In this paper, we developed softcomputing models for on-line automatic speech recognition (ASR) based on Bayesian on-line inference techniques. Bayesian on-line inference for change point detection (BOCPD) is tested ...
详细信息
ISBN:
(纸本)9783642196430
In this paper, we developed softcomputing models for on-line automatic speech recognition (ASR) based on Bayesian on-line inference techniques. Bayesian on-line inference for change point detection (BOCPD) is tested for on-line environmental learning using highly non-stationary noisy speech samples from the Aurora2 speech database. Significant improvement in predicting and adapting to new acoustic conditions is obtained for highly non-stationary noises. The simulation results show that the Bayesian on-line inference-based softcomputing approach would be one of the possible solutions to on-line ASR for real-time applications.
In this study a confidence measure for probability density functions (pdfs) is presented. The measure can be used in one-class classification to select a pdf threshold for class inclusion. In addition, confidence info...
详细信息
ISBN:
(纸本)0769525210
In this study a confidence measure for probability density functions (pdfs) is presented. The measure can be used in one-class classification to select a pdf threshold for class inclusion. In addition, confidence information can be used to verify correctness of a decision in a multi-class case where for example the Bayesian decision rule reveals which class is the most probable. Additionally, using confidence values - which represent in which quantile of the probability mass a pdf value resides ([0, 1]) - is often straightforward compared to using arbitrarily scaled pdf values. As the main contributions, use of confidence information in classification is described and a method for confidence estimation is presented.
In this paper, Multi-class classification using an Improved Multiobjective Simultaneous learning framework (MCIMSDC) is proposed. This learning algorithm is used to solve any multiclass classification problem. It is b...
详细信息
ISBN:
(纸本)9788132204909
In this paper, Multi-class classification using an Improved Multiobjective Simultaneous learning framework (MCIMSDC) is proposed. This learning algorithm is used to solve any multiclass classification problem. It is based on the framework proposed by Cai, Chen and Zhang [1] in 2010. In [1], the multiple objective functions are utilized to simultaneously optimize the clustering and classification learning by employing Bayesian theory. In [1], the selection of learning parameter i.e., clusters membership degree u(j)(x(i)) is initially chosen at random due to which the number of iteration and training time achieve while obtaining the stable cluster center is comparatively higher, but here in the proposed methodology, the value of clusters membership degree u(j)(x(i)) is calculated on the basis of randomly initialized cluster centers. Thus, these cluster centers are updated and corresponding u(j)(x(i)) is calculated iteratively. Experimental results show that, this method improve the performance by significantly reducing the number of iterations and training time required to obtain the cluster center. The same is being verified with six benchmark datasets.
Chemical taste is indispensable information in food testing. The technical of electronic tongue system is one of research directions to identify different chemical tastes. This paper focuses on the patternrecognition...
详细信息
ISBN:
(纸本)9781424437092
Chemical taste is indispensable information in food testing. The technical of electronic tongue system is one of research directions to identify different chemical tastes. This paper focuses on the patternrecognition method based on learning vector quantization (LVQ) neural network. The electronic tongue system designed could identify all the samples of beer, fruit juice and milk successfully in the experiments. The result shows that LVQ neural network is applicable in the patternrecognition of electronic tongue system and can also be used on condition that information is gathered by multisensors array. The patternrecognition methods of the universal electronic tongue are proposed in this paper. The effective universal electronic tongue has much advantage over others such as simple methods of patternrecognition and classification, easy training approaches and wider application fields.
Social network services (SNSs) serve numerous users with large amounts of information of different kinds. On an SNS, information will propagate on a user network, which is represented as a complex network in general b...
详细信息
ISBN:
(纸本)9781479959556
Social network services (SNSs) serve numerous users with large amounts of information of different kinds. On an SNS, information will propagate on a user network, which is represented as a complex network in general but can be reformed as a tree by using the direction of propagation and allowing duplication. Our goal in this study was to show the propagation of a particular kind of information on an SNS, as well as the clustering of a similar propagation scheme for each user. For this goal, we used elastic tree pattern matching to calculate the similarity of two tree structures. A set of users are propagated from source to destination in the same or similar way, and these users are given information from a similar source. We also aimed to find the high-influence person who is at the start of the same or similar propagation, which will indicate that she/he is the moderator of a topic. We used tumblr data for the experiment. Findings indicated that the similar part of each information propagation tree on tumblr was too small for the clustering propagation pattern.
The paper proposes a face detection system that locates and extracts faces from the background using the multilayer feedforward perceptron. Facial features are extracted from the local image using filters. In this app...
