The problem of searching the elements of a set which are close to a given query element under some similarity criterion has a vast number of applications in many branches of computer science, from pattern recognition to textual and multimedia information retrieval. We are interested in the rather general case where the similarity criterion defines a metric space, instead of the more restricted case of a vector space. A large number of solutions have been proposed in different areas, in many cases without cross-knowledge. Because of this, the same ideas have been reinvented several times, and very different presentations have been given for the same approaches. We

We present CiteSeer: an autonomous citation indexing system which indexes academic literature in electronic format (e.g. Postscript files on the Web). CiteSeer understands how to parse citations, identify citations to the same paper in different formats, and identify the context of citations in the body of articles. CiteSeer provides most of the advantages of traditional (manually constructed) citation indexes (e.g. the ISI citation indexes), including: literature retrieval by following citation links (e.g. by providing a list of papers that cite a given paper), the evaluation and ranking of papers, authors, journals, etc. based on the number of citations, and the identification of research trends. CiteSeer has many advantages over traditional citation indexes, including the ability to create more up-to-date databases which are not limited to a preselected set of journals or restricted by journal publication delays, completely autonomous operation with a corresponding reduction in cost, and powerful interactive browsing of the literature using the context of citations. Given a particular paper of interest, CiteSeer can display the context of how the paper is cited in subsequent publications. This context may contain a brief summary of the paper, another author's response to the paper, or subsequent work which builds upon the original article. CiteSeer allows the location of papers by keyword search or by citation links. Papers related to a given paper can be located using common citation information or word vector similarity. CiteSeer will soon be available for public use.

Given user data, one often wants to find approximate matches in a large database. A good example of such a task is finding images similar to a given image in a large collection of images. We focus on the important and technically difficult case where each data element is high dimensional, or more generally, is represented by a point in a large metric spaceand distance calculations are computationally expensive. In this paper we introduce a data structure to solve this problem called a GNAT- Geometric Near-neighbor Access Tree. It is based on the philosophy that the data structure should act as a hierarchical geometrical model of the data as opposed to a simple decomposition of the data that does not use its intrinsic geometry. In experiments, we find that GNAT’s outperform previous data structures in a number of applications.

This paper presents the theory, design principles, implementation, and performance results of PicHunter, a prototype content-based image retrieval (CBIR) system that has been developed over the past three years. In addition, this document presents the rationale, design, and results of psychophysical experiments that were conducted to address some key issues that arose during PicHunter’s development. The PicHunter project makes four primary contributions to research on content-based image retrieval. First, PicHunter represents a simple instance of a general Bayesian framework we describe for using relevance feedback to direct a search. With an explicit model of what users would do, given what target image they want, PicHunter uses Bayes’s rule to predict what is the target they want, given their actions. This is done via a probability distribution over possible image targets, rather than by refining a query. Second, an entropy-minimizing display algorithm is described that attempts to maximize the information obtained from a user at each iteration of the search. Third, PicHunter makes use of hidden annotation rather than a possibly inaccurate/inconsistent annotation structure that the user must learn and make queries in. Finally, PicHunter introduces two experimental paradigms to quantitatively evaluate the performance of the system, and psychophysical experiments are presented that support the theoretical claims.

Most research on nearest neighbor algorithms in the literature has been focused on the Euclidean case. In many practical search problems however, the underlying metric is non-Euclidean. Nearest neighbor algorithms for general metric spaces are quite weak, which motivates a search for other classes of metric spaces that can be tractably searched.

Similarity search is a very important operation in multimedia databases and other database applications involving complex objects, and involves finding objects in a data set S similar to a query object q, based on some similarity measure. In this article, we focus on methods for similarity search that make the general assumption that similarity is represented with a distance metric d. Existing methods for handling similarity search in this setting typically fall into one of two classes. The first directly indexes the objects based on distances (distance-based indexing), while the second is based on mapping to a vector space (mapping-based approach). The main part of this article is dedicated to a survey of distance-based indexing methods, but we also briefly outline how search occurs in mapping-based methods. We also present a general framework for performing search based on distances, and present algorithms for common types of queries that operate on an arbitrary “search hierarchy. ” These algorithms can be applied on each of the methods presented, provided a suitable search hierarchy is defined.

