The retrospective fault analysis of complex technical devices based on documents emerging in the advanced steps of the product life cycle can reveal error sources and problems, which have not been discovered by simula...
详细信息
ISBN:
(纸本)9789728865726
The retrospective fault analysis of complex technical devices based on documents emerging in the advanced steps of the product life cycle can reveal error sources and problems, which have not been discovered by simulations or other test methods in the early stages of the product life cycle. This paper presents a novel approach to support the failure analysis through (i) a semi-automatic analysis of databases containing product-related documents in natural language (e. g., problem and error descriptions, repair and maintenance protocols, service bills) using information retrieval and text mining techniques and (ii) an interactive exploration of the data mining results. Our system supports visual data mining by mapping the results of analyzing failure-related documents onto corresponding 3D models. Thus, visualization of statistics about failure sources can reveal problem sources resulting from problematic spatial configurations.
Speaker variability is an important source of speech variations which makes continuous speech recognition a difficult *** automatic speech recognition(ASR) models to the speaker variations is a well-known strategy to ...
详细信息
Speaker variability is an important source of speech variations which makes continuous speech recognition a difficult *** automatic speech recognition(ASR) models to the speaker variations is a well-known strategy to cope with the *** all such techniques focus on developing adaptation solutions within the acoustic models of the ASR *** variations of the acoustic features constitute an important portion of the inter-speaker variations,they do not cover variations at the phonetic *** variations are known to form an important part of variations which are influenced by both micro-segmental and suprasegmental ***-speaker phonetic variations are influenced by the structure and anatomy of a speaker's articulatory system and also his/her speaking style which is driven by many speaker background characteristics such as accent,gender,age,socioeconomic and educational *** effect of inter-speaker variations in the feature space may cause explicit phone recognition *** errors can be compensated later by having appropriate pronunciation variants for the lexicon entries which consider likely phone misclassifications besides *** this paper,we introduce speaker adaptive dynamic pronunciation models,which generate different lexicons for various speaker clusters and different ranges of speech *** models are hybrids of speaker adapted contextual rules and dynamic generalized decision trees,which take into account word phonological structures,rate of speech,unigram probabilities and stress to generate pronunciation variants of *** the set of speaker adapted dynamic lexicons in a Farsi(Persian) continuous speech recognition task results in word error rate reductions of as much as 10.1% in a speaker-dependent scenario and 7.4% in a speaker-independent scenario.
With Semantic Web technologies and Linked Data datasets we are able to not only retrieve the textual content of a document but also to automatically create formal semantic descriptions of its content. In this paper we...
详细信息
The K2 metric is a well-known evaluation measure (or scoring function) for learning Bayesian networks from data [7]. It is derived by assuming uniform prior distributions on the values of an attribute for each possibl...
详细信息
We present a new approach to enriching under-specified representations of content to be realized as text. Our approach uses an attribute grammar to propagate missing information where needed in a tree that represents ...
详细信息
In this paper we consider the problem of inducing causal relations from statistical data. Although it is well known that a correlation does not justify the claim of a causal relation between two measures, the question...
详细信息
The field of natural language processing (NLP) has dramatically expanded within the last decade. Many human-being applications are conducted daily via NLP tasks, starting from machine translation, speech recognition, ...
详细信息
The FP-growth algorithm is currently one of the fastest approaches to frequent item set mining. In this paper I describe a C implementation of this algorithm, which contains two variants of the core operation of compu...
详细信息
ISBN:
(纸本)1595932100
The FP-growth algorithm is currently one of the fastest approaches to frequent item set mining. In this paper I describe a C implementation of this algorithm, which contains two variants of the core operation of computing a projection of an FP-tree (the fundamental data structure of the FP-growth algorithm). In addition, projected FP-trees are (optionally) pruned by removing items that have become infrequent due to the projection (an approach that has been called FP-Bonsai). I report experimental results comparing this implementation of the FP-growth algorithm with three other frequent item set mining algorithms I implemented (Apriori, Eclat, and Relim). Copyright 2005 ACM.
暂无评论