40 results about "Decision tree learning" patented technology

The invention is a computer implemented technique for id entifying anomalous data in a data set. The method uses cascaded k-Means clustering and the ID3 decision tree learning methods to characterize a training data set having data points with known characterization. The k-Means clustering method first partitions the training instances into k clusters using Euclidean distance similarity. On each training cluster, representing a density region of normal or anomaly instances, the invention builds an ID3 decision tree. The decision tree on each cluster refines the decision boundaries by learning the sub-groups within the cluster. A test data point is then subjected to the clustering and decision trees constructed form the training instances. To obtain a final decision on classification, the decisions of the k-Means and ID3 methods are combined using rules: (1) the Nearest-neighbor rule, and (2) the Nearest-consensus rule.

The invention discloses website operation state monitoring and abnormal detection based on MapReduce. By utilizing a MapReduce parallel programming model, the state monitoring and abnormal detection can be finished, an optimal information point is captured from massive journal files, and abnormal behaviours in the access process are effectively and accurately captured by adopting an effective strategy. A website journal is analyzed by adopting four parallel strategies, namely state monitoring, characteristic abnormal detection, flow peak value detection and decision tree learning access rules, so that the monitoring and abnormal access detection of the website operation state can be achieved.

A system and method facilitating lithography defect solution generation is provided. The invention includes a defect solution component and a defect alert component. The defect solution component provides potential solution(s) to a defect within the lithography process utilizing artificial intelligence technique(s) (e.g., Bayesian learning methods that perform analysis over alternative dependent structures and apply a score, Bayesian classifiers and other statistical classifiers, including decision tree learning methods, support vector machines, linear and non-linear regression and/or neural network).

The invention relates to a method and a device for classifying network traffic on line based on decision tree high-speed parallel processing. The method comprises the following steps of: performing acquisition, distribution and manual classification on early real traffic data, extracting the packet characteristics of an early transmission control protocol (TCP) stream set, establishing a decision tree classification model, converting a data structure, performing distribution and class judgment on a data packet to be classified, tagging a current data packet, extracting the packet characteristics of a TCP stream to be classified, and searching for a decision tree. The device comprises a decision tree construction module, a structure conversion module, a classification result processing module, a medium access control (MAC) layer processing module, a data packet polling management module, a distribution judgment module, a traffic information extraction and tagging module, and a decision tree searching module. The method and the device are low in algorithm complexity and high in processing speed, classification accuracy and stability, and can be used for equipment and systems with requirements for online traffic classification in a high speed backbone network; and online classification can be realized.

A multi-task machine learning method is performed to generate a multi-task (MT) predictor for a set of tasks including at least two tasks. The machine learning method includes: learning a multi-task decision tree (MT-DT) including learning decision rules for nodes of the MT-DT that optimize an aggregate information gain (IG) that aggregates single-task IG values for tasks of the set of tasks; and constructing the MT predictor based on the learned MT-DT. In some embodiments the aggregate IG is the largest single-task IG value of the single-task IG values. In some embodiments the machine learning method includes repeating the MT-DT learning operation for different subsets of a training set to generate a set of learned MT-DT's, and the constructing comprises constructing the MT predictor as a weighted combination of outputs of the set of MT-DT's.

The invention discloses an automatic image annotation and translation method based on decision tree learning. A new image is automatically annotated, and a text word list with a visualized content is translated by a machine so as to realize the machine retrieval of image data, comprising a training annotation image set and image automatic annotations, wherein the training annotation image set utilizes an image segmentation algorithm to segment a training image set into sub areas and extract low-level visual features of each sub area; the feature data is discretized, and then the training annotation image set is classified by a clustering algorithm based on a low-level feature discrete value to construct a semantic dictionary; the low-level feature discrete value is used as an input attribute of the decision tree learning; and self training learning is carried out on the constructed dictionary by a decision tree machine learning corresponding to preset semantic concepts so as to generate a decision tree and obtain a corresponding decision rule. The training annotation image set has expandability and robustness and can improve the recall ratio and the precision ratio of the retrieval when the training annotation image set is applied to semantic image retrievals.

