In this paper we explore the power of surface text patterns for open-domain question answering systems. In order to obtain an optimal set of patterns, we have developed a method for learning such patterns automatically. A tagged corpus is built from the Internet in a bootstrapping process by providing a few hand-crafted examples of each question type to Altavista. Patterns are then automatically extracted from the returned documents and standardized. We calculate the precision of each pattern, and the average precision for each question type. These patterns are then applied to find answers to new questions.

A new system for aligning whole genome sequences is described. Using an efficient data structure called a suffix tree, the system is able to rapidly align sequences containing millions of nucleotides. Its use is demonstrated on two strains of Mycobacterium tuberculosis, on two less similar species of Mycoplasma bacteria and on two syntenic sequences from human chromosome 12 and mouse chromosome 6. In each case it found an alignment of the input sequences, using between 30 s and 2 min of computation time. From the system output, information on single nucleotide changes, translocations and homologous genes can easily be extracted. Use of the algorithm should facilitate analysis of syntenic chromosomal regions, strain-to-strain comparisons, evolutionary comparisons and genomic duplications. INTRODUCTION Since the first successful whole-genome shotgun sequence of Haemophilus influenzae (1), the number of organisms whose genomes have been completely sequenced has been increasing rapidly e...

In this paper we explore object recognition in clutter. We test our object recognition techniques on Gimpy and EZGimpy, examples of visual CAPTCHAs. A CAPTCHA ("Completely Automated Public Turing test to Tell Computers and Humans Apart") is a program that can generate and grade tests that most humans can pass, yet current computer programs can't pass. EZ-Gimpy (see Fig. 1, 5), currently used by Yahoo, and Gimpy (Fig. 2,9) are CAPTCHAs based on word recognition in the presence of clutter. These CAPTCHAs provide excellent test sets since the clutter they contain is adversarial; it is designed to confuse computer programs. We have developed efficient methods based on shape context matching that can identify the word in an EZGimpy image with a success rate of 92%, and the requisite 3 words in a Gimpy image 33% of the time. The problem of identifying words in such severe clutter provides valuable insight into the more general problem of object recognition in scenes. The methods that we present are instances of a framework designed to tackle this general problem.

In order to increase retrieval precision, some new search engines provide manually verified answers to Frequently Asked Queries (FAQs). An underlying task is the identification of FAQs. This paper describes our attempt to cluster similar queries according to their contents as well as user logs. Our preliminary results show that the resulting clusters provide useful information for FAQ identification.

Motivation Classification of proteins sequences into functional and structural families based on sequence homology is a central problem in computational biology. Discriminative supervised machine learning approaches provide good performance, but simplicity and computational efficiency of training and prediction are also important concerns. Results We introduce a class of string kernels, called mismatch kernels, for use with support vector machines (SVMs) in a discriminative approach to the problem of protein classification and remote homology detection. These kernels measure sequence similarity based on shared occurrences of fixed-length patterns in the data, allowing for mutations between patterns. Thus the kernels provide a biologically well-motivated way to compare protein sequences without relying on family-based generative models such as hidden Markov models. We compute the kernels efficiently using a mismatch tree data structure, allowing us to calculate the contributions of all patterns occurring in the data in one pass while traversing the tree. When used with an SVM, the kernels enable fast prediction on test sequences. We report experiments on two benchmark SCOP data sets, where we show that the mismatch kernel used with an SVM classifier performs competitively with state-of-the-art methods for homology detection, particularly when very few training examples are available. Examination of the highestweighted patterns learned by the SVM classifier recovers biologically important motifs in protein families and superfamilies. Availability SVM software is publically available at

Conventional document retrieval systems (e.g., Alta Vista) return long lists of ranked documents in response to user queries. Recently, document clustering has been put forth as an alternative method of organizing retrieval results (Cutting et al. 1992). A person browsing the clusters can discover patterns that could be overlooked in the traditional presentation. This paper describes two novel clustering methods that intersect the documents in a cluster to determine the set of words (or phrases) shared by all the documents in the cluster. We report on experiments that evaluate these intersectionbased clustering methods on collections of snippets returned from Web search engines. First, we show that word-intersection clustering produces superior clusters and does so faster than standard techniques. Second, we show that our O(n log n) time phrase-intersection clustering method produces comparable clusters and does so more than two orders of magnitude faster than all methods tested. I...

