David H. Glass

Coherence, Explanation, and Bayesian Networks

This paper discusses the relevance of coherence to deciding between competing explanations. It provides a basic definition of coherence in probabilistic terms, which yields a coherence measure and can easily be extended from the coherence of two beliefs to the coherence of n beliefs. Using this definition, the coherence of a set of beliefs can be obtained by making simple extensions to a Bayesian network. The basic definition suggests a strategy for revising beliefs since a decision to reject a belief can be based on maximising the coherence of the remaining beliefs. It is also argued that coherence can provide a suitable approach for inference to the best explanation.

Integration of Genomic Data for Inferring Protein Complexes from Global Protein–Protein Interaction Networks

Protein-protein interactions (PPIs) play crucial roles in virtually every aspect of cellular function within an organism. One important objective of modern biology is the extraction of functional modules, such as protein complexes from global protein interaction networks. This paper describes how seven genomic features and four experimental interaction data sets were combined using a Bayesian-networks-based data integration approach to infer PPI networks in yeast. Greater coverage and higher accuracy were achieved than in previous high-throughput studies of PPI networks in yeast. A Markov clustering algorithm was then used to extract protein complexes from the inferred protein interaction networks. The quality of the computed complexes was evaluated using the hand-curated complexes from the Munich Information Center for Protein Sequences database and gene-ontology-driven semantic similarity. The results indicated that, by integrating multiple genomic information sources, a better clustering result was obtained in terms of both statistical measures and biological relevance.

Confirmation measures of association rule interestingness

This paper considers advantages of measures of confirmation or evidential support in the context of interestingness of association rules. In particular, it is argued that the way in which they characterize positive/negative association has advantages over other measures such as null-invariant measures. Several properties are reviewed and proposed as requirements for an adequate confirmation measure in a data mining context. While none of the well-known confirmation measures satisfy all of these requirements, two new measures are proposed which do and one of these is shown to have a further advantage. Some results suggest that these measures are relatively stable when the number of null transactions varies.

Coherence Measures and Inference to the Best Explanation

This paper considers an application of work on probabilistic measures of coherence to inference to the best explanation (IBE). Rather than considering information reported from different sources, as is usually the case when discussing coherence measures, the approach adopted here is to use a coherence measure to rank competing explanations in terms of their coherence with a piece of evidence. By adopting such an approach IBE can be made more precise and so a major objection to this mode of reasoning can be addressed. Advantages of the coherence-based approach are pointed out by comparing it with several other ways to characterize ‘best explanation’ and showing that it takes into account their insights while overcoming some of their problems. The consequences of adopting this approach for IBE are discussed in the context of recent discussions about the relationship between IBE and Bayesianism.

Inferring Adaptive Regulation Thresholds and Association Rules from Gene Expression Data through Combinatorial Optimization Learning

There is a need to design computational methods to support the prediction of gene regulatory networks (GRNs). Such models should offer both biologically meaningful and computationally accurate predictions which, in combination with other techniques, may improve large-scale integrative studies. This paper presents a new machine-learning method for the prediction of putative regulatory associations from expression data which exhibit properties never or only partially addressed by other techniques recently published. The method was tested on a Saccharomyces cerevisiae gene expression data set. The results were statistically validated and compared with the relationships inferred by two machine-learning approaches to GRN prediction. Furthermore, the resulting predictions were assessed using domain knowledge. The proposed algorithm may be able to accurately predict relevant biological associations between genes. One of the most relevant features of this new method is the prediction of adaptive regulation thresholds for the discretization of gene expression values, which is required prior to the rule association learning process. Moreover, an important advantage consists of its low computational cost to infer association rules. The proposed system may significantly support exploratory large-scale studies of automated identification of potentially relevant gene expression associations.

Inference to the best explanation: does it track truth?

