The marginal likelihood of the data computed using Bayesian score metrics is at the core of score+search methods when learning Bayesian networks from data. However, common formulations of those Bayesian score metrics ...
详细信息
The marginal likelihood of the data computed using Bayesian score metrics is at the core of score+search methods when learning Bayesian networks from data. However, common formulations of those Bayesian score metrics rely on free parameters which are hard to assess. Recent theoretical and experimental works have also shown that the commonly employed BDe score metric is strongly biased by the particular assignments of its free parameter known as the equivalent sample size. This sensitivity means that poor choices of this parameter lead to inferred BN models whose structure and parameters do not properly represent the distribution generating the data even for large sample sizes. In this paper we argue that the problem is that the BDe metric is based on assumptions about the BN model parameters distribution assumed to generate the data which are too strict and do not hold in real settings. To overcome this issue we introduce here an approach that tries to marginalize the meta-parameter locally, aiming to embrace a wider set of assumptions about these parameters. It is shown experimentally that this approach offers a robust performance, as good as that of the standard BDe metric with an optimum selection of its free parameter and, in consequence, this method prevents the choice of wrong settings for this widely applied Bayesian score metric. (C) 2012 Elsevier Inc. All rights reserved.
We present the work that allowed us to win the Next-Place Prediction task of the Nokia Mobile Data Challenge. Using data collected from the smartphones of 80 users, we explore the characteristics of their mobility tra...
详细信息
We present the work that allowed us to win the Next-Place Prediction task of the Nokia Mobile Data Challenge. Using data collected from the smartphones of 80 users, we explore the characteristics of their mobility traces. We then develop three families of predictors, including tailored models and generic algorithms, to predict, based on instantaneous information only, the next place a user will visit. These predictors are enhanced with aging techniques that allow them to adapt quickly to the users' changes of habit. Finally, we devise various strategies to blend predictors together and take advantage of their diversity, leading to relative improvements of up to 4%. (C) 2013 Elsevier B.V. All rights reserved.
The marginal likelihood of the data computed using Bayesian score metrics is at the core of score+search methods when learning Bayesian networks from data. However, common formulations of those Bayesian score metrics ...
详细信息
The marginal likelihood of the data computed using Bayesian score metrics is at the core of score+search methods when learning Bayesian networks from data. However, common formulations of those Bayesian score metrics rely on free parameters which are hard to assess. Recent theoretical and experimental works have also shown that the commonly employed BDe score metric is strongly biased by the particular assignments of its free parameter known as the equivalent sample size. This sensitivity means that poor choices of this parameter lead to inferred BN models whose structure and parameters do not properly represent the distribution generating the data even for large sample sizes. In this paper we argue that the problem is that the BDe metric is based on assumptions about the BN model parameters distribution assumed to generate the data which are too strict and do not hold in real settings. To overcome this issue we introduce here an approach that tries to marginalize the meta-parameter locally, aiming to embrace a wider set of assumptions about these parameters. It is shown experimentally that this approach offers a robust performance, as good as that of the standard BDe metric with an optimum selection of its free parameter and, in consequence, this method prevents the choice of wrong settings for this widely applied Bayesian score metric. (C) 2012 Elsevier Inc. All rights reserved.
This paper introduces a new probabilisticgraphical model called gated Bayesian network (GBN). This model evolved from the need to represent real world processes that include several distinct phases. In essence a GBN ...
详细信息
ISBN:
(纸本)9781614993308;9781614993292
This paper introduces a new probabilisticgraphical model called gated Bayesian network (GBN). This model evolved from the need to represent real world processes that include several distinct phases. In essence a GBN is a model that combines several Bayesian networks (BN) in such a manner that they may be active or inactive during queries to the model. We use objects called gates to combine BNs, and to activate and deactivate them when predefined logical statements are satisfied. These statements are based on combinations of posterior probabilities of the variables in the BNs. Although GBN is a new formalism there are features of GBNs that are similar to other formalisms and research, including influence diagrams, context-specific independence and structural adaptation.
Sentiment analysis and opinion mining are often addressed as a text classification or entity recognition problem, involving the detection or classification of aspects and subjective phrases. Many approaches do not mod...
详细信息
ISBN:
(纸本)9780769551098
Sentiment analysis and opinion mining are often addressed as a text classification or entity recognition problem, involving the detection or classification of aspects and subjective phrases. Many approaches do not model the relation between aspects and subjective phrases explicitly, implicitly assuming that a subjective phrase refers to a certain aspect if they co-occur together in the same sentence, thus potentially sacrificing accuracy. Instead, in the approach presented in this paper, we model the relation between aspects and subjective phrases explicitly, exploiting a flexible model based on imperatively defined factor graphs (IDF). The extraction of subjective phrases, aspects and the relation between them is modeled as a joint inference problem and compared to a pure pipeline architecture. Our goal is to analyse and quantify to what extent a joint model outperforms a pipeline model in terms of extraction of aspects, subjective phrases and the relation between them. Our results show that, while we have a substantial improvement on predicting targets using a joint inference model, the performance on subjective phrase detection and relation extraction actually decreases only slightly.
