Narrow Results

In this paper, we study the problem of multicast routing on directed graphs. We define the asymmetry of a graph to be the maximum ratio of weights on opposite directed edges between a pair of nodes for all node-pairs. We examine three types of problems according the membership behavior: (i) the static, (ii) the join-only, (iii) the join-leave problems. We study the effect of the asymmetry on the worst case performance of two algorithms: the Greedy and Shortest Paths algorithms. The worst case performance of Shortest Paths is poor, but it is affected by neither the asymmetry nor the membership behavior. In contrast, the worst case performance of Greedy is a proportional to the asymmetry in a some cases. We prove an interesting result for the join-only problem: the Greedy algorithm has near-optimal on-line performance.

Drinking water systems are vulnerable to attack. While homeland security initiatives commonly focus on aerial CW/BW attacks, they have tended to ignore the far more inexpensive and easy to orchestrate attacks on drinking water distribution networks. An approach that utilizes off-the-shelf broad-spectrum analytical instruments coupled with advanced interpretive algorithms to provide detection-response networks for water is described. This report summarizes the development of interpretive algorithms applied to drinking water instrumentation and the implications for immediately engineering and deploying distribution based detection-protection systems. Data obtained from in house testing along with testing at ECBC was used to produce fingerprint response data from traditional as well as non-military threat agents. Data obtained from a Battelle/EPA ETV study addresses issues such as long-term deployment and ability to detect and characterize contaminants. Loop testing at ECBC and Battelle demonstrates extrapolation of beaker testing fingerprints to flowing systems. Real world deployment data is used to demonstrate recognition and classification of actual events and heuristic capabilities of the system along with its potential role in enhancing water quality above and beyond its obvious security aspects. The system is shown to be a practical measure to detect and characterize backflow events involving both chemicals and bioagents.

This report presents the status of the use of World-Wide Web (WWW) in High Energy Physics (HEP) experiments. The use of WWW in general, for ‘Online Datataking Systems’ and for ‘Offline Analysis Systems’ is discussed. In each of these cases the current use and a possible outlook for the future is described. Statistics on the actual use of WWW in HEP experiments as well as its current problems and future needs are also presented.

This paper presents a real-time constraint-free handprinted character recognition system based on a structural approach. After the preprocessing operation, a chain code is extracted to represent the character. The classification is based on the use of a processor dedicated to string comparison. The average computation time to recognize a character is about 0.07 seconds. During the learning step, the user can define any set of characters or symbols to be recognized by the system. Thus there are no constraints on the handprinting. The experimental tests show a high degree of accuracy (96%) for writer-dependent applications. Comparisons with other system and methods are discussed. We also present a comparison between the processor used in this system and the Wagner and Fischer algorithm. Finally, we describe some applications of the system.

Blended teaching is a kind of teaching that combines online teaching with traditional teaching, which is defined as “online and offline”. Through the organic combination of these two teaching forms, students’ learning can be from shallow to deep. Therefore, based on the data mining algorithm, this paper designs the method of College English online and offline blended teaching effect. First, it collects the College English blended teaching resources, then builds the College English online and offline teaching support, debugs the College English teaching environment, and finally designs the College English blended teaching model based on the data mining algorithm, so as to realize the College English online and offline blended teaching, The experiment shows that the method designed in this paper can effectively improve the reading ability of College English, and has certain application value.

In this chapter, we analyze several on-line cursive handwriting recognition systems. We find that virtually all such systems involve (a) a preprocessor, (b) a trainable classifier, and (c) a language modeling post-processor. Such architectures are described within the framework of Weighted Finite State Transductions, previously used in speech recognition by Pereira et al. We describe in some detail a recognition system built in our laboratory. It is a writer independent system which can handle a variety of writing styles including cursive script and handprint. The input to the system encodes the pen trajectory as a time-ordered sequence of feature vectors. A Time Delay Neural Network is used to estimate a posteriori probabilities for characters in a word. A Hidden Markov Model segments the word in a way which optimizes the global word score, taking a lexicon into account. The last part of the chapter is devoted to bibliographical notes.

Drinking water systems are vulnerable to attack. While homeland security initiatives commonly focus on aerial CW/BW attacks, they have tended to ignore the far more inexpensive and easy to orchestrate attacks on drinking water distribution networks. An approach that utilizes off-the-shelf broad-spectrum analytical instruments coupled with advanced interpretive algorithms to provide detection-response networks for water is described. This report summarizes the development of interpretive algorithms applied to drinking water instrumentation and the implications for immediately engineering and deploying distribution based detection-protection systems. Data obtained from in house testing along with testing at ECBC was used to produce fingerprint response data from traditional as well as non-military threat agents. Data obtained from a Battelle/EPA ETV study addresses issues such as long-term deployment and ability to detect and characterize contaminants. Loop testing at ECBC and Battelle demonstrates extrapolation of beaker testing fingerprints to flowing systems. Real world deployment data is used to demonstrate recognition and classification of actual events and heuristic capabilities of the system along with its potential role in enhancing water quality above and beyond its obvious security aspects. The system is shown to be a practical measure to detect and characterize backflow events involving both chemicals and bioagents.

This paper presents a real-time constraint-free handprinted character recognition system based on a structural approach. After the preprocessing operation, a chain code is extracted to represent the character. The classification is based on the use of a processor dedicated to string comparison. The average computation time to recognize a character is about 0.07 seconds. During the learning step, the user can define any set of characters or symbols to be recognized by the system. Thus there are no constraints on the handprinting. The experimental tests show a high degree of accuracy (96%) for writer-dependent applications. Comparisons with other systems and methods are discussed. We also present a comparison between the processor used in this system and the Wagner and Fischer algorithm. Finally, we describe some applications of the system.