if they represent known or unknown attack patterns, classify the attacks, and initiate appropriate responses to prevent and / or mitigate the attack, including sending warnings and / or modifying resource pool(s). The information layer agent may consult a knowledge base comprising information associated with known attack patterns, including state-action mappings. An attack tree model and an overlay network (over which detection and / or response messages may be sent) may be constructed for the distributed system. They may be dynamically modified in response to changes in system configuration, state, and / or workload. Reinforcement learning may be applied to the tuning of attack detection and classification techniques and to the identification of appropriate responses.

The invention discloses an internet public opinion analysis method. The method includes the following steps: determining a public opinion event, and collecting public opinion articles related to the public opinion event; preprocessing the collected public opinion articles, and obtaining a vocabulary collection in the public opinion articles to characterize the public opinion articles; performing clustering analysis on the vocabulary collection by adopting a clustering algorithm, generating a plurality of viewpoints of the public opinion event, and calculating a word vector of the viewpoints; extracting a core topic from the vocabulary collection contained in the viewpoints; calculating an emotional score of the viewpoints through an emotional scoring model, and calculating the level of popularity of the viewpoints; and calculating a public opinion index of the viewpoints according to the emotional score and the level of popularity, and determining that the viewpoint of which the absolute value of the public opinion index is greater than a preset threshold is an abnormal viewpoint, generating early warning information according to the abnormal viewpoint and the core topic, and outputting the early warning information. The invention also provides an internet public opinion analysis device and a computer readable storage medium. According to the scheme of the invention, the monitoring and early warning capabilities for public opinions can be improved.

The invention discloses a printing image defect detection method, which comprises the following steps: carrying out real-time image model learning in a gray scale region and a gradient region by aiming at specific images on a large image; comparing the large image to be detected to the a gray scale region model and a gradient region model established during learning; realizing small-dimension strong-contrast defect detection through Blob cluster analysis; when no small-dimension strong-contrast defect is detected, dividing the large image into sub regions, and respectively calculating image integrated features; adopting a variable threshold method for carrying out threshold division on each sub region; carrying out Blob cluster analysis on divided images; and realizing the large-area weak-contrast defect detection. Compared with the prior art, the invention does not rely on a reference template, is not sensitive on the real-time imaging brightness change, can overcome the defects of missing detection, error detection and poor self adaptation of the template by using a reference template comparison detection method, and can simultaneously solve the problems of eliminating tiny wrinkles and blackspots in printing and papermaking industries.