Distributed flow algorithms for scalable similarity visualization

We describe simple yet scalable and distributed algorithms for solving the maximum flow problem and its minimum cost flow variant, motivated by problems of interest in objects similarity visualization. We formulate the fundamental problem as a convex-concave saddle point problem. We then show that this problem can be efficiently solved by a first order method or by exploiting faster quasi-Newton steps. Our proposed approach costs at most O(|ε|) per iteration for a graph with |ε| edges. Further, the number of required iterations can be shown to be independent of number of edges for the first order approximation method. We present experimental results in two applications: mosaic generation and color similarity based image layouting. © 2010 IEEE.

Encoding of mixtures in a simple olfactory system.

Natural odors are usually mixtures; yet, humans and animals can experience them as unitary percepts. Olfaction also enables stimulus categorization and generalization. We studied how these computations are performed with the responses of 168 locust antennal lobe projection neurons (PNs) to varying mixtures of two monomolecular odors, and of 174 PNs and 209 mushroom body Kenyon cells (KCs) to mixtures of up to eight monomolecular odors. Single-PN responses showed strong hypoadditivity and population trajectories clustered by odor concentration and mixture similarity. KC responses were much sparser on average than those of PNs and often signaled the presence of single components in mixtures. Linear classifiers could read out the responses of both populations in single time bins to perform odor identification, categorization, and generalization. Our results suggest that odor representations in the mushroom body may result from competing optimization constraints to facilitate memorization (sparseness) while enabling identification, classification, and generalization.

Encoding of mixtures in a simple olfactory system

Natural odors are usually mixtures; yet, humans and animals can experience them as unitary percepts. Olfaction also enables stimulus categorization and generalization. We studied how these computations are performed with the responses of 168 locust antennal lobe projection neurons (PNs) to varying mixtures of two monomolecular odors, and of 174 PNs and 209 mushroom body Kenyon cells (KCs) to mixtures of up to eight monomolecular odors. Single-PN responses showed strong hypoadditivity and population trajectories clustered by odor concentration and mixture similarity. KC responses were much sparser on average than those of PNs and often signaled the presence of single components in mixtures. Linear classifiers could read out the responses of both populations in single time bins to perform odor identification, categorization, and generalization. Our results suggest that odor representations in the mushroom body may result from competing optimization constraints to facilitate memorization (sparseness) while enabling identification, classification, and generalization

Graphical representation and the similarity of DNA primary sequences

In this article, graphical representations of DNA primary sequences were generated. Topological indices and molecular connectivity indices were calculated and used for the comparison of similarities among eight different DNA segments. The satisfactory results were achieved by this analysis.

Local similarity in organic crystals and the non-uniqueness of X-ray powder patterns

Two new concepts for molecular solids, 'local similarity' and 'boundary-preserving isometry', are defined mathematically and a theorem which relates these concepts is formulated. 'Locally similar' solids possess an identical short-range structure and a 'boundary-preserving isometry' is a new mathematical operation on a finite region of a solid that transforms mathematically a given solid to a locally similar one. It is shown further that the existence of such a 'boundary-preserving isometry' in a given solid has infinitely many 'locally similar' solids as a consequence. Chemical implications, referring to the similarity of X-ray powder patterns and patent registration, are discussed as well. These theoretical concepts, which are first introduced in a schematic manner, are proved to exist in nature by the elucidation of the crystal structure of some diketopyrrolopyrrole (DPP) derivatives with surprisingly similar powder patterns. Although the available powder patterns were not indexable, the underlying crystals could be elucidated by using the new technique of ab initio prediction of possible polymorphs and a subsequent Rietveld refinement. Further ab initio packing calculations on other molecules reveal that 'local crystal similarity' is not restricted to DPP derivatives and should also be exhibited by other molecules such as quinacridones. The 'boundary-preserving isometry' is presented as a predictive tool for crystal engineering purposes and attempts to detect it in crystals of the Cambridge Structural Database (CSD) are reported.

