In this paper, we present an image and text mining tool named TNT. This tool is based on Contextual Exploration and work on different points of view. It can process a corpus of all sizes in French or in English. The web interface associated with this tool, offers a reorganization of the text guided by the images and annotated segments that are associated.

We consider the problem of planning, on the fly, a path whereby a robotic vehicle will cover every point in an initially unknown spatial environment. We describe four strategies (Iterated WaveFront, Greedy-Scan, Delayed Greedy-Scan and Closest-First Scan) for generating cost-effective coverage plans in real time for unknown environments. We give theorems showing the correctness of our planning strategies. Our experiments demonstrate that some of these strategies work significantly better than others, and that the best ones work very well; e.g., in environments having an average of 64,000 locations for the robot to cover, the best strategy returned plans with less than 6% redundant coverage, and took only an average of 0.1 milliseconds per action.

This paper presents a novel multimodal approach to automatically detect learner’s uncertainty through the integration of multiple sensors. An acquisition protocol was established to record participants’ electrical brain activity and physiological signals while interacting with a problem solving system specifically designed for uncertainty elicitation. Data were collected from 38 subjects using 8 sensors and two video feeds. Results from machine learning classifiers support the feasibility of our approach. 81% of accuracy was reached using Support Vector Machine (SVM) algorithm.

Bayesian Network Classifiers are popular approaches for classification problems where instances have to be assigned to one of several classes. However, in many domains, it is necessary to assign instances to multiple classes at the same time. This task has been normally addressed either by (i) transforming the problem into a single-class scenario by defining a new class variable with all of the possible combinations of classes or, (ii) by building an independent classifier for each class variable. Either way, the resulting models do not capture all the relations and dependencies between classes and features resulting into unprecise multidimensional classifiers. In this paper, we introduce a two-step method for learning Multidimensional Bayesian Network Classifiers (MBC) from data based on mutual information measures. The first step of the method learns an initial MBC structure which then, in the second step, is refined. Our approach is simple and keeps all the interactions and dependencies among classes and features. The method was tested on three benchmark multidimensional data-sets. Preliminary experimental results show how our method outperforms state-of-the-art methods used in multidimensional classification.

We examine a corpus of reflective tutorial dialogues between human tutor and student after the student completed introductory physics problems, to predict when the tutor abstracted from the student's preceding turn or when the tutor specialized from the student's preceding turn. Tutor abstraction occurs when the tutor repeats a segment of the student's turn using more general terms. Tutor specialization occurs when the tutor repeats a segment of the student's turn using more concrete terms. We find that features extracted from the reflective dialogue context produce the most predictive models. Also, the tutor abstracts more often when the student shows signs of working at a very detailed level for awhile, and prompts for specification when the student's responses are imprecise.

We present in this paper a novel solution to the problem of paraphrase identification based on lexical dissimilarity kernels. Lexical kernels in conjunction with Support Vector Machines are preferred over other learning methods, e.g. decision trees, due to their ability to handle a high number of features. Dissimilarity-based kernels emphasize dissimilarities among text fragments and therefore are appropriate for text similarity tasks characterized by high lexical overlap. We conducted experiments with our kernels on the Microsoft Research (MSR) Paraphrase Corpus, a standardized data set used for assessing approaches to paraphrase identification. Our reported accuracy results are competitive and robust when compared to state-of-the-art single-model approaches. The results were obtained using 10-fold cross-validation over the entire corpus. We also report competitive results on the test portion of the MSR Paraphrase Corpus, which is the standard way to report results on this corpus.

Learning by observation allows a software agent to learn by watching an expert perform a task. This transfers the burden of training from the expert, who would traditionally need to program the agent, to the agent itself. Most existing approaches to learning by observation perform their observation in a purely passive manner. We propose a case-based reasoning agent that is able to observe passively but can also use mixed-initiative control to request assistance from the expert for difficult input problems. Our agent uses mixed-initiative case acquisition in the game of Tetris. We show that the agent is able to obtain cases it would not have been able to with passive observation alone, is able to improve its performance and places less burden on the expert.

This volume contains the papers accepted for presentation at ICAPS 2011, the Twenty-First International Conferenceon Automated Planning and Scheduling, held in Freiburg, Germany, on June 11–16, 2011. The annual ICAPS conference series was established in 2003 through the merger of two pre-existing biennial conferences, the International Conference on Artificial Intelligence Planning and Scheduling (AIPS) and the European Conference on Planning (ECP). ICAPS continues the traditional high standards of AIPS and ECP as an archival forum for new research in the rapidly developing field of automated planning andscheduling. This volume contains the papers accepted at the conference.

Since the reinvigoration of emotions research, many computationalmodels of emotion have been developed. None ofthese models, however, fully address the integration of emotiongeneration and emotional effect in the context of cognitiveprocesses. This paper seeks to unify various modelsof computational emotions while fully integrating with workdone in cognitive architectures. We propose a perspective onhow this integration would occur and EmoCog, a cognitivearchitecture with mechanisms for emotion generation and effects.

Behavior-based and reactive control methods are popular choices for building fast and lightweight intelligent controllers for resource-constrained systems. Reactive methods are extremely useful in highly resource-constrained applications, but at a cost: they tend to be even more susceptible to certain types of failures than deliberative techniques. Without a planner to adapt to changes, even a small failure can result in incorrect behavior from the entire controller. In this paper, we propose extensions to behavior-based subsumption that can detect four types of failures.