Boolean Features Research Articles

Relevance feedback for human motion retrieval using a boosting approach

Content-based human motion retrieval (CBMR) has been more and more important with the rapid growth of motion capture data, but the gap between high-level semantic concepts and low-level features hinders further performance improvement. Relevance feedback is an effective tool to narrow the semantic gap and enhance the retrieval performance. However, as a type of variable-length multivariate time series (VLMTS), motion capture data has its own characteristics including high-dimensionality, demand for elastic matching, and difficulty representing different movements in a uniform feature space, which make it much more challenging to design an effective relevance feedback approach. This paper presents a novel boosting approach for CBMR and the main contributions include three aspects. First, to fit in with the characteristics of VLMTS data and meet the real-time requirement of relevance feedback, the ensemble learning framework RankBoost is introduced and k-nearest neighbors combining with dynamic time warping (KNN-DTW) is employed as its weak ranker. Second, the set of extended Boolean geometry features containing much richer geometry elements and measures is used to represent motion content, and it provides a comparatively complete feature set for designing the weak ranker of RankBoost. Third, to solve the over-fitting problem caused by the small-sample training of relevance feedback, a novel learning objective composed of minimizing empirical ranking loss and minimizing the maximum generalization loss is proposed for RankBoost ensemble learning. Experimental results on CMU database and its extended database verify the effectiveness of the proposed approach.

Comparison of machine learning algorithms for classification of the sentences in three clinical practice guidelines.

ObjectivesClinical Practice Guidelines (CPGs) are an effective tool for minimizing the gap between a physician's clinical decision and medical evidence and for modeling the systematic and standardized pathway used to provide better medical treatment to patients.MethodsIn this study, sentences within the clinical guidelines are categorized according to a classification system. We used three clinical guidelines that incorporated knowledge from medical experts in the field of family medicine. These were the seventh report of the Joint National Committee (JNC7) on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure from the National Heart, Lung, and Blood Institute; the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults from the same institution; and the Standards of Medical Care in Diabetes 2010 report from the American Diabetes Association. Three annotators each tagged 346 sentences hand-chosen from these three clinical guidelines. The three annotators then carried out cross-validations of the tagged corpus. We also used various machine learning-based classifiers for sentence classification.ResultsWe conducted experiments using real-valued features and token units, as well as a Boolean feature. The results showed that the combination of maximum entropy-based learning and information gain-based feature extraction gave the best classification performance (over 98% f-measure) in four sentence categories.ConclusionsThis result confirmed the contribution of the feature reduction algorithm and optimal technique for very sparse feature spaces, such as the sentence classification problem in the clinical guideline document.

A multiple window-based co-location pattern mining approach for various types of spatial data

Co-location pattern analysis represents the subsets of spatial events whose instances are found in close geographic proximity. Given a collection of Boolean spatial features, the co-location pattern discovery process finds the subsets of features frequently located together. Key challenges in co-location pattern analysis are modelling of neighbourhood in spatial domain, minimum prevalent threshold to generate collocation patterns and analysing extended spatial objects. We discuss the above key challenges using event centric approach and N-most prevalent co-location patterns approach. We propose a window-based model to find the neighbourhood for point spatial datasets and the multiple window model for extended spatial data objects. We also use N-most prevalent co-location patterns approach to filter the number of co-location pattern generation. We propose a more generic and efficient window-based model algorithm to find colocation patterns. Towards the end, we have done a comparative study of the existing approaches with our proposed approach.

Exploratory analysis of the interrelations between co-located boolean spatial features using network graphs

Visual data mining of spatial data is a challenging task. As exploratory analysis is fundamental, it is beneficial to explore the data using different potential visualisations. In this article, we propose and analyse network graphs as a useful visualisation tool to mine spatial data. Due to their ability to represent complex systems of relationships in a visually insightful and intuitive way, network graphs offer a rich structure that has been recognised in many fields as a powerful visual representation. However, they have not been sufficiently exploited in spatial data mining, where they have principally been used on data that come with an explicit pre-specified network graph structure. This research presents a methodology with which to infer relationship network graphs for large collections of boolean spatial features. The methodology consists of four principal stages: (1) define a co-location model, (2) select the type of co-association of interest, (3) compute statistical diagnostics for these co-associations and (4) construct and visualise a network graph of the statistic from step (3). We illustrate the potential usefulness of the methodology using an example taken from an ecological setting. Specifically, we use network graphs to understand and analyse the potential interactions between potential vector and reservoir species that enable the propagation of leishmaniasis, a disease transmitted by the bite of sandflies.