A computerized decision tree training system may include a distributed control processing unit configured to receive input of training data for training a decision tree. The system may further include a plurality of data batch processing units, each data batch processing unit being configured to evaluate each of a plurality of split functions of a decision tree for respective data batch of the training data, to thereby compute a partial histogram for each split function, for each datum in the data batch. The system may further include a plurality of node batch processing units configured to aggregate the associated partial histograms for each split function to form an aggregated histogram for each split function for each of a subset of frontier tree nodes and to determine a selected split function for each frontier tree node by computing the split function that produces highest information gain for the frontier tree node.

The invention provides a facial beauty classification method of a woman image by adopting a computer, wherein, a C4.5 decision tree learning method is used for building a classifier, and 17-dimensional characteristic quantity is used as a superficial characteristic quantity of facial beauty degree to classify the facial beauty of the woman image. The method in the invention is simple and effective, proving the practicality of intelligent perception of facial beauty by utilizing the method of machine learning, in particular having higher and exciting classification accuracy degree towards the perception of obvious beauty; the test result simultaneously proves that obvious beauty really has objectively existent and quantitative regulations.

The invention discloses a nighttime road vehicle detection method based on luminance variance characteristics. For vehicle detection of nighttime road monitoring, a road surveillance camera is used as video source input, threshold segmentation of images is carried out using a histogram two-peak method, and nighttime vehicle images are preprocessed based on the atmospheric scattering principle; then, characteristic difference is statistically processed according to the data of luminance variance of a vehicle light and reflected light in an image, the reflected light and the vehicle light in the image are distinguished using a learning method based on a decision tree, and the reflected light in the image is eliminated; and finally, nighttime vehicle light detection is realized. The accuracy rate of nighttime vehicle detection and the time performance of detection are improved.

The invention provides a classification decision tree-based intelligent content distribution classification method, and belongs to the field of content management systems and data mining. A universal system architecture for the method comprises a content data acquisition module, a classification strategy module and a classification decision tree training module. According to the method, a supervised learning method for machine learning is combined, existing content information in a content management system is used as a training dataset, and the training dataset is used for generating a classifier on the basis of a decision tree learning algorithm; all distributed contents are based on the same distribution way, and types of the distributed contents are determined by the classifier according to characteristic values of the distributed contents after the contents are distributed, so that the whole distribution process is finished. According to the method, user experiences in content distribution are improved; the types of the distributed contents are determined by the system rather than selected by a user, so that a distribution flow is more intelligent.

The invention discloses a prediction method for a correlation between circular RNA and a disease based on a gradient enhancement decision-making tree. A circular RNA-disease relationship network is converted into an undirected graph, the circular RNA base sequence similarity, the function annotation semantic similarity and the expression similarity are calculated, and the disease function similarity and the semantic similarity are calculated; a multi-network integration algorithm is adopted for integrating a plurality of kinds of circular RNA similarity networks and conducting weighted averageintegration on a disease similarity network, statistical characteristics of the integrated circular RNA and disease similarity network and the circular RNA-disease relationship network are extracted,and the integrated circular RNA and disease similarity network is converted into an unweighted graph-related characteristic, a circular RNA base sequence characteristic and a circular RNA-disease relationship network implicit vector characteristic; a gradient enhancement decision-making tree learning machine is trained, and a potential circular RNA-disease relationship is predicted. By means of the method, the potential circular RNA-disease relationship can be accurately predicted; and the prediction accuracy of the circular RNA-disease relationship is improved.

The present invention discloses a decision tree mechanism-based multi-copy routing method in vehicular Ad hoc networks, and belongs to the technical field of vehicular Ad hoc networks. In the method, a machine is used for learning a learning method in a decision tree C4.5; a vehicle firstly collects historical data, then performs integrated processing on the collected historical data, and performs classification and grading on selected corresponding attributes by using the learning method of the decision tree to generate a decision rule tree; and the constructed decision rule tree is used in transmitting between an initial message and a message copy to enable the transmission of the message to be relatively directional and purposeful. The method is used for solving the problem of overlarge network cost brought from blind flooding retransmission of the message copy number in multi-copy routing, and simultaneously solving the problems of low delivering rate and large time delay as direct transmitting is passively adopted; and compared with a conventional method, the method disclosed by the present invention has higher performances such as timeliness and reliability, and is more suitable for the vehicular Ad hoc networks having high-dynamic topology and a relatively high network density.