详细信息
ISBN:
(纸本)9788132226710;9788132226697
The paper proposes a face detection system that locates and extracts faces from the background using the multilayer feedforward perceptron. Facial features are extracted from the local image using filters. In this approach, feature vector from Gabor filter acts as an input for the multilayer feedforward perceptron. The points holding high information on face image are used for extraction of feature vectors. Since Gabor filter extracts features from varying scales and orientations, the feature points are extracted with high accuracy. Experimental results show the multilayer feedforward perceptron discriminates and detects faces from non-face patterns irrespective of the illumination changes.
Biometrics is the measurement of person's physiological or behavioral characteristics. It enables authentication of a person's identity using such measurements. Biometric-based authentication is thus becoming ...
详细信息
Biometrics is the measurement of person's physiological or behavioral characteristics. It enables authentication of a person's identity using such measurements. Biometric-based authentication is thus becoming increasingly important in computer-based applications because the amount of sensitive data stored in such systems is growing. Particularly challenging is the implementation of biometric-based authentication in embedded computer system applications, because the resources of such systems are scarce. Reliability and performance are two primary requirements to be satisfied in embedded system applications. Single-mode and hard-feature-based biometrics do not offer enough reliability and performance to satisfy such requirements. Multimode biometrics is a primary level of improvement. soft-biometric features can thus be considered along with hard-biometric features to further improve performance. A combination of soft-computing methods and soft-biometric data can yield more improvements in authentication performance by limiting requirements for memory and processing power. The multi-biometric approach also increases system reliability, since most embedded systems can capture more than one physiological or behavioral characteristic. A multi-biometric platform that combines voiceprint and fingerprint authentication was developed as a reference model to demonstrate the potential of soft-computing methods and soft-biometric data. Hard-computingpattern-matching algorithms were applied to match hard-biometric features. Artificial neural network (ANN) processing was applied to match soft-biometric features. Both hard-computing and soft-computing matching results are inferred by a fuzzy logic engine to perform smart authentication using a decision-fusion paradigm. The embedded implementation was based on a single-chip, floating-point, digital signal processor (DSP) to demonstrate the practical embeddability of such an approach and the improved performance that can be attaine
soft set theory in combination with the interval-valued fuzzy set has been proposed as the concept of the interval-valued fuzzy soft set. However, up to the present, few documents have focused on parameter reduction o...
详细信息
ISBN:
(纸本)9783642218804
soft set theory in combination with the interval-valued fuzzy set has been proposed as the concept of the interval-valued fuzzy soft set. However, up to the present, few documents have focused on parameter reduction of the interval-valued fuzzy soft sets. In this paper, we propose a definition of normal parameter reduction of interval-valued fuzzy soft sets, which considers the problems of sub-optimal choice and added parameters. Then, a heuristic algorithm of normal parameter reduction for interval-valued fuzzy soft sets is presented. Finally, an illustrative example is employed to show our contribution.
patterns are crucial for efficiently scheduling microservice workflow applications to containers in cloud computing scenarios. However, it is challenging to learn patterns of microservice workflows because of their co...
详细信息
patterns are crucial for efficiently scheduling microservice workflow applications to containers in cloud computing scenarios. However, it is challenging to learn patterns of microservice workflows because of their complex precedence constrained structures provided by users with more lightweighted, diversified, and personalized services. In this paper, we propose a graph neural network is designed to identify patterns within a set of microservice workflows by mining the common substructures of workflows. Based on the learned patterns, a pattern-based scheduling algorithm framework is developed for microservice workflows with soft deadline constraints to minimize the average tardiness. A sorting strategy is introduced based on urgency and pattern coverage rate. For simplification of the task sorting process, the pattern-based task sorting algorithm (PB-TS) is devised. Furthermore, a resource selection phase is incorporated to the pattern-based resource selection algorithm (PB-RS) to minimize the candidate resource space. Experimental results demonstrate the proposed method is much efficient as compared to three classical algorithms.
Kernel tracing facilitates to demonstrate various activities running inside the Operating System. Kernel tracing tools like LTT, LTTng, DTrace, FTrace provide details about processes and their resources but these tool...
详细信息
ISBN:
(纸本)9781509015603
Kernel tracing facilitates to demonstrate various activities running inside the Operating System. Kernel tracing tools like LTT, LTTng, DTrace, FTrace provide details about processes and their resources but these tools lack to extract knowledge from it. patternrecognition is a major field of data mining and knowledge discovery. This paper presents a survey of widely used algorithms like Apriori, Tree-projection, FP-growth, Eclat for finding frequent pattern over the database. This paper presents a comparative study of frequent pattern mining algorithm and suggests that the FP-growth algorithm is suitable for finding patterns in kernel trace data.
暂无评论