Abstract—The problem of finding the closest point in high-dimensional spaces is common in pattern recognition. Unfortunately, the complexity of most existing search algorithms, such as k-d tree and R-tree, grows exponentially with dimension, making them impractical for dimensionality above 15. In nearly all applications, the closest point is of interest only if it lies within a user-specified distance e. We present a simple and practical algorithm to efficiently search for the nearest neighbor within Euclidean distance e. The use of projection search combined with a novel data structure dramatically improves performance in high dimensions. A complexity analysis is presented which helps to automatically determine e in structured problems. A comprehensive set of benchmarks clearly shows the superiority of the proposed algorithm for a variety of structured and unstructured search problems. Object recognition is demonstrated as an example application. The simplicity of the algorithm makes it possible to construct an inexpensive hardware search engine which can be 100 times faster than its software equivalent. A C++ implementation of our algorithm is available upon request to search@cs.columbia.edu/CAVE/.

Given a set S of n sites (points), and a distance measure d, the nearest neighbor searching problem is to build a data structure so that given a query point q, the site nearest to q can be found quickly. This paper gives data structures for this problem when the sites and queries are in a metric space. One data structure, D(S), uses a divide-and-conquer recursion. The other data structure, M(S, Q), is somewhat like a skiplist. Both are simple and implementable. The data structures are analyzed when the metric space obeys a certain sphere-packing bound, and when the sites and query points are random and have distributions with an exchangeability property. This property implies, for example, that query point q is a random element of S ∪ {q}. Under these conditions, the preprocessing and space bounds for the algorithms are close to linear in n. They depend also on the sphere-packing bound, and on the logarithm of the distance ratio Υ(S) of S, the ratio of the distance between the farthest pair of points in S to the distance between the closest pair. The data structure M(S, Q) requires as input data an additional set Q, taken to be representative of the query points. The resource bounds of M(S, Q) have a dependence on the distance ratio of S ∪ Q. While M(S, Q) can return wrong answers, its failure probability can be bounded, and is decreasing in a parameter K. Here K ≤ |Q|/n is chosen when building M(S, Q). The expected query time for M(S, Q) is O(K log n) log Υ(S ∪ Q), and the resource bounds increase linearly in K. The data structure D(S) has expected O(log n) O(1) query time, for fixed distance ratio. The preprocessing algorithm for M(S, Q) can be used to solve the all-nearest-neighbor problem for S in O(n(log n) 2 (log Υ(S)) 2) expected time. 1

We present a novel approach to finding a correspondence (alignment) between two curves. The correspondence is based on a notion of an alignment curve which treats both curves symmetrically. We then define a similarity metric based on the alignment curve using two intrinsic properties of the curve, namely, length and curvature. The optimal correspondence is found by an efficient dynamic-programming method both for aligning pairs of curve segments and pairs of closed curves, and is effective in the presence of a variety of transformations of the curve. Finally, the correspondence is shown in application to handwritten character recognition, prototype formation, and object recognition, and is potentially useful in other applications such as registration and tracking.

Given a set S of n sites (points), and a distance measure d, the nearest neighbor searching problem is to build a data structure so that given a query point q, the site nearest to q can be found quickly. This paper gives a data structure for this problem; the data structure is built using the distance function as a “black box”. The structure is able to speed up nearest neighbor searching in a variety of settings, for example: points in low-dimensional or structured Euclidean space, strings under Hamming and edit distance, and bit vector data from an OCR application. The data structures are observed to need linear space, with a modest constant factor. The preprocessing time needed per site is observed to match the query time. The data structure can be viewed as an application of a “kd-tree ” approach in the metric space setting, using Voronoi regions of a subset in place of axis-aligned boxes. 1