Spatial Pattern Mining for Soil Erosion Characterization

The protection and the maintenance of the exceptional environment of New Caledonia are major goals for this territory. Among environmental problems, erosion has a strong impact on terrestrial and coastal ecosystems. However, due to the volume of data and its complexity, assessment of hazard at a regional scale is time-consuming, costly and rarely updated. Therefore, understanding and predicting environmental phenomenons need advanced techniques of analysis and modelization. In order to improve the understanding of the erosion phenomenon, this paper proposes a spatial approach based on co-location mining and GIS. Considering a set of Boolean spatial features, the goal of co-location mining is to find subsets of features often located together. This system provides useful and interpretable knowledge based on a new interestingness measure for co-locations and a new visualization of the discovered knowledge. The interestingness measure better reflects the importance of a co-location for the experts, and is completely integrated in the mining process. The visualization approach is a simple, concise and intuitive representation of the co-locations that takes into consideration the spatial nature of the underlying objects and the experts practice.

Non-crossing linked partitions, the partial order ≪ on 𝑁𝐶(𝑛), and the 𝑆-transform

The paper establishes a connection between two recent combinatorial developments in free probability: the non-crossing linked partitions introduced by Dykema in 2007 to study the S S -transform, and the partial order ≪ \ll on N C ( n ) NC(n) introduced by Belinschi and Nica in 2008 in order to study relations between free and Boolean probability. More precisely, one has a canonical bijection between N C L ( n ) NCL(n) (the set of all non-crossing linked partitions of { 1 , … , n } \{ 1, \ldots , n \} ) and the set { ( α , β ) ∣ α , β ∈ N C ( n ) , α ≪ β } \{ ( \alpha , \beta ) \mid \alpha , \beta \in NC(n), \ \alpha \ll \beta \} . As a consequence of this bijection, one gets an alternative description of Dykema’s formula expressing the moments of a non-commutative random variable a a in terms of the coefficients of the reciprocal S S -transform 1 / S a 1/S_a . Moreover, due to the Boolean features of ≪ \ll , this formula can be simplified to a form which resembles the moment-cumulant formula from c c -free probability.

Intestinal Diversion (Colostomy or Ileostomy) in Patients With Severe Bowel Dysfunction Following Spinal Cord Injury

Spinal cord injury (SCI) affects motor and sensory nervous integrity resulting in paralysis of lower or both upper and lower extremities, as well as autonomic nervous system function resulting in neurogenic bowel. SCI leads to diminished or lost sensations of the need to defecate or inability to distinguish the presence of gas versus liquid versus solid stool in the rectal vault. Sensory loss, incomplete evacuation of stool from the rectal vault, immobility, and reduced anal sphincter tone increase the risk of fecal incontinence. Gastrointestinal symptoms are associated with depression, anxiety, and significant impairments in quality of life (QOL) in a significant portion of persons with SCI. 1. To compare clinical, functional, or quality of life outcomes in spinal cord injured patients with gastrointestinal symptoms managed by conservative measures versus intestinal diversion (colostomy or ileostomy). 2. To identify complications associated with ostomy surgery in patients with bowel dysfunction and SCI. A systematic review of electronic databases MEDLINE and CINAHL (from January 1960 to November 2007) was undertaken using the following key words: (1) ostomy, (2) stoma, (3) colostomy, and (4) ileostomy. Boolean features of these databases were used to combine these terms with the key word "spinal cord injuries." Prospective and retrospective studies that directly compared clinical, functional, QOL outcomes or satisfaction among patients with intestinal diversions to patients managed by conservative means were included. Creation of an ostomy in selected patients provides equivocal or superior QOL outcomes when compared to conservative bowel management strategies. Both colostomy and ileostomy surgery significantly reduce the amount of time required for bowel management. Patients who undergo ostomy surgery tend to be satisfied with their surgery, and a significant portion report a desire to be counseled about this option earlier. There are no clear advantages when functional, clinical, or QOL outcomes associated with colostomy are compared to those seen in SCI patients undergoing ileostomy. 1. The WOC nurse plays a pivotal role in both conservative bowel management and the decision to undergo ostomy surgery. 2. Preoperative stoma site marking is vital for the best surgical outcome. 3. The system best suited to an individual is based on a variety of factors including but not limited to stoma location, type of effluent, peristomal plane and contours, and the individual's capabilities and preferences. 4. Some individuals with a sigmoid or descending colostomy may benefit from colostomy irrigation as a management method. 5. Postoperatively, assessment of pressure points for signs of tissue breakdown, evaluation of treatment methods for existing pressure ulcers with suitable modification, and support surface assessment should be included in ongoing annual follow-up visits.