In the form of inference known as inference to the best explanation there are various ways to characterise what is meant by the best explanation. This paper considers a number of such characterisations including several based on confirmation measures and several based on coherence measures. The goal is to find a measure which adequately captures what is meant by ‘best’ and which also yields the truth with a high degree of probability. Computer simulations are used to show that the overlap coherence measure achieves this goal, enabling the true explanation to be identified almost as often as an approach which simply selects the most probable explanation. Further advantages to this approach are also considered in the case where there is uncertainty in the prior probability distribution.

R-matrix Floquet theory of molecular multiphoton processes: II. Multiphoton ionization of H2

Multiphoton ionization rates for H2 immersed in an intense linearly polarized laser field are calculated using the recently developed R-matrix Floquet theory of molecular multiphoton processes. We assume that the H2 molecule is aligned along the laser polarization direction and we adopt the fixed-nuclei approximation, in which the motion of the target electrons is calculated in the laser field and in the field of the nuclei, which are assumed to be fixed in space. An accurate multi-state wavefunction is employed to calculate one-, two- and four-photon ionization rates for H2 at several internuclear separations over a range of frequencies and intensities. Analysis of the ionization rates reveals the important role played both by resonances corresponding to Rydberg bound states converging to the H2+ ion ground state and by doubly excited states converging to the H2+ ion first excited state. These resonances give rise to resonant enhanced multiphoton ionization peaks in many of the ionization rates studied in this paper, and their possible role in controlling the vibrational population of the final H2+ ion is discussed.

R-matrix-Floquet theory of molecular multiphoton processes

In this paper we describe a unified R -matrix-Floquet theory which can be used to analyse both multiphoton ionization of diatomic molecules and laser-assisted electron-diatomic molecule scattering. Our treatment is non-perturbative and can be applied to arbitrary multi-electron diatomic molecules. We assume that the laser field is monochromatic, monomode, spacially homogeneous and linearly polarized, where the molecular axis can be oriented in an arbitrary direction relative to this polarization direction. The theory takes advantage of the natural division of configuration space into internal and external regions occurring in the R -matrix method, to choose the most appropriate form of the interaction Hamiltonian in each region. This enables standard multi-centre electron-molecule scattering programs to be modified in a straightforward way to solve the problem in the internal region and single-centre atomic multiphoton propagator programs to be extended to solve the problem in the external region. We illustrate our theory by considering the form of the equations for homonuclear diatomic molecules. We also present results for H2 using a simple target wavefunction which provides an important test of the theory and the computer programs and illustrates the role of resonances in two-photon ionization.

Collaborative filtering: the aim of recommender systems and the significance of user ratings

This paper investigates the significance of numeric user ratings in recommender systems by considering their inclusion / exclusion in both the generation and evaluation of recommendations. When standard evaluation metrics are used, experimental results show that inclusion of numeric rating values in the recommendation process does not enhance the results. However, evaluating the accuracy of a recommender algorithm requires identifying the aim of the system. Evaluation metrics such as precision and recall evaluate how well a system performs at recommending items that have been previously rated by the user. By contrast, a new metric, known as Approval Rate, is intended to evaluate how well a system performs at recommending items that would be rated highly by the user. Experimental results demonstrate that these two aims are not synonymous and that for an algorithm to attempt both obscures the investigation. The results also show that appropriate use of numeric rating valuesin the process of calculating user similarity can enhance the performance when Approval Rate is used.

Entailment and symmetry in confirmation measures of interestingness

Abstract In a recent paper Greco et al. (2012) propose a number of properties for measures of rule interestingness. The most fundamental of these properties is that such measures should be Bayesian confirmation measures and this criterion provides the context for the current paper as well. They also propose a number of properties relating to entailment and symmetry in order to discriminate between various confirmation measures which have been proposed in the literature. Working within the same framework of confirmation measures, several limitations of their proposed properties are discussed and a motivation provided for alternative properties. Two new measures of interestingness are proposed and then compared with two other recently proposed measures which also satisfy these properties.