gusfield,eddhu¡ A phylogenetic network is a generalization of a phylogenetic tree, allowing structural properties that are not treelike. With the growth of genomic data, much of which does not fit ideal tree models, there is greater need to understand the algorithmics and combinatorics of phylogenetic networks [10, 11]. However, to date, very little has been published on this, with the notable exception of the paper by Wang et al.[12]. Other related papers include [4, 5, 7] We consider the problem introduced in [12], of determining whether the sequences can be derived on a phylogenetic network where the recombination cycles are node disjoint. In this paper, we call such a phylogenetic network a “galled-tree”. By more deeply analysing the combinatorial constraints on cycle-disjoint phylogenetic networks, we obtain an efficient algorithm that is guaranteed to be both a necessary and sufficient test for the existence of a galledtree for the data. If there is a galled-tree, the algorithm constructs one and obtains an implicit representation of all the galled trees for the data, and can create these in linear time for each one. We also note two additional results related to galled trees: first, any set of sequences that can be derived on a galled tree can be derived on a true tree (without recombination cycles), where at most one back mutation is allowed per site; second, the site compatibility problem (which is NP-hard in general) can be solved in linear time for any set of sequences that can be derived on a galled tree. The combinatorial constraints we develop apply (for the most part) to node-disjoint cycles in any phylogenetic network (not just galled-trees), and can be used for example to prove that a given site cannot be on a node-disjoint cycle in any phylogenetic network. Perhaps more important than the specific results about galled-trees, we introduce an approach that can be used to study recombination in phylogenetic networks that go beyond galled-trees.

Multiple alignment of macromolecular sequences generalizes from N = 2 to N # 3 the comparison of N sequences which have diverged through the local processes of insertion, deletion and substitution. Gene-order sequences diverge through non-local genome rearrangement processes such as inversion (or reversal) and transposition. In this paper we show which formulations of multiple alignment have counterparts in multiple rearrangement. Based on di#culties inherent in rearrangement edit-distance calculation and interpretation, we argue for the simpler "breakpoint analysis ". Consensus-based multiple rearrangement of N # 3 orders can be solved exactly through reduction to instances of the Travelling Salesman Problem (TSP). We propose a branch-and-bound solution to TSP particularly suited to these instances. Simulations show how non-uniqueness of the solution is attenuated with increasing numbers of data genomes. Treebased multiple alignment can be achieved to a great degree o...

Anomaly detection is an important problem that has been researched within diverse research areas and application domains. Many anomaly detection techniques have been specifically developed for certain application domains, while others are more generic. This survey tries to provide a structured and comprehensive overview of the research on anomaly detection. We have grouped existing techniques into different categories based on the underlying approach adopted by each technique. For each category we have identified key assumptions, which are used by the techniques to differentiate between normal and anomalous behavior. When applying a given technique to a particular domain, these assumptions can be used as guidelines to assess the effectiveness of the technique in that domain. For each category, we provide a basic anomaly detection technique, and then show how the different existing techniques in that category are variants of the basic technique. This template provides an easier and succinct understanding of the techniques belonging to each category. Further, for each category, we identify the advantages and disadvantages of the techniques in that category. We also provide a discussion on the computational complexity of the techniques since it is an important issue in real application domains. We hope that this survey will provide a better understanding of the di®erent directions in which research has been done on this topic, and how techniques developed in one area can be applied in domains for which they were not intended to begin with.

We present a lossless compression algorithm, GenCompress, for genetic sequences, based on searching for approximate repeats. Our algorithm achieves the best compression ratios for benchmark DNA sequences. Significantly better compression results show that the approximate repeats are one of the main hidden regularities in DNA sequences. We then describe a theory of measuring the relatedness between two DNA sequences. Using our algorithm, we present strong experimental support for this theory, and demonstrate its application in comparing genomes and constructing evolutionary trees. 1