The winning mindset

Suggests that search engines are useful but limited in their application for competitive intelligence searching on the internet, and highlights the importance and effectiveness not just of structured searching but also of creativity. Explains some of the technical limitations of internet searching and suggests conditions in which a competitive intelligence search may be made more effective, pointing out that the value an information professional adds is in having some idea in advance of what they are likely to find. Gives details of what search engines will and will not retrieve, and illustrates how search strategies can be improved through use of the available filtering syntax. Suggests that using Boolean logical operators and other features directly in the search box is likely to produce better results than simply relying on the search engine's advanced search feature. Concludes by re-emphasizing the need for a creative mindset, building on some structure.

Efficient information theoretic extraction of higher order features for improving neural network-based spam e-mail categorization

A novel approach for spam e-mail filtering is herein considered based on information theoretic extraction of higher order features and the committee machines neural network models. An extensive experimental study is organized, the most extensive so far in the literature, based on widely accepted benchmarking e-mail data sets, comparing the proposed methodology with the Naïve Bayes spam filter as well as with the Boosting tree methodology, the linear models-based classification (classification via regression) and the nonlinear models-based classification using simple neural network models, including Multilayer Perceptrons. Moreover, several feature extraction approaches based on information theory are evaluated, comparing mainly the proposed higher order feature extraction methodology with information theoretic extraction of single features. It is shown that the former outperforms the latter and, moreover, that the proposed information theoretic Boolean features present a remarkably high spam categorization performance compared to that of their analog counterparts. Finally, it is shown that the committee machines mail categorization performance compares very favorably to the other rival methods' performance, including the Bayes spam filter which is the most widely used approach in the e-mail services market.

Efficiency versus Convergence of Boolean Kernels for On-Line Learning Algorithms

The paper studies machine learning problems where each example is described using a set of Boolean features and where hypotheses are represented by linear threshold elements. One method of increasing the expressiveness of learned hypotheses in this context is to expand the feature set to include conjunctions of basic features. This can be done explicitly or where possible by using a kernel function. Focusing on the well known Perceptron and Winnow algorithms, the paper demonstrates a tradeoff between the computational efficiency with which the algorithm can be run over the expanded feature space and the generalization ability of the corresponding learning algorithm. We first describe several kernel functions which capture either limited forms of conjunctions or all conjunctions. We show that these kernels can be used to efficiently run the Perceptron algorithm over a feature space of exponentially many conjunctions; however we also show that using such kernels, the Perceptron algorithm can provably make an exponential number of mistakes even when learning simple functions. We then consider the question of whether kernel functions can analogously be used to run the multiplicative-update Winnow algorithm over an expanded feature space of exponentially many conjunctions. Known upper bounds imply that the Winnow algorithm can learn Disjunctive Normal Form (DNF) formulae with a polynomial mistake bound in this setting. However, we prove that it is computationally hard to simulate Winnow's behavior for learning DNF over such a feature set. This implies that the kernel functions which correspond to running Winnow for this problem are not efficiently computable, and that there is no general construction that can run Winnow with kernels.