Considering the current state in service-robotics, an expert is still necessary to add new tasks and execution behaviors by textual and error-prone programming. Under the consideration that humans typically execute sa...
详细信息
ISBN:
(纸本)9781479929719
Considering the current state in service-robotics, an expert is still necessary to add new tasks and execution behaviors by textual and error-prone programming. Under the consideration that humans typically execute same activities almost identical (or at least similar) and further combine simple behaviors to more complex activities, we follow the constitutive assumption that all complex behaviors are composed of a limited set of atomic behaviors. This work introduces a generic framework for spatial-temporal analysis and classification of arbitrary atomic behaviors. Therefore, we propose the combination of Self-Organizing Maps (SOM) and probabilistic graphical models (PGM) in order to exploit the advantages of both concepts. In this work, we describe the essential methods of the framework briefly, whereas the data-driven training of the spatial-temporal model and the reasoning process are described in detail. In order to demonstrate the potential and to emphasize the high level of generalization and flexibility in real-world environments, the framework is evaluated in an exemplary scenario.
Any regular Gaussian probability distribution that can be represented by an AMP chain graph (CG) can be expressed as a system of linear equations with correlated errors whose structure depends on the CG. However, the ...
详细信息
ISBN:
(纸本)9781614993308;9781614993292
Any regular Gaussian probability distribution that can be represented by an AMP chain graph (CG) can be expressed as a system of linear equations with correlated errors whose structure depends on the CG. However, the CG represents the errors implicitly, as no nodes in the CG correspond to the errors. We propose in this paper to add some deterministic nodes to the CG in order to represent the errors explicitly. We call the result an EAMP CG. We will show that, as desired, every AMP CG is Markov equivalent to its corresponding EAMP CG under marginalization of the error nodes. We will also show that every EAMP CG under marginalization of the error nodes is Markov equivalent to some LWF CG under marginalization of the error nodes, and that the latter is Markov equivalent to some directed and acyclic graph (DAG) under marginalization of the error nodes and conditioning on some selection nodes. This is important because it implies that the independence model represented by an AMP CG can be accounted for by some data generating process that is partially observed and has selection bias. Finally, we will show that EAMP CGs are closed under marginalization. This is a desirable feature because it guarantees parsimonious models under marginalization.
All data and information are not always available at the time of a system design and implementation. Especially in knowledge-based systems, training data could be limited at the early stage and more training data migh...
详细信息
ISBN:
(纸本)9781479903863
All data and information are not always available at the time of a system design and implementation. Especially in knowledge-based systems, training data could be limited at the early stage and more training data might be acquired after the system deployment. This paper is concerned with a method to keep track of knowledge evolution and to detect the changes in the knowledge as more training data are provided. The method assumes that the knowledge is expressed in Bayesian networks and makes use of an agent framework for autonomous processing of knowledge evolution and change detection. It maintains sufficient statistics using a tiled sliding window structure. In order to flexibly encode the strategy for detecting the changes in the joint probability distributions, a set of fuzzy rules are used with which application domains specify their own strategy.
Robots need to effectively use multimodal behaviors, including speech, gaze, and gestures, in support of their users to achieve intended interaction goals, such as improved task performance. This proposed research con...
详细信息
ISBN:
(纸本)9781450321297
Robots need to effectively use multimodal behaviors, including speech, gaze, and gestures, in support of their users to achieve intended interaction goals, such as improved task performance. This proposed research concerns designing effective multimodal behaviors for robots to interact with humans using a data-driven approach. In particular, probabilistic graphical models (PGMs) are used to model the interdependencies among multiple behavioral channels and generate complexly contingent multimodal behaviors for robots to facilitate human-robot interaction. This data-driven approach not only allows the investigation of hidden and temporal relationships among behavioral channels but also provides a holistic perspective on how multimodal behaviors as a whole might shape interaction outcomes. Three studies are proposed to evaluate the proposed data-driven approach and to investigate the dynamics of multimodal behaviors and interpersonal interaction. This research will contribute to the multimodal interaction community in theoretical, methodological, and practical aspects.
Exploiting label correlations is a challenging and crucial problem especially in multi-label learning context. Labels correlations are not necessarily shared by all instances and have generally a local definition. Thi...
详细信息
ISBN:
(纸本)9781479929719
Exploiting label correlations is a challenging and crucial problem especially in multi-label learning context. Labels correlations are not necessarily shared by all instances and have generally a local definition. This paper introduces LOC-LDA, which is a latent variable model that adresses the problem of modeling annotated data by locally exploiting correlations between annotations. In particular, we represent explicitly local dependencies to define the correspondence between specific objects, i.e. regions of images and their annotations. We conducted experiments on a collection of pictures provided by the Wikipedia "Picture of the day" website (1), and evaluated our model on the task of "automatic image annotation". The results validate the effectiveness of our approach.
暂无评论