Abstract: In this paper, we propose an improved fast search
algorithm using combined histogram features and temporal division
method for short MPEG video clips from large video database. There
are two types of histogram features used to generate more robust
features. The first one is based on the adjacent pixel intensity
difference quantization (APIDQ) algorithm, which had been reliably
applied to human face recognition previously. An APIDQ histogram is
utilized as the feature vector of the frame image. Another one is
ordinal feature which is robust to color distortion. Combined with
active search [4], a temporal pruning algorithm, fast and robust video
search can be realized. The proposed search algorithm has been
evaluated by 6 hours of video to search for given 200 MPEG video
clips which each length is 30 seconds. Experimental results show the
proposed algorithm can detect the similar video clip in merely 120ms,
and Equal Error Rate (ERR) of 1% is achieved, which is more
accurately and robust than conventional fast video search algorithm.
Abstract: Web applications have become complex and crucial for many firms, especially when combined with areas such as CRM (Customer Relationship Management) and BPR (Business Process Reengineering). The scientific community has focused attention to Web application design, development, analysis, testing, by studying and proposing methodologies and tools. Static and dynamic techniques may be used to analyze existing Web applications. The use of traditional static source code analysis may be very difficult, for the presence of dynamically generated code, and for the multi-language nature of the Web. Dynamic analysis may be useful, but it has an intrinsic limitation, the low number of program executions used to extract information. Our reverse engineering analysis, used into our WAAT (Web Applications Analysis and Testing) project, applies mutational techniques in order to exploit server side execution engines to accomplish part of the dynamic analysis. This paper studies the effects of mutation source code analysis applied to Web software to build application models. Mutation-based generated models may contain more information then necessary, so we need a pruning mechanism.
Abstract: This paper presents an effective framework for Chinesesyntactic parsing, which includes two parts. The first one is a parsing framework, which is based on an improved bottom-up chart parsingalgorithm, and integrates the idea of the beam search strategy of N bestalgorithm and heuristic function of A* algorithm for pruning, then get multiple parsing trees. The second is a novel evaluation model, which integrates contextual and partial lexical information into traditional PCFG model and defines a new score function. Using this model, the tree with the highest score is found out as the best parsing tree. Finally,the contrasting experiment results are given. Keywords?syntactic parsing, PCFG, pruning, evaluation model.
Abstract: The belief decision tree (BDT) approach is a decision
tree in an uncertain environment where the uncertainty is represented
through the Transferable Belief Model (TBM), one interpretation
of the belief function theory. The uncertainty can appear either in
the actual class of training objects or attribute values of objects to
classify. In this paper, we develop a post-pruning method of belief
decision trees in order to reduce size and improve classification
accuracy on unseen cases. The pruning of decision tree has a
considerable intention in the areas of machine learning.
Abstract: Feature selection study is gaining importance due to its contribution to save classification cost in terms of time and computation load. In search of essential features, one of the methods to search the features is via the decision tree. Decision tree act as an intermediate feature space inducer in order to choose essential features. In decision tree-based feature selection, some studies used decision tree as a feature ranker with a direct threshold measure, while others remain the decision tree but utilized pruning condition that act as a threshold mechanism to choose features. This paper proposed threshold measure using Manhattan Hierarchical Cluster distance to be utilized in feature ranking in order to choose relevant features as part of the feature selection process. The result is promising, and this method can be improved in the future by including test cases of a higher number of attributes.
Abstract: Knowledge is indispensable but voluminous knowledge becomes a bottleneck for efficient processing. A great challenge for data mining activity is the generation of large number of potential rules as a result of mining process. In fact sometimes result size is comparable to the original data. Traditional data mining pruning activities such as support do not sufficiently reduce the huge rule space. Moreover, many practical applications are characterized by continual change of data and knowledge, thereby making knowledge voluminous with each change. The most predominant representation of the discovered knowledge is the standard Production Rules (PRs) in the form If P Then D. Michalski & Winston proposed Censored Production Rules (CPRs), as an extension of production rules, that exhibit variable precision and supports an efficient mechanism for handling exceptions. A CPR is an augmented production rule of the form: If P Then D Unless C, where C (Censor) is an exception to the rule. Such rules are employed in situations in which the conditional statement 'If P Then D' holds frequently and the assertion C holds rarely. By using a rule of this type we are free to ignore the exception conditions, when the resources needed to establish its presence, are tight or there is simply no information available as to whether it holds or not. Thus the 'If P Then D' part of the CPR expresses important information while the Unless C part acts only as a switch changes the polarity of D to ~D. In this paper a scheme based on Dempster-Shafer Theory (DST) interpretation of a CPR is suggested for discovering CPRs from the discovered flat PRs. The discovery of CPRs from flat rules would result in considerable reduction of the already discovered rules. The proposed scheme incrementally incorporates new knowledge and also reduces the size of knowledge base considerably with each episode. Examples are given to demonstrate the behaviour of the proposed scheme. The suggested cumulative learning scheme would be useful in mining data streams.
Abstract: Clustering large populations is an important problem
when the data contain noise and different shapes. A good clustering
algorithm or approach should be efficient enough to detect clusters
sensitively. Besides space complexity, time complexity also gains
importance as the size grows. Using hierarchies we developed a new
algorithm to split attributes according to the values they have and
choosing the dimension for splitting so as to divide the database
roughly into equal parts as much as possible. At each node we
calculate some certain descriptive statistical features of the data
which reside and by pruning we generate the natural clusters with a
complexity of O(n).
Abstract: This article outlines conceptualization and
implementation of an intelligent system capable of extracting
knowledge from databases. Use of hybridized features of both the
Rough and Fuzzy Set theory render the developed system flexibility
in dealing with discreet as well as continuous datasets. A raw data set
provided to the system, is initially transformed in a computer legible
format followed by pruning of the data set. The refined data set is
then processed through various Rough Set operators which enable
discovery of parameter relationships and interdependencies. The
discovered knowledge is automatically transformed into a rule base
expressed in Fuzzy terms. Two exemplary cancer repository datasets
(for Breast and Lung Cancer) have been used to test and implement
the proposed framework.
Abstract: A novel path planning approach is presented to solve
optimal path in stochastic, time-varying networks under priori traffic
information. Most existing studies make use of dynamic programming
to find optimal path. However, those methods are proved to
be unable to obtain global optimal value, moreover, how to design
efficient algorithms is also another challenge.
This paper employs a decision theoretic framework for defining
optimal path: for a given source S and destination D in urban transit
network, we seek an S - D path of lowest expected travel time
where its link travel times are discrete random variables. To solve
deficiency caused by the methods of dynamic programming, such as
curse of dimensionality and violation of optimal principle, an integer
programming model is built to realize assignment of discrete travel
time variables to arcs. Simultaneously, pruning techniques are also
applied to reduce computation complexity in the algorithm. The final
experiments show the feasibility of the novel approach.