Atom space distance and computing of molecular similarity

A mathematical model on computation of molecular similarity was suggested, The algorithmic techniques for measuring the degree of similarity between pairs of three-dimensional chemical molecules was represented by modified interatomic distance matrices. Current work was carried out on Indigo 2 work station with Sybyl software. Four groups of molecules were used to compute the molecules similarity to testing the mathematical model with satisfactory results.

Chemical environment code and measure of molecular similarity

A new scheme for the code of chemical environments of compounds is described in this paper, and three molecular similarity methods have been used to select nearest neighbors from four different types of probe compounds. One of the methods is based on the C-13 NMR spectra. The second method is based on the code of chemical environments and molecular topological index A(x). The third approach, i.e. the Tanimoto coefficient, is also based on the code of chemical environments, but not to use the topological index. Five nearest neighbors for each probe compound using these three molecular similarity methods were determined and taken from the database of 7309 structures. The results indicate that the scheme of the chemical environment code and the method for similarity measure of intermolecules suggested in this study are reasonable. (C) 1998 Published by Elsevier Science Ltd. All rights reserved.

Code for chemical environment and computation of molecular similarity

A new code for chemical environment and an empirical mathematical pattern Sa(m) on computation of molecular similarity were suggested. Seven molecules which referred to as the probe compounds and the nearest neighbors of each probe structure were determined by the methods of Sa(m) and Tanimoto, The results show an intuitive notion of chemical similarity.

Homology Induction: the use of machine learning to improve sequence similarity searches

Karwath, A. King, R. Homology induction: the use of machine learning to improve sequence similarity searches. BMC Bioinformatics. 23rd April 2002. 3:11 Additional File Describes the title organims species declaration in one string [http://www.biomedcentral.com/content/supplementary/1471- 2105-3-11-S1.doc] Sponsorship: Andreas Karwath and Ross D. King were supported by the EPSRC grant GR/L62849.

Semi-Supervised OWA Aggregation for Link-Based Similarity Evaluation and Alias Detection

T.Boongoen and Q. Shen. Semi-Supervised OWA Aggregation for Link-Based Similarity Evaluation and Alias Detection. Proceedings of the 18th International Conference on Fuzzy Systems (FUZZ-IEEE'09), pp. 288-293, 2009. Sponsorship: EPSRC

BoostMap: A Method for Efficient Approximate Similarity Rankings

This paper introduces BoostMap, a method that can significantly reduce retrieval time in image and video database systems that employ computationally expensive distance measures, metric or non-metric. Database and query objects are embedded into a Euclidean space, in which similarities can be rapidly measured using a weighted Manhattan distance. Embedding construction is formulated as a machine learning task, where AdaBoost is used to combine many simple, 1D embeddings into a multidimensional embedding that preserves a significant amount of the proximity structure in the original space. Performance is evaluated in a hand pose estimation system, and a dynamic gesture recognition system, where the proposed method is used to retrieve approximate nearest neighbors under expensive image and video similarity measures. In both systems, BoostMap significantly increases efficiency, with minimal losses in accuracy. Moreover, the experiments indicate that BoostMap compares favorably with existing embedding methods that have been employed in computer vision and database applications, i.e., FastMap and Bourgain embeddings.

Explaining World Wide Web Traffic Self-Similarity

Recently the notion of self-similarity has been shown to apply to wide-area and local-area network traffic. In this paper we examine the mechanisms that give rise to self-similar network traffic. We present an explanation for traffic self-similarity by using a particular subset of wide area traffic: traffic due to the World Wide Web (WWW). Using an extensive set of traces of actual user executions of NCSA Mosaic, reflecting over half a million requests for WWW documents, we show evidence that WWW traffic is self-similar. Then we show that the self-similarity in such traffic can be explained based on the underlying distributions of WWW document sizes, the effects of caching and user preference in file transfer, the effect of user "think time", and the superimposition of many such transfers in a local area network. To do this we rely on empirically measured distributions both from our traces and from data independently collected at over thirty WWW sites.