this paper presents a new approach to constructing a neural tree with partial incremental learning capability. the proposed neural tree, called a (q) under bar uadratic-ne (u) under bar ron-b (a) under bar sed (n) und...
详细信息
ISBN:
(纸本)9781424409723
this paper presents a new approach to constructing a neural tree with partial incremental learning capability. the proposed neural tree, called a (q) under bar uadratic-ne (u) under bar ron-b (a) under bar sed (n) under bar eural (t) under bar ree (QUANT), is a tree structured neural network composed of neurons with quadratic neural-type junctions for pattern classification. the proposed QUANT integrates t e advantages of decision trees and neural networks. Via a batch-mode training algorithm, the QUANT grows a neural tree containing quadratic neurons in its nodes. these quadratic neurons recursively partition the feature space into hyper-ellipsoidal-shaped sub:-regions. the QUANT has the partial incremental capability so that it does not need to re-construct a new neural tree to accommodate new training data whenever new data are introduced to a trained QUANT. To demonstrate the performance of the proposed QUANT, several patternrecognition problems were tested.
Pair data is a kind of data, which consists of two correlative data components. Book title and its author, product name and its price, bilingual translation term and Chinese couplet (a unit of verse consisting of two ...
详细信息
ISBN:
(纸本)9781424409723
Pair data is a kind of data, which consists of two correlative data components. Book title and its author, product name and its price, bilingual translation term and Chinese couplet (a unit of verse consisting of two successive lines) are of this type data. In this paper, based on the observation that pair data tend to co-occur in the same block of the same web page following similar patterns, we propose a new approach to extract the collective pair data. A recursive process is used to extract collective pair data from Web. An automatic algorithm of discovering repeated patterns based on a data structure called PAT tree is proposed to discover all repeated patterns first, then all these repeated patterns are ranked with a ranking SVM to get the trusty pair data extraction patterns. Finally the patterns are transformed with some predefined surface pattern classes anti then applied to extract collective pair data. Experimental results demonstrate our model gains higher extraction precision and recall than previous approach.
People often try to smooth or eliminate load outliers all together in traditional power load forecasting. this, however, could result in the loss of important hidden information. In other words, the power load outlier...
详细信息
ISBN:
(纸本)9781424410651
People often try to smooth or eliminate load outliers all together in traditional power load forecasting. this, however, could result in the loss of important hidden information. In other words, the power load outliers themselves may be particular important. Hence there is a beforehand estimate to change and characteristic of power load, is a precondition of power system carry through economy dispatch, reduce production cost and prevent widespread blackout or collapse on electric system. In this paper propose a novel method for special periods power peak load detection, mining and forecasting. It incorporates the characteristic of high level load and maximum peak load analysis with optimum forecasting algorithm based on support vector machine. the validity of the method is proved by real data calculation.
Many applications require the discovery of items which have occur frequently within multiple distributed data streams. Past solutions for this problem either require a high degree of error tolerance or can only provid...
详细信息
ISBN:
(数字)9783540734994
ISBN:
(纸本)9783540734987
Many applications require the discovery of items which have occur frequently within multiple distributed data streams. Past solutions for this problem either require a high degree of error tolerance or can only provide results periodically. In this paper we introduce a new algorithm designed for continuously tracking frequent items over distributed data streams providing either exact or approximate answers. We tested the efficiency of our method using two real-world data sets. the results indicated significant reduction in communication cost when compared to naive approaches and an existing efficient algorithm called Top-K Monitoring. Since our method does not rely upon approximations to reduce communication overhead and is explicitly designed for tracking frequent items, our method also shows increased quality in its tracking results.
this paper presents a data preprocessing procedure to select support vector (SV) candidates. We select decision boundary region vectors (BRVs) as SV candidates. Without the need to use the decision boundary, BRVs can ...
详细信息
ISBN:
(数字)9783540734994
ISBN:
(纸本)9783540734987
this paper presents a data preprocessing procedure to select support vector (SV) candidates. We select decision boundary region vectors (BRVs) as SV candidates. Without the need to use the decision boundary, BRVs can be selected based on a vector's nearest neighbor of opposite class (NNO). To speed up the process, two spatial approximation sample hierarchical (SASH) trees are used for estimating the BRVs. Empirical results show that our data selection procedure can reduce a full dataset to the number of SVs or only slightly higher. Training withthe selected subset gives performance comparable to that of the full dataset. For large datasets, overall time spent in selecting and training on the smaller dataset is significantly lower than the time used in training on the full dataset.
