The recognition of the goal a user is pursing when interacting with a software application is a crucial task for an interface agent as it serves as a context for making opportune interventions to provide assistance to the user. The prediction of the user goal must be fast and a goal recognizer must be able to make early predictions with few observations of the user actions. In this work we propose an approach to automatically build an intention model from a plan corpus using Variable Order Markov models. We claim that following our approach, an interface agent will be capable of accurately ranking the most probable user goals in a time linear to the number of goals modeled.

We show how to translate HTN domain descriptions (if they satisfy certain restrictions) into PDDL so that they can be used by classical planners.  We provide correctness results for our translation algorithm, and show that it runs in linear time and space. We also show that even small and incomplete amounts of HTN knowledge, when translated into PDDL using our algorithm, can greatly improve a classical planner's performance. In experiments on several thousand randomly generated problems in three different planning domains, such knowledge speeded up the well-known Fast-Forward planner by several orders of magnitude, and enabled it to solve much larger problems than it could otherwise solve.

P. This compilation, however, is linear in the number of possible initial states that is exponential in the number of fluents. The problem of planning in the presence of sensing We show nonetheless that even in such cases, a sound, has been addressed in recent years as a nondeterministic complete, and polynomial translation X(P) is possible, provided search problem in belief space. In this that the problem P has bounded contingent width, and work, we use ideas advanced recently for compiling show that the contingent width of almost all existing benchmarks conformant problems into classical ones for introducing is 1; a result that parallels the one reported by Palacios a different approach where contingent problems and Geffner for conformant planning. We then show how the P are mapped into non-deterministic problems non-deterministic but fully observable problem X(P) can be X(P) in state space.

While the most accurate word sense disambiguation systems are built using supervised learning from sense-tagged data, scaling them up to all words of a language has proved elusive, since preparing a sense-tagged corpus for all words of a language is time-consuming and human labor intensive. In this paper, we propose and implement a completely automatic approach to scale up word sense disambiguation to all words of English.  Our approach relies on English-Chinese parallel corpora, English-Chinese bilingual dictionaries, and automatic methods of finding synonyms of Chinese words. No additional human sense annotations or word translations are needed. We conducted a large-scale empirical evaluation on more than 29,000 noun tokens in English texts annotated in OntoNotes 2.0, based on its coarse-grained sense inventory.  The evaluation results show that our approach is able to achieve high accuracy, outperforming the first-sense baseline and coming close to a prior reported approach that requires manual human efforts to provide Chinese translations of English senses.

This paper deals with evolutionary clustering, which refers to the problem of clustering data with distribution drifting along time. Starting from a density estimation view to clustering problems, we propose two general on-line frameworks. In the first framework, i.e., historical data dependent (HDD), current model distribution is designed to approximate both current and historical data distributions. In the second framework, i.e., historical model dependent (HMD), current model distribution is designed to approximate both current data distribution and historical model distribution. Both frameworks are based on the general exponential family mixture (EFM) model. As a result, all conventional clustering algorithms based on EFMs can be extended to evolutionary setting under the two frameworks. Empirical results validate the two frameworks.

In this paper we present an approach to the task of generating and resolving referring expressions (REs) for conversational mobile robots. It is based on a spatial knowledge base encompassing both robot-and human-centric representations. Existing algorithms for the generation of referring expressions (GRE) try to find a description that uniquely identifies the referent with respect to other entities that are in the current context. Mobile robots, however, act in large-scale space, that is environments that are larger than what can be perceived at a glance, e.g. an office building with different floors, each containing several rooms and objects. One challenge when referring to elsewhere is thus to include enough information so that the interlocutors can extend their context appropriately. We address Figure 1: Situated dialogue with a campus service robot this challenge with a method for context construction 2. "the area" that can be used for both generating and resolving 3. "Peter's office at the end of the corridor on the third floor REs - two previously disjoint aspects. Our approach of the Acme Corp. building 7 in the Acme Corp. complex, is embedded in a bidirectional framework 47 Evergreen Terrace, Calisota, Earth, (...)" for natural language processing for robots. Clearly, these REs are valid descriptions of the respective entities in the robot's world representation.

Computing the semantic distance between real-world concepts is crucial for many intelligent applications. We present a novel method that leverages data from `Wikispeedia', an online game played on Wikipedia; players have to reach an article from another, unrelated article, only by clicking links in the articles encountered. In order to automatically infer semantic distances between everyday concepts, our method effectively extracts the common sense displayed by humans during play, and is thus more desirable, from a cognitive point of view, than purely corpus-based methods. We show that our method significantly outperforms Latent Semantic Analysis in a psychometric evaluation of the quality of learned semantic distances.

We introduce a nonparametric approach to multiscale analysis of document corpora using a hierarchical matrix analysis framework called diffusion wavelets. In contrast to eigenvector methods, diffusion wavelets construct multiscale basis functions. In this framework, a hierarchy is automatically constructed by an iterative series of dilation and orthogonalization steps beginning with an initial set of orthogonal basis functions, such as the unit-vector bases. Each set of basis functions at a given level is constructed from the bases at the lower level by dilation using the dyadic powers of a diffusion operator. A novel aspect of our work is that the diffusion analysis is conducted on the space of variables (words), instead of  instances (documents). This approach can automatically and efficiently determine the number of levels of the topical hierarchy, as well as the topics at each level. Multiscale analysis of document corpora is achieved by using the projections of the documents onto the spaces spanned by basis functions at different levels. Further, when the input term-term matrix is a local diffusion operator, the algorithm runs in time approximately linear in the number of non-zero elements of the matrix. The approach is illustrated on various data sets including NIPS conference papers, 20 Newsgroups and TDT2 data.

Graph-based manifold-ranking methods have been successfully applied to topic-focused multi-document summarization. This paper further proposes to use the multi-modality manifold-ranking algorithm for extracting topic-focused summary from multiple documents by considering the within-document sentence relationships and the cross-document sentence relationships as two separate modalities (graphs). Three different fusion schemes, namely linear form, sequential form and score combination form, are exploited in the algorithm. Experimental results on the DUC benchmark datasets demonstrate the effectiveness of the proposed multi-modality learning algorithms with all the three fusion schemes.

We propose a novel approach to context sensitive semantic smoothing by making use of an intermediate, "semantically light" representation for sentences, called Semantically Relatable Sequences (SRS). SRSs of a sentence are tuples of words appearing in the semantic graph of the sentence as linked nodes depicting dependency relations. In contrast to patterns based on consecutive words, SRSs make use of groupings of non-consecutive but semantically related words. Our experiments on TREC AP89 collection show that the mixture model of SRS translation model and Two Stage Language Model (TSLM) of Lafferty and Zhai achieves MAP scores better than the mixture model of MultiWord Expression (MWE) translation model and TSLM. Furthermore, a system, which for each test query selects either the SRS or the MWE mixture model based on better query MAP score, shows significant improvements over the individual mixture models.