The invention provides an emergency evacuation system and method for people in a key area based on a traffic flow model, particularly relates to an emergency early warning and evacuation method for people in the key area, belongs to the field of road traffic planning. and aims to solve the problems that in the prior art, obvious errors exist in the traffic demand simulation process, and an existing prediction model is long in prediction time. The system comprises a real-time people flow monitoring and early warning layer and a scheduling scheme determination layer for traffic demand prediction. Real-time people flow data are accessed to realize real-time people flow monitoring in key areas, real-time people flow and people flow early warning thresholds are compared, and if the thresholds are exceeded, a scheduling scheme is triggered. The people flow early warning threshold value of each key area can be obtained by constructing a simple decision tree learning model; According to the method, massive historical data can be processed, correlation analysis of various traffic characteristics is supported, and the influence of various factors on travel demands of people is reflected more truly; and the model established based on a decision tree has the characteristics of simple algorithm and accurate prediction.

The invention provides a method for generating a lane changing rule of an automatic driving automobile. The method comprises the following steps: S1, acquiring traffic data of a selected road area; S2, screening from the traffic data according to lane changing behavior influence factors to obtain sample data; S3, calculating the grey entropy correlation degree of each lane changing behavior influence factor according to the sample data, and obtaining the condition attribute of the lane changing behavior; and S4, constructing a lane changing behavior decision tree based on the information gainand gain rate of the condition attribute of the lane changing behavior, and generating a lane changing rule according to the lane changing behavior decision tree. According to the method, the machinelearning-decision tree-learning is effectively utilized to simulate the human driving behavior, so that the uninterpretability of the driving behavior learned by the neural network algorithm is avoided, and meanwhile, the over-constraint phenomenon and the under-constraint phenomenon of the lane changing behavior caused by manually setting rules are also avoided.

The invention relates to an emotion recognition method based on a deep fuzzy forest. The method comprises the following steps: S1, collecting electroencephalogram signals; S2, preprocessing the electroencephalogram signals to remove noise; S3, inputting the electroencephalogram signals into a pre-trained deep fuzzy forest model to obtain an emotion recognition result. In the step S3, multi-granularity scanning is adopted for the deep fuzzy forest model, and probability vectors of electroencephalogram signal features are obtained from the electroencephalogram signals to serve as input of a cascade forest; and a cascade forest is adopted to recognize the probability vectors of the electroencephalogram signal features to obtain an emotion recognition result, and the multi-granularity scanningand the cascade forest are both constructed by adopting a fuzzy decision tree. Compared with the prior art, the emotion recognition method combines a fuzzy set theory with a traditional decision treelearning strategy, and has the advantages of originality, high recognition degree, few parameters, capability of being used for a small sample data set, accurate and reliable result and the like.

The invention discloses a steak grade division method based on decision tree induction learning, which mainly comprises the following steps: A. establishing a steak grade division model, uniformly collecting a cross-sectional image of beef tenderloin, preprocessing the image, and eliminating image background; B, calculate that edge distribution of the steak image, carrying out level set evolutionon the steak image to obtain an image gradient, thereby extract the edge of the image and carrying out muscle and fat segmentation; C. establishing pixel gray geometric moments of steak marbling, andextracting the pixel gray moments as marbling image features; D. Establish the correlation function of marbling feature as the test attribute to construct the decision tree, carry on the inductive learning, classify the steak automatically, finish the steak grade division. The method has good stability and robustness, can achieve large-scale, standardized, batch, muscle and fat segmentation accurate, objective evaluation results, evaluation speed, high accuracy, so as to ensure the efficient and continuous supply of goods.

A learning device is configured to perform learning of a decision tree by gradient boosting. The learning device includes a plurality of learning units and a plurality of model memories. The plurality of learning units are configured to perform learning of the decision tree using learning data divided to be stored in a plurality of data memories. The plurality of model memories are each configured to store data of the decision tree learned by corresponding one of the plurality of learning units.

A mechanism is provided in a data processing system for distributed tree learning. A source processing instance distributes data record instances to a plurality of model update processing items. The plurality of model update processing items determine candidate leaf splitting actions in a decision tree in parallel based on the data record instances. The plurality of model update processing items send the candidate leaf splitting actions to a plurality of conflict resolve processing items. The plurality of conflict resolve processing items identifies conflict leaf splitting actions. The plurality of conflict resolve processing items applies tree structure changes to the decision tree in the plurality of model update processing items.