WebFountain is a platform for very large-scale text analytics applications that allows uniform access to a wide variety of sources. It enables the deployment of a variety of document-level and corpus-level miners in a scalable manner, and feeds information that drives end-user applications through a set of hosted Web services. Sentiment (or opinion) mining is one of the most useful analyses for various end-user applications, such as reputation management. Instead of classifying the sentiment of an entire document about a subject, our sentiment miner determines sentiment of each subject reference using natural language processing techniques. In this paper, we describe the fully functional system environment and the algorithms, and report the performance of the sentiment miner. The performance of the algorithms was verified on online product review articles, and more general documents including Web pages and news articles.

We describe a system for extracting mentions of terms such as company and product names, in a large and noisy corpus of documents, such as the World Wide Web. Since natural language terms are highly ambiguous, a significant challenge in this task is disambiguating which occurrences of each term are truly related to the right meaning, and which are not. We describe our approach for disambiguation, and show that it achieves very high accuracy with only limited training. This serves as a necessary first step for applications that strive to do analytics on term mentions.

We present sentiment analyzer (SA) that extracts sentiment (or opinion) about a subject from online text documents. Instead of classifying the sentiment of an entire document about a subject, SA detects all references to the given subject, and determines sentiment in each of the references using natural language processing (NLP) techniques. Our sentiment analysis consists of 1) a topic specific feature term extraction, 2) sentiment extraction, and 3) (subject, sentiment) association by relationship analysis. SA utilizes two linguistic resources for the analysis: the sentiment lexicon and the sentiment pattern database. The performance of the algorithms was verified on online product review articles ("digital camera" and "music" reviews), and more general documents including general Webpages and news articles.

This research addresses the issue of automatically segmenting color images into foreground (F) and background (B) regions with the assumption that background regions are relatively smooth but may have gradually varying colors or be slightly textured. A multi-level segmentation scheme is used that involves color clustering, unsupervised segmentation using MDL (minimum description length) principle, edge-based F/B separation, integrated initial F/B segmentation, background modeling, and final F/B separation. The approach has been tested on more than 100 images. Some of the experimental results are presented in this paper.

This research addresses the issue of automatically segmenting color images into foreground (F) and back-ground (B) regions with the assumption that background regions are relatively smooth but may have gradually varying colors or be slightly textured. A multi-level segmentation scheme is used that involves color clustering, unsupervised segmentation using MDL (minimum description length) principle, edge-based F/B separation, and integrated F/B segmentation. The approach has been tested on more than 100 images. Some of the experimental results are presented.

We describe a pseudo-distance function for planar shapes that can be used for similarity retrieval based on shape in image database applications. A shape is represented as a vector of turning angles, and the distance between two vectors is computed using a dynamic programming algorithm. We improve the method over previous similar approaches by allowing multiple starting points along the object perimeter. The results of shape retrieval to match either user hand-drawn shapes or stored object shapes in a database of approximately 2300 shapes demonstrate the methods capabilities.

In image retrieval based on color, the weighted distance between color histograms of two images, represented as a quadratic form, may be defined as a match measure. However, this distance measure is computationally expensive and it operates on high dimensional features (O(N)). We propose the use of low-dimensional, simple to compute distance measures between the color distributions, and show that these are lower bounds on the histogram distance measure. Results on color histogram matching in large image databases show that prefiltering with the simpler distance measures leads to significantly less time complexity because the quadratic histogram distance is now computed on a smaller set of images. The low-dimensional distance measure can also be used for indexing into the database.>

We consider the problem of image segmentation and describe an algorithm that is based on the Minimum Description Length (MDL) principle, is fast, is applicable to multi-band images, and guarantees closed regions. We construct an objective function that, when minimized, yields a partitioning of the image into regions where the pixel values in each band of each region are described by a polynomial surface plus noise. The polynomial orders and their coefficients are determined by the algorithm. The minimization is difficult because (1) it involves a search over a very large space and (2) there is extensive computation required at each stage of the search. To address the first of these problems we use a region-merging minimization algorithm. To address the second we use an incremental polynomial regression that uses computations from the previous stage to compute results in the current stage, resulting in a significant speed up over the non-incremental technique. The segmentation result obtained is suboptimal in general but of high quality. Results on real images are shown.>

On-line collections of images are growing larger and more common, and tools are needed to efficiently manage, organize, and navigate through them. The authors have developed a prototype system called QBIC which allows complex multi-object and multi-feature queries of large image databases. The queries are based on image content-the colors, textures, shapes, and positions of images and the objects/regions they contain. The system computes numeric features to represent the image properties and uses similarity measures based on these features for image retrieval. The focus of the paper is its user interface which allows a user to graphically pose and refine queries based on multiple visual properties of images and their objects.>

We describe how the QBIC (Query By Image Content) system handles "multi-*" queries-queries on large image collections involving multifeatures of each image as a whole and of multiple objects within each image. The queries are based on properties of image content-such as colors, textures, shapes, and edges. The system computes a set of features to describe the above properties, uses distance-like measures on the features to provide similarity based retrieval, and has a graphical interface that enable users pose queries visually. In this paper, we present QBIC indexing algorithms that allow these "multi-*" queries to run efficiently.