In order to arising the safety and reliabillity of numerical control system, aiming at the conventionally used encoders in the system as the equipment of feedback in position-loop, the datamining of its fault is anal...
详细信息
ISBN:
(纸本)9781424409723
In order to arising the safety and reliabillity of numerical control system, aiming at the conventionally used encoders in the system as the equipment of feedback in position-loop, the datamining of its fault is analyzed. Benefiting from the reluctant information of velocity in the normal servo system, the principle component analysis is introduced to solving the fault of losing codes and pausing codes in the encoder. At the same time, the basic principles and flow process are presented. In order to validate the effectiveness of the mining way based on the software of Matlab., simulation shows that it can perform a very exact diagnosis at the time of fault occurring. this idea provides a believable way for the fault datamining in numeric control system.
We propose a novel approach which extracts consistent (100% confident) rules and builds a classifier withthem. Recently, associative classifiers which utilize association rules have been widely studied. Indeed, the a...
详细信息
ISBN:
(数字)9783540734994
ISBN:
(纸本)9783540734987
We propose a novel approach which extracts consistent (100% confident) rules and builds a classifier withthem. Recently, associative classifiers which utilize association rules have been widely studied. Indeed, the associative classifiers often outperform the traditional classifiers. In this case, it is important to collect high quality (association) rules. Many algorithms find only high support rules, because decreasing the minimum support to be satisfied is computationally demanding. However, it may be effective to collect low support but high confidence rules. therefore, we propose an algorithm that produces a wide variety of 100% confident rules including low support rules. To achieve this goal, we adopt a specific-to-general rule searching strategy, in contrast to the previous many approaches. Our experimental results show that the proposed method achieves higher accuracies in several datasets taken from UCI machinelearning repository.
In this paper, a reinforcement learning method called DAQL is proposed to solve the problem of seeking and homing onto a fast maneuvering target, within the context of mobile robots. this Q-learning based method consi...
详细信息
ISBN:
(数字)9783540734994
ISBN:
(纸本)9783540734987
In this paper, a reinforcement learning method called DAQL is proposed to solve the problem of seeking and homing onto a fast maneuvering target, within the context of mobile robots. this Q-learning based method considers both target and obstacle actions when determining its own action decisions, which enables the agent to learn more effectively in a dynamically changing environment. It particularly suits fast-maneuvering target cases, in which maneuvers of the target are unknown a priori. Simulation result depicts that the proposed method is able to choose a less convoluted path to reach the target when compared to the ideal proportional navigation (IPN) method in handling fast maneuvering and randomly moving target. Furthermore, it can learn to adapt to the physical limitation of the system and do not require specific initial conditions to be satisfied for successful navigation towards the moving target.
Based on the definitions of extensible set and the constructing method of its dependent function, a sort of classification method under extension tranformation, which is called extension classification method, is stud...
详细信息
ISBN:
(纸本)9781424410651
Based on the definitions of extensible set and the constructing method of its dependent function, a sort of classification method under extension tranformation, which is called extension classification method, is studied. It is different from the classification methods based on classical set, fuzzy set and rough set, and it is a sort of alterable classification method According to a certain transformation, it can divide a universe of discourse into 5 ports: positive extension field, negative extension field, positive stable field, negative stable field and extension boundary. Moreover, the universe of discourse and the dependent function describing the degree that an object possesses certain character are alterable. It makes the classification more elaborate. the phenomenon that "there is a corresponding classification pattern for a given transformation" is illuminated from the angle of set theory. Taking the extension classification management. on human resources as an example, its applied value will be explained. the classification method is a basic Method of extension datamining. It makes the classification function of datamining richer.
this paper presents a novel solution for the problem of building text classifier using positive documents (P) and unlabeled documents (U). Here, the unlabeled documents are mixed with positive and negative documents. ...
详细信息
ISBN:
(数字)9783540734994
ISBN:
(纸本)9783540734987
this paper presents a novel solution for the problem of building text classifier using positive documents (P) and unlabeled documents (U). Here, the unlabeled documents are mixed with positive and negative documents. this problem is also called PU-learning. the key feature of PU-learning is that there is no negative document for training. Recently, several approaches have been proposed for solving this problem. Most of them are based on the same idea, which builds a classifier in two steps. Each existing technique uses a different method for each step. Generally speaking, these existing approaches do not perform well when the size of P is small. In this paper, we propose a new approach aiming at improving the system when the size of P is small. this approach combines the graph-based semi-supervised learning method withthe two-step method. Experiments indicate that our proposed method performs well especially when the size of P is small.
暂无评论