Adapting Hidden Naive Bayes for Text Classification

Due to its simplicity, efficiency, and effectiveness, multinomial naive Bayes (MNB) has been widely used for text classification. As in naive Bayes (NB), its assumption of the conditional independence of features is often violated and, therefore, reduces its classification performance. Of the numerous approaches to alleviating its assumption of the conditional independence of features, structure extension has attracted less attention from researchers. To the best of our knowledge, only structure-extended MNB (SEMNB) has been proposed so far. SEMNB averages all weighted super-parent one-dependence multinomial estimators; therefore, it is an ensemble learning model. In this paper, we propose a single model called hidden MNB (HMNB) by adapting the well-known hidden NB (HNB). HMNB creates a hidden parent for each feature, which synthesizes all the other qualified features’ influences. For HMNB to learn, we propose a simple but effective learning algorithm without incurring a high-computational-complexity structure-learning process. Our improved idea can also be used to improve complement NB (CNB) and the one-versus-all-but-one model (OVA), and the resulting models are simply denoted as HCNB and HOVA, respectively. The extensive experiments on eleven benchmark text classification datasets validate the effectiveness of HMNB, HCNB, and HOVA.

The genetics-BIDS extension: Easing the search for genetic data associated with human brain imaging

Metadata are what makes databases searchable. Without them, researchers would have difficulty finding data with features they are interested in. Brain imaging genetics is at the intersection of two disciplines, each with dedicated dictionaries and ontologies facilitating data search and analysis. Here, we present the genetics Brain Imaging Data Structure extension, consisting of metadata files for human brain imaging data to which they are linked, and describe succinctly the genomic and transcriptomic data associated with them, which may be in different databases. This extension will facilitate identifying micro-scale molecular features that are linked to macro-scale imaging repositories, facilitating data aggregation across studies.

IPCAS: a direct-method-based pipeline from phasing to model building and refinement for macromolecular structure determination

A new version (2.0) of the pipeline IPCAS (Iterative Protein Crystal structure Automatic Solution) has been released, in which the program OASIS performs direct-method single-wavelength anomalous diffraction/single isomorphous replacement phasing and direct-method-aided partial-structure extension. IPCAS incorporates the widely used packages CCP4 and PHENIX for locating heavy atoms, density modification, molecular replacement, model building and refinement. Important extensions to the previous version of IPCAS include a resolution screening method for non-crystallographic symmetry searching, an alternate model-building protocol for avoiding premature convergence and direct-method image processing for electron microscopy maps, including single-particle cryo-EM maps. Moreover, a new graphical user interface is provided for controlling and real-time monitoring of the whole dual-space iterative process, which works as a plugin to CCP4i. Applications of the new IPCAS to difficult cases have yielded promising results, including `direct-method phasing and fragment extension' from weak anomalous diffraction signal data and `direct-method-aided partial-structure extension' from low-homology models.

Structure Extension of Tree-Augmented Naive Bayes

Due to the simplicity and competitive classification performance of the naive Bayes (NB), researchers have proposed many approaches to improve NB by weakening its attribute independence assumption. Through the theoretical analysis of Kullback–Leibler divergence, the difference between NB and its variations lies in different orders of conditional mutual information represented by these augmenting edges in the tree-shaped network structure. In this paper, we propose to relax the independence assumption by further generalizing tree-augmented naive Bayes (TAN) from 1-dependence Bayesian network classifiers (BNC) to arbitrary k-dependence. Sub-models of TAN that are built to respectively represent specific conditional dependence relationships may “best match” the conditional probability distribution over the training data. Extensive experimental results reveal that the proposed algorithm achieves bias-variance trade-off and substantially better generalization performance than state-of-the-art classifiers such as logistic regression.

Competition and selectivity in supramolecular synthesis: structural landscape around 1-(pyridylmethyl)-2,2′-biimidazoles

Three isomeric forms of 1-(pyridylmethyl)-2,2′-biimidazole, A1–A3, have been synthesized and subjected to systematic co-crystallizations with selected hydrogen- and halogen-bond donors in order to explore the impact of electrostatics and geometry on the resulting supramolecular architectures. The solid-state supramolecular behavior of A1–A3 is largely consistent in halogen-bonded co-crystals. Only two types of primary interactions, the N–H⋯N/N⋯H–N homomeric hydrogen-bond interactions responsible for the pairing of biimidazole moieties and the I⋯N(pyridine) halogen bonds responsible for the co-crystal formation and structure extension, are present in these systems. The co-crystallizations with hydrogen-bond donors (carboxylic acids), however, lead to multiple possible structural outcomes because of the presence of the biimidazole–acid N–H⋯OC/N⋯H–O heterosynthon that can compete with biimidazole–biimidazole N–H⋯N/N⋯H–N homosynthon. In addition, the somewhat unpredictable nature of proton transfer makes the hydrogen-bonded co-crystals structurally less consistent than their halogen-bonded counterparts.

Development of Simulation Technology for Production of Porous Polymeric Membranes

Porous polymeric membranes are used for ion exchange membranes, membrane filter and separators of batteries owing to its micro-porous structure. Extension method is one of the inexpensive processes of such membrane. However, any suitable stability condition of the process has not yet been clarified. In this study, SEM (Scanning Electron Microscope) observations in production process are carried out and the simulation technology for production is developed for improvement in productivity. In this simulation model, the evolution equation of microscopic damage, constitutive equation depending on microscopic damage and the homogenization method are used for representation of evolution of micro-porous structure of crystalline polymer. It is indicated that numerical results obtained here are in good agreement with the SEM observations.

Crystal structures of three anhydrous salts of the Lewis base 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) with the ring-substituted benzoic acid analogues 4-aminobenzoic acid, 3,5-dinitrobenzoic acid and 3,5-dinitrosalicylic acid

The anhydrous salts of the Lewis base 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) with 4-aminobenzoic acid [1-aza-8-azoniabicyclo[5.4.0]undec-7-ene 4-aminobenzoate, C9H17N2+·C7H6NO2−(I)], 3,5-dinitrobenzoic acid [1-aza-8-azoniabicyclo[5.4.0]undec-7-ene 3,5-dinitrobenzoate, C9H17N2+·C7H3N2O6−, (II)] and 3,5-dinitrosalicylic acid (DNSA) [1-aza-8-azoniabicyclo[5.4.0]undec-7-ene 2-hydroxy-3,5-dinitrobenzoate, C9H17N2+·C7H3N2O7−, (III)] have been determined and their hydrogen-bonded structures are described. In both (II) and (III), the DBU cations have a common disorder in three of the C atoms of the six-membered ring moieties [site-occupancy factors (SOF) = 0.735 (3)/0.265 (3) and 0.686 (4)/0.314 (4), respectively], while in (III), there is additional rotational disorder in the DNSA anion, giving two sites (SOF = 0.72/0.28, values fixed) for the phenol group. In the crystals of (I) and (III), the cation–anion pairs are linked through a primary N—H...Ocarboxylhydrogen bond [2.665 (2) and 2.869 (3) Å, respectively]. In (II), the ion pairs are linked through an asymmetric three-centreR12(4), N—H...O,O′ chelate association. In (I), structure extension is through amine N—H...Ocarboxylhydrogen bonds between the PABA anions, giving a three-dimensional structure. The crystal structures of (II) and (III) are very similar, the cation–anion pairs being associated only through weak C—H...O hydrogen bonds, giving in both overall two-dimensional layered structures lying parallel to (001). No π–π ring associations are present in any of the structures.