In this paper, the concept of long memory systems for forecasting is developed. the pattern Modelling and recognition System and Fuzzy Single Nearest Neighbour methods are introduced as local approximation tools for f...
详细信息
ISBN:
(纸本)0780364295
In this paper, the concept of long memory systems for forecasting is developed. the pattern Modelling and recognition System and Fuzzy Single Nearest Neighbour methods are introduced as local approximation tools for forecasting. Such systems are used for matching current state of the time-series with past states to make a forecast. In the past, the PMRS system has been successfully used for forecasting the Santa Fe competition data. In this paper, we forecast the FTSE 100 and 250 financial returns indices, as well as the stock returns of five FTSE 100 companies and compare the results of the two different systems, withthat of Exponential Smoothing and Random Walk on seven different error measures. the results show that patternrecognition based approaches in time-series forecasting are highly accurate. Simple theoretical trading strategies are also mentioned, highlighting real applications of the system.
In this work the construction of a neural network to perform the task of classification from a set of data for which the true classes are known is investigated. Depending upon the knowledge strored in the architecture...
ISBN:
(纸本)1853128104
In this work the construction of a neural network to perform the task of classification from a set of data for which the true classes are known is investigated. Depending upon the knowledge strored in the architecture of an Adaptive Logic Network (ALN) a specialized neurochip is built. the performance of this architecture is evaluated using a challenging medical data set and conclusions are drawn for the expandability of this neurochip for more general cases.
In this paper, we address the problem of high performance speaker-independent continuous Mandarin digital string recognizer and focus on exploiting context information and prosody knowledge. Data-driven decision tree ...
详细信息
ISBN:
(纸本)7801501144
In this paper, we address the problem of high performance speaker-independent continuous Mandarin digital string recognizer and focus on exploiting context information and prosody knowledge. Data-driven decision tree method to train tri-phone acoustic model was proposed. According to Chinese language property, digital specific question set was designed and the derived tri-phone model is more accurate to describe acoustic observation. For prosody cue, a novel Gaussian Mixture Density Duration Model (GMDDM) was presented. Unlike traditional normalizing or single parameter strategy, proposed duration model is context independent. the context variation is naturally embodied into multiple Gaussian distribution mixture. the number of mixture is automatically selected according maximum likelihood criteria. this simple but effective duration model's likelihood score is combined with acoustic score as heuristic information for the backward A∗ decoding of word graph. Experimental results show the tri-phone acoustic model could lead to average 12.9% reduce of string error rate. When GMDDM model is applied, the string error rate is further reduced by 22.7%, which demonstrates the very usefulness of GMDDM model.
A domain-independent, reusable, general text generation system for Chinese is presented in this paper. this system combines the template method and generation technology in a single formalism, which enables the system...
详细信息
ISBN:
(纸本)7801501144
A domain-independent, reusable, general text generation system for Chinese is presented in this paper. this system combines the template method and generation technology in a single formalism, which enables the system to maintain both flexibility and efficiency. At the same time, in order to maintain the generator's independence of application domains, An upper model is designed to interface the different application and the general generation system, which enables the system's adaptability to different application domains. the upper model is a kind of semantic hierarchy. It is organized according to the semantic relationship between predicates and their arguments, nouns and their modifiers. Tests show that the generation system embodies good adaptability to different application domains and good performances.
this paper investigates the potential for automatic mapping of typical embedded applications to architectures with multimedia instruction set extensions. For this purpose a (pattern matching based) code transformation...
详细信息
ISBN:
(纸本)3540679561
this paper investigates the potential for automatic mapping of typical embedded applications to architectures with multimedia instruction set extensions. For this purpose a (pattern matching based) code transformation engine is used, which involves a three-step process of matching, condition checking and replacing of the source code. Experiments with DSP and the MPEG2 encoder benchmarks, show that about 85% of the loops which are suitable for Single Instruction Multiple Data (SIMD) parallelization can be automatically recognized and mapped.