An algorithm for generating skeletons of objects in a binary image is described. The algorithm produces a well-centered skeleton with the same simple connectivity as the object, and it allows the object to be either exactly or approximately (to within a known, user-selectable error) reconstructed. Its connectivity and reconstructability properties can be rigorously proved. For approximate reconstruction, the skeleton can also be (almost always) thin and is insensitive to border noise without image prefiltering or skeleton post-pruning, while maintaining the precise error bounds for reconstruction. Because of these properties, its robustness to rotation, pleasing visual appearance, and flexibility, it is well suited for such applications as data compression, image analysis, character recognition, and circuit board inspection.>

A method for detecting curve segments in a digital image is described. The method takes as input a set of edges, and produces as output the number of and parameters for the segments. The method is robust, requiring no thresholds. In place of thresholds, a model class must be provided. Using the information-theoretic minimum description length (MDL) principle, it evaluates each model in the model class, computing the optimal parameters for that model, and selects the best model as the one that gives the shortest encoding of the data and the model. Typical of such methods, the search space is extremely large. It is shown how the Hough transform (HT) may be used to reduce this search space greatly, yielding an efficient (although suboptimal) search. The result is an algorithm in which MDL overcomes standard problems with the HT, while the HT overcomes problems with MDL, and which produces a pleasing set of line segments.>

The paper presents the first parallel skeletonization algorithm whose running time is not dependent on the width of the objects in the image. The algorithm runs in O (log n) time on an EREW PRAM using n/log n processors; previous methods require Omega ( square root n) time.>

Considers the problem of unsupervised multiband image segmentation specifically, using the MDL criterion, an associated complexity measure. According to the MDL principle, the best estimates are those that result in the most compact encoding of the image. An important advantage of such an approach is that it is virtually free of the need to choose arbitrary thresholds, which are typical of many segmentation techniques, and which, in many cases, need to be interactively adjusted in order to get satisfactory results.>

An algorithm for generating connected skeletons of objects in binary images is described. Three main properties of the algorithm are that: (1) it is noniterative, taking a fixed number of passes through the image to produce the skeleton regardless of the width of the objects; (2) it is based on a distance transform that uses a good approximation to the Euclidean distance, giving skeletons that are well centered and robust with respect to rotation; and (3) the skeletons it produces are connected. In addition, the skeletons are thin and allow the objects to be nearly reconstructed. The algorithm can also be run in a mode to produce centerlines, a connected approximation to the skeleton that is less sensitive to border noise and that is useful in image analysis applications.>

An information-theoretic approach is used to derive a new feature selection criterion capable of detecting features that are totally useless. Since the number of useless features is initially unknown, traditional class-separability and distance measures are not capable of coping with this problem. The useless feature-subset is detected by fitting a probability model to a given training set of classified feature-vectors using the minimum-description-length criterion (MDLC) for model selection. The resulting criterion for the Gaussian case is a simple closed-form expression, having a plausible geometric interpretation, and is proved to be consistent, i.e., it yields the true useless subset with probability 1 as the size of the training set grows to infinity. Simulations show excellent results compared to the cross-validation method and other information-theoretic criteria, even for small-sized training sets.>

The application of J. Rissanen's theory (1986) of stochastic complexity to the problem of features selection in statistical pattern recognition (SPR) is discussed. Stochastic complexity provides a general framework for statistical problems such as coding, prediction, estimation, and classification. A brief review of the SPR paradigm and traditional methods of feature selection is presented, followed by a discussion of the basic of stochastic complexity. Two forms of stochastic complexity, minimum description length and an integral form, are applied to the problem of feature selection. Experimental results using simulated data generated with Gaussian distributions are given and compared with results from cross validation, a traditional technique. The stochastic complexity measures give superior results, as measured by their ability to select subsets of relevant features, as well as probability of error computed based on the selected feature subset.>

The authors describe applications and give evaluations of rule-based methods for image analysis. They used rule-based programming methods for two image-analysis applications: segmenting and interpreting images of printed circuit boards, and segmenting satellite images of ice floes. Such problems are often divided into low-, mid-, and high-level processing, and they consider rule-based, or declarative, programming at the mid and high levels. They used Knowledge Tool, an OPS-like language that allows an excellent mix of procedural and declarative code. This ability to mix procedural and declarative code was very useful, but the authors nevertheless found limitations using the rule-based programming model. They describe these and list characteristics of image problems in which declarative coding can be applied.>

The authors present two methods for very-high-precision estimation of straight-line parameters from the Hough transform and compare them with the standard method of taking the absolute peak in the Hough array and with least-squares fitting using both extensive simulation and a number of tests with real target images. Both methods use preprocessing and interpolation in the Hough array, and are based on compensating for effects that cause a spreading of the peak in Hough space. By interpolation, the authors achieve accuracy better than the accumulator cell size. A complete set of simulations show that the two methods produce similar results, which are much better than taking the absolute peak in Hough space. They also compare well with least-square fitting, which was considered optimal in the case of zero mean noise. Results of experiments with real images are reported, confirming that the Hough transform can yield very accurate results, almost as good as least-squares fitting for zero mean noise.>