Efficient information theoretic strategies for classifier combination, feature extraction and performance evaluation in improving false positives and false negatives for spam e-mail filtering

Spam emails are considered as a serious privacy-related violation, besides being a costly, unsolicited communication. Various spam filtering techniques have been so far proposed, mainly based on Naïve Bayesian algorithms. Other Machine Learning algorithms like Boosting trees, or Support Vector Machines (SVM) have already been used with success. However, the number of False Positives (FP) and False Negatives (FN) resulting through applying various spam e-mail filters still remains too high and the problem of spam e-mail categorization cannot be solved completely from a practical viewpoint. In this paper, we propose a novel approach for spam e-mail filtering based on efficient information theoretic techniques for integrating classifiers, for extracting improved features and for properly evaluating categorization accuracy in terms of FP and FN. The goal of the presented methodology is to empirically but explicitly minimize these FP and FN numbers by combining high-performance FP filters with high-performance FN filters emerging from a previous work of the authors [Zorkadis, V., Panayotou, M., & Karras, D. A. (2005). Improved spam e-mail filtering based on committee machines and information theoretic feature extraction. Proceedings of the International Joint Conference on Neural Networks, July 31–August 4, 2005, Montreal, Canada]. 1 1 An abbreviated version of some portions of this article appeared in Zorkadis et al. (2005), published under the IEEE copyright. To this end, Random Committee-based filters along with ADTree-based ones are efficiently combined through information theory, respectively. The experiments conducted are of the most extensive ones so far in the literature, exploiting widely accepted benchmarking e-mail data sets and comparing the proposed methodology with the Naive Bayes spam filter as well as with the Boosting tree methodology, the classification via regression and other machine learning models. It is illustrated by means of novel information theoretic measures of FP & FN filtering performance that the proposed approach is very favorably compared to the other rival methods. Finally, it is found that the proposed information theoretic Boolean features present a remarkably high spam categorization performance.

Construction of all non-reducible descriptors

We present a computational procedure based on a decision-tree model for the identification and construction of all non-reducible descriptors in a supervised pattern recognition problem in which pattern descriptions consist of Boolean features. We illustrate the use of the procedure through its application to the standard problem of recognizing Arabic numerals. We also present an analysis of the computational complexity of the procedure.

Converting numerical classification into text classification

Consider a supervised learning problem in which examples contain both numerical- and text-valued features. To use traditional feature-vector-based learning methods, one could treat the presence or absence of a word as a Boolean feature and use these binary-valued features together with the numerical features. However, the use of a text-classification system on this is a bit more problematic—in the most straight-forward approach each number would be considered a distinct token and treated as a word. This paper presents an alternative approach for the use of text classification methods for supervised learning problems with numerical-valued features in which the numerical features are converted into bag-of-words features, thereby making them directly usable by text classification methods. We show that even on purely numerical-valued data the results of text classification on the derived text-like representation outperforms the more naive numbers-as-tokens representation and, more importantly, is competitive with mature numerical classification methods such as C4.5, Ripper, and SVM. We further show that on mixed-mode data adding numerical features using our approach can improve performance over not adding those features.

WebSail: From On-line Learning to Web Search

We investigate the applicability of on-line learning algorithms to the real-world problem of Web search. Consider that Web documents are indexed using n Boolean features. We first present a practically efficient online learning algorithm TW2 to search for Web documents represented by a disjunction of at most k relevant features. We then design and implement WebSail, a real-time adaptive Web search learner, with TW2 as its learning component. WebSail learns from the user's relevance feedback in real-time and helps the user to search for the desired Web documents. The architecture and performance of WebSail are also discussed.