We address the issue of detecting automatically occurrences of high level patterns in audiovisual documents. these patterns correspond to recurring sequences of shots, which are considered as first class entities by d...
详细信息
A design pattern is a description of a high-quality solution to a frequently occurring problem in some domain. A pattern language is a collection of design patterns that are carefully organized to embody a design meth...
详细信息
To analyze the statistical characteristics of partial discharges (PDs) in power transformers, 7 types of experimental models simulating PD in transformers and 3 types of models simulating interfering PD in air are des...
详细信息
To analyze the statistical characteristics of partial discharges (PDs) in power transformers, 7 types of experimental models simulating PD in transformers and 3 types of models simulating interfering PD in air are designed and model experiments are conducted in a screened room. Using a digital measuring device withthe sampling rate of 50kHz, the quantity-phase information of PD pulse current in models is obtained after the signal is processed by a peak-holding hardware. the PD features are extracted using the 3D pattern chart and then a group of three-layer back-propagation ANNs (artificial neural networks) is used to recognize the PD patterns. the investigation shows that ANN has enough ability to recognize different PD in oil-paper insulation of power transformers and PD interference from air can be recognized and eliminated.
the proceedings contain 33 papers. the special focus in this conference is on Graph Languages, Graph theory;Categorical Approaches, Concurrency and Distribution. the topics include: Some remarks on the generative powe...
ISBN:
(纸本)3540672036
the proceedings contain 33 papers. the special focus in this conference is on Graph Languages, Graph theory;Categorical Approaches, Concurrency and Distribution. the topics include: Some remarks on the generative power of collage grammars and chain-code grammars;tree languages generated by context-free graph grammars;neighborhood-preserving node replacements;the power of local computations in graphs with initial knowledge;double-pushout approach with injective matching;pushout complements for arbitrary partial algebras;local views on distributed systems and their communication;dynamic change management by distributed graph transformation;redundancy and subsumption in high-level replacement systems;utilizing constraint satisfaction techniques for efficient graph pattern matching;hypergraphs as a uniform diagram representation model;a new graph rewrite language based on the unified modeling language and java;more about control conditions for transformation units;a framework for adding packages to graph transformation approaches;UML packages for programmed graph rewriting systems;incremental development of safety properties in Petri net transformations;graph-based models for managing development processes, resources, and products and deriving software performance models from architectural patterns by graph transformations.
the proceedings contain 121 papers. the special focus in this conference is on Foundations of AI, Induction, Logic Programming, Reinforcement Learning and Machine Learning. the topics include: Knowledge representation...
ISBN:
(纸本)3540679251
the proceedings contain 121 papers. the special focus in this conference is on Foundations of AI, Induction, Logic Programming, Reinforcement Learning and Machine Learning. the topics include: Knowledge representation, belief revision, and the challenge of optimality;towards a next-generation search engine;an attempt to represent a database by predicate formulae;a unifying semantics for causal ramifications;inductive inference of chess player strategy;compiling logical features into specialized state-evaluators by partial evaluation, Boolean tables and incremental calculation;using domain knowledge in ILP to discover protein functional models;knowledge reformation for efficient first-order hypothetical reasoning;determination of general concept in learning default rules;a theory of profit sharing in dynamic environment;a region selecting method which performs observation and action in the multi-resolution environment;the lumberjack algorithm for learning linked decision forests;efficient iris recognition system by optimization of feature vectors and classifier;an efficient learning algorithm using natural gradient and second order information of error surface;fast and robust general purpose clustering algorithms;tropical cyclone intensity forecasting model;trading off granularity against complexity in predictive models for complex domains;efficient inference in dynamic belief networks with variable temporal resolution;epistemic states guiding the rational dynamics of information;perceiving environments for intelligent agents;constructing an autonomous agent with an interdependent heuristics;from brain theory to autonomous robotic agents and a multi-agent approach for optical inspection technology.
暂无评论