The need for appropriate ways to measure the distance or similarity between data is ubiquitous in machine learning, pattern recognition and data mining, but handcrafting such good metrics for specific problems is generally difficult. This has led to the emergence of metric learning, which aims at automatically learning a metric from data and has attracted a lot of interest in machine learning and related fields for the past ten years. This survey paper proposes a systematic review of the metric learning literature, highlighting the pros and cons of each approach. We pay particular attention to Mahalanobis distance metric learning, a well-studied and successful framework, but additionally present a wide range of methods that have recently emerged as powerful alternatives, including nonlinear metric learning, similarity learning and local metric learning. Recent trends and extensions, such as semi-supervised metric learning, metric learning for histogram data and the derivation of generalization guarantees, are also covered. Finally, this survey addresses metric learning for structured data, in particular edit distance learning, and attempts to give an overview of the remaining challenges in metric learning for the years to come.

The recent developments on the syntactical and morphological analysis of natural language text constitute the first step towards a more ambitious goal, that of assigning a proper form of meaning to arbitrary text compounds. Indeed, for certain really "intelligent" applications, such as machine translation, question-answering systems, paraphrase detection, or automatic essay scoring, to name just a few, there will always exist a gap between raw linguistic information (such as part-of-speech labels, for example) and the knowledge of the real world that is needed for the completion of the task in a satisfactory way. Semantic analysis has exactly this role, aiming to close (or reduce as much as possible) this gap by linking the linguistic information with semantic representations that embody this elusive real-world knowledge. The traditional way of adding semantics to sentences is a syntax-driven compositional approach: every word in the sentence is associated with a primitive symbol or a predicate, and these are combined to larger and larger logical forms based on the syntactical rules of the grammar. At the end of the syntactical analysis, the logical representation of the whole sentence is a complex formula that can be fed to a theorem prover for further processing. Although such an approach seems intuitive, it has been shown that it is rather inefficient for any practical application (for example, Bos and Markert (2006) get very low recall scores for a textual entailment task).

The recently introduced series of description logics under the common moniker'DL-Lite' has attracted attention of the description logic and semantic web communities due to the low computational complexity of inference, on the one hand, and the ability to represent conceptual modeling formalisms, on the other. The main aim of this article is to carry out a thorough and systematic investigation of inference in extensions of the original DL-Lite logics along five axes: by (i) adding the Boolean connectives and (ii) number restrictions to concept constructs, (iii) allowing role hierarchies, (iv) allowing role disjointness, symmetry, asymmetry, reflexivity, irreflexivity and transitivity constraints, and (v) adopting or dropping the unique name assumption. We analyze the combined complexity of satisfiability for the resulting logics, as well as the data complexity of instance checking and answering positive existential queries. Our approach is based on embedding DL-Lite logics in suitable fragments of the one-variable first-order logic, which provides useful insights into their properties and, in particular, computational behavior.

In this survey, we present and compare different approaches to estimate Mutual Information (MI) from data to analyse general dependencies between variables of interest in a system. We demonstrate the performance difference of MI versus correlation analysis, which is only optimal in case of linear dependencies. First, we use a piece-wise constant Bayesian methodology using a general Dirichlet prior. In this estimation method, we use a two-stage approach where we approximate the probability distribution first and then calculate the marginal and joint entropies. Here, we demonstrate the performance of this Bayesian approach versus the others for computing the dependency between different variables. We also compare these with linear correlation analysis. Finally, we apply MI and correlation analysis to the identification of the bias in the determination of the aerosol optical depth (AOD) by the satellite based Moderate Resolution Imaging Spectroradiometer (MODIS) and the ground based AErosol RObotic NETwork (AERONET). Here, we observe that the AOD measurements by these two instruments might be different for the same location. The reason of this bias is explored by quantifying the dependencies between the bias and 15 other variables including cloud cover, surface reflectivity and others.

Articulating thought in computer-based media is a powerful means for humans to develop their understanding of phenomena. We have created DynaLearn, an Intelligent Learning Environment that allows learners to acquire conceptual knowledge by constructing and simulating qualitative models of how systems behave. DynaLearn uses diagrammatic representations for learners to express their ideas. This article presents an overview of the DynaLearn system.

In robotics and video games, one often discretizes continuous terrain into a grid with blocked and unblocked grid cells and then uses path-planning algorithms to find a shortest path on the resulting grid graph. This path, however, is typically not a shortest path in the continuous terrain. In this overview article, we discuss a path-planning methodology for quickly finding paths in continuous terrain that are typically shorter than shortest grid paths. Any-angle path-planning algorithms are variants of the heuristic path-planning algorithm A* that find short paths by propagating information along grid edges (like A*, to be fast) without constraining the resulting paths to grid edges (unlike A*, to find short paths).

Articulating thought in computer-based media is a powerful means for humans to develop their understanding of phenomena. We have created DynaLearn, an Intelligent Learning Environment that allows learners to acquire conceptual knowledge by constructing and simulating qualitative models of how systems behave. DynaLearn uses diagrammatic representations for learners to express their ideas. The environment is equipped with semantic technology components capable of generating knowledge-based feedback, and virtual characters enhancing the interaction with learners. Teachers have created course material, and successful evaluation studies have been performed. This article presents an overview of the DynaLearn system.

Virtual humans are computer-generated characters designed to look and behave like real people. Studies have shown that virtual humans can mimic many of the social effects that one finds in human-human interactions such as creating rapport, and people respond to virtual humans in ways that are similar to how they respond to real people. We believe that virtual humans represent a new metaphor for interacting with computers, one in which working with a computer becomes much like interacting with a person and this can bring social elements to the interaction that are not easily supported with conventional interfaces. We present two systems that embody these ideas. The first, the Twins are virtual docents in the Museum of Science, Boston, designed to engage visitors and raise their awareness and knowledge of science. The second SimCoach, uses an empathetic virtual human to provide veterans and their families with information about PTSD and depression.

In robotics and video games, one often discretizes continuous terrain into a grid with blocked and unblocked grid cells and then uses path-planning algorithms to find a shortest path on the resulting grid graph. This path, however, is typically not a shortest path in the continuous terrain. In this overview article, we discuss a path-planning methodology for quickly finding paths in continuous terrain that are typically shorter than shortest grid paths. Any-angle path-planning algorithms are variants of the heuristic path-planning algorithm A* that find short paths by propagating information along grid edges (like A*, to be fast) without constraining the resulting paths to grid edges (unlike A*, to find short paths).

During the last two decades there has been a growing interest in Particle Filtering (PF). However, PF suffers from two long-standing problems that are referred to as sample degeneracy and impoverishment. We are investigating methods that are particularly efficient at Particle Distribution Optimization (PDO) to fight sample degeneracy and impoverishment, with an emphasis on intelligence choices. These methods benefit from such methods as Markov Chain Monte Carlo methods, Mean-shift algorithms, artificial intelligence algorithms (e.g., Particle Swarm Optimization, Genetic Algorithm and Ant Colony Optimization), machine learning approaches (e.g., clustering, splitting and merging) and their hybrids, forming a coherent standpoint to enhance the particle filter. The working mechanism, interrelationship, pros and cons of these approaches are provided. In addition, Approaches that are effective for dealing with high-dimensionality are reviewed. While improving the filter performance in terms of accuracy, robustness and convergence, it is noted that advanced techniques employed in PF often causes additional computational requirement that will in turn sacrifice improvement obtained in real life filtering. This fact, hidden in pure simulations, deserves the attention of the users and designers of new filters.