Improving heuristic mini-max search by supervised learning

This article surveys three techniques for enhancing heuristic game-tree search pioneered in the author's Othello program Logistello, which dominated the computer Othello scene for several years and won against the human World-champion 6–0 in 1997. First, a generalized linear evaluation model (GLEM) is described that combines conjunctions of Boolean features linearly. This approach allows an automatic, data driven exploration of the feature space. Combined with efficient least squares weight fitting, GLEM greatly eases the programmer's task of finding significant features and assigning weights to them. Second, the selective search heuristic ProbCut and its enhancements are discussed. Based on evaluation correlations ProbCut can prune probably irrelevant sub-trees with a prescribed confidence. Tournament results indicate a considerable playing strength improvement compared to full-width α- β search. Third, an opening book framework is presented that enables programs to improve upon previous play and to explore new opening lines by constructing and searching a game-tree based on evaluations of played variations. These general methods represent the state-of-the-art in computer Othello programming and begin to attract researchers in related fields.

Random Boolean nets and features of language

Describes an attempt to cast several abstract properties of natural languages in the framework of Kauffman's (1993, 1995) random Boolean nets (RBN). The properties are complexity, interconnectedness, stability, diversity, and underdeterminedness. A language is modeled as a Boolean net attractor. (Groups of) net nodes are linguistic principles or parameters as posited by Chomskyan theory, according to which the language learner sets parameters to appropriate values on the basis of very limited experience of the language. The setting of one parameter can have a complex effect on the settings of others. A RBN is generated to find an attractor. A state from this attractor is degraded, which represents the degenerate input of language to the learner, and this state is then input to a net with the same connectivity and activation functions as the original net to see whether it converges on the same attractor. Many nets degenerate into attractors representing complete uncertainty. Others settle at intermediate levels of uncertainty, and some manage to overcome the incompleteness of input and converge on attractors identical to that from which the original inputs were (de)generated. Finally, an attempt was made to select a population of such successful nets, using a genetic algorithm where fitness was correlated with an ability to acquire several different languages faithfully. This has so far proved impossible, supporting the Chomskyan suggestion that the human language acquisition capacity is not the outcome of natural selection.

Probably Almost Bayes Decisions

In this paper, we investigate the problem of classifying objects which are given by feature vectors with Boolean entries. Our aim is to “(efficiently) learn probably almost optimal classifications” from examples. A classical approach in pattern recognition uses empirical estimations of the Bayesian discriminant functions for this purpose. We analyze this approach for different classes of distribution functions of Boolean features: k th order Bahadur–Lazarsfeld expansions and k th order Chow expansions. In both cases, we obtain upper bounds for the required sample size which are small polynomials in the relevant parameters and which match the lower bounds known for these classes. Moreover, the learning algorithms are efficient.

Adaptive-predictive game-playing programs

Abstract predictive search (APS) is a new method by which systems can improve their search performance through experience. It is believed that the development of such methods is critical, as currently a tremendous number of computational results are potentially wasted by not integrating search and partial search results into the knowledge of a problem-solving system. In the APS model, pattern formation and associative recall are used to improve or replace search. In this paper, the theory, background, and motivations behind the model are presented and its application to two-player game-playing programs is discussed. In these programs, the system develops a knowledge base of patterns (boolean features) coupled with weights and, using pattern-oriented evaluation, performs only 1-ply search, yet competes respectably with programs that search more. The learning mechanism is a hybrid of machine learning and artificial intelligence techniques that have been successful in other settings. Specific examples and performance results are taken from the domains of Hexapawn, Tic-Tac-Toe, Pente, Othello, and chess.

Criterion for detection of regularities in a Boolean feature space

An inductive-learning criterion is proposed for detection of regularities that describe classes of objects in the Boolean feature space. The criterion decides between hypotheses of random and regular occurrence of combinations of feature values in descriptions of objects in the learning sample.

Connectionist hashed associative memory

This paper proposes the use of simple connectionist networks as hashing functions for sparse associative or content addressable memory. The robustness of such networks in the presence of noisy inputs, and the property that “similar inputs lead to similar outputs” permits (in a probabilistic sense) faster-than-linear retrieval of data which best fits the input. The input may be noisy or have partially specified feature vectors. Mathematical analysis is presented for the Boolean feature case using a network with randomly selected connection strengths.