一种基于安全状态转移的简并测试集生成方法

在总结前人工作的基础上,结合安全操作系统对测试的特殊需求,提出了简并测试集(degenerate test set,简称DTS)的概念,设计了一种使用模型检测的基于安全状态转移的高效测试集生成方法.该方法以状态转移为化简对象,在利用模型检测技术生成测试用例的同时,归并相同的状态转移并化简需求集中的冗余属性,从而最终达到化简测试集的目的.在此基础上,探讨了单个用例失败时用例集的有效性问题,并对DTS生成算法进行了改进.实验结果表明,该方法可以有效地对测试集中的冗余进行化简.

近红外光谱测定溶液中甲醇的含量

测量了含微量甲醇(体积分数为0.04%～0.24%)的系列乙醇水溶液的近红外光谱,利用近红外光谱分析建立了预测甲醇含量的定量分析模型。比较了用外部检验法(Test Set-Validation)和交叉检验法(Cross-Validaton)建立的数学模型,研究了使用外部检验法时,校正集和检验集样品数的改变对模型预测结果的影响。结果发现,当校正集样品数为15检验集样品数为6(总样品数为21)时,使用外部检验法建立的数学模型预测结果较好,其校正集的均方根误差和检验集的预测均方根误差(分别为RMSEE和RMSEP)均较小(分别为0.0115和0.0105),而且很接近。结果表明,近红外光谱方法简单,准确而且实用。

Classification of valence changes of trivalent rare earth ions in alkaline earth borates using artificial neural networks

The investigations of classification on the valence changes from RE3+ to RE2+ (RE = Eu, Sm, Yb, Tm) in host compounds of alkaline earth berate were performed using artificial neural networks (ANNs). For comparison, the common methods of pattern recognition, such as SIMCA, KNN, Fisher discriminant analysis and stepwise discriminant analysis were adopted. A learning set consisting of 24 host compounds and a test set consisting of 12 host compounds were characterized by eight crystal structure parameters. These parameters were reduced from 8 to 4 by leaps and bounds algorithm. The recognition rates from 87.5 to 95.8% and prediction capabilities from 75.0 to 91.7% were obtained. The results provided by ANN method were better than that achieved by the other four methods. (C) 1999 Elsevier Science B.V. All rights reserved.

SEARCH FOR THE BASIC REGULARITIES OF THE TRANSITION EMISSION OF EUROPIUM(II) ION IN COMPLEX FLUORIDES USING PATTERN-RECOGNITION

The relationship between structures of complex fluorides and spectral structure of Eu(II) ion in complex fluorides (AB(m)F(n)) is investigated by means of pattern recognition methods, such as KNN, ALKNN, BAYES, LLM, SIMCA and PCA. A learning set consisting of 32 f-f transition emission host compounds and 31 d-f transition emission host compounds and a test set consisting of 27 host compounds were characterized by 12 crystal structural parameters. These parameters, i.e. features, were reduced from 12 to 6 by multiple criteria for the classification of these host compounds as f-f transition emission or d-f transition emission. A recognition rate from 79.4 to 96.8% and prediction capabilities from 85.2 to 92.6% were obtained. According to the above results, the spectral structures of Eu(II) ion in seven unknown host lattices were predicted.

Learning object segmentation from video data

This memo describes the initial results of a project to create a self-supervised algorithm for learning object segmentation from video data. Developmental psychology and computational experience have demonstrated that the motion segmentation of objects is a simpler, more primitive process than the detection of object boundaries by static image cues. Therefore, motion information provides a plausible supervision signal for learning the static boundary detection task and for evaluating performance on a test set. A video camera and previously developed background subtraction algorithms can automatically produce a large database of motion-segmented images for minimal cost. The purpose of this work is to use the information in such a database to learn how to detect the object boundaries in novel images using static information, such as color, texture, and shape. This work was funded in part by the Office of Naval Research contract #N00014-00-1-0298, in part by the Singapore-MIT Alliance agreement of 11/6/98, and in part by a National Science Foundation Graduate Student Fellowship.

An in silico approach for screening flavonoids as P-glycoprotein inhibitors based on a Bayesian-regularized neural network

P-glycoprotein (P-gp), an ATP-binding cassette (ABC) transporter, functions as a biological barrier by extruding cytotoxic agents out of cells, resulting in an obstacle in chemotherapeutic treatment of cancer. In order to aid in the development of potential P-gp inhibitors, we constructed a quantitative structure-activity relationship (QSAR) model of flavonoids as P-gp inhibitors based on Bayesian-regularized neural network (BRNN). A dataset of 57 flavonoids collected from a literature binding to the C-terminal nucleotide-binding domain of mouse P-gp was compiled. The predictive ability of the model was assessed using a test set that was independent of the training set, which showed a standard error of prediction of 0.146 +/- 0.006 (data scaled from 0 to 1). Meanwhile, two other mathematical tools, back-propagation neural network (BPNN) and partial least squares (PLS) were also attempted to build QSAR models. The BRNN provided slightly better results for the test set compared to BPNN, but the difference was not significant according to F-statistic at p = 0.05. The PLS failed to build a reliable model in the present study. Our study indicates that the BRNN-based in silico model has good potential in facilitating the prediction of P-gp flavonoid inhibitors and might be applied in further drug design.

Default ARTMAP 2

Default ARTMAP combines winner-take-all category node activation during training , distributed activation during testing, and a set of default parameter values that define a ready-to-use, general-purpose neural network system for supervised learning and recognition. Winner-take-all ARTMAP learning is designed so that each input would make a correct prediction if re-presented immediately after its training presentation, passing the "next-input test." Distributed activation has been shown to improve test set prediction on many examples, but an input that made a correct winner-take-all prediction during training could make a different prediction with distributed activation. Default ARTMAP 2 introduces a distributed next-input test during training. On a number of benchmarks, this additional feature of the default system increases accuracy without significantly decreasing code compression. This paper includes a self-contained default ARTMAP 2 algorithm for implementation.

ARTSCENE: A Neural System for Natural Scene Classification

How do humans rapidly recognize a scene? How can neural models capture this biological competence to achieve state-of-the-art scene classification? The ARTSCENE neural system classifies natural scene photographs by using multiple spatial scales to efficiently accumulate evidence for gist and texture. ARTSCENE embodies a coarse-to-fine Texture Size Ranking Principle whereby spatial attention processes multiple scales of scenic information, ranging from global gist to local properties of textures. The model can incrementally learn and predict scene identity by gist information alone and can improve performance through selective attention to scenic textures of progressively smaller size. ARTSCENE discriminates 4 landscape scene categories (coast, forest, mountain and countryside) with up to 91.58% correct on a test set, outperforms alternative models in the literature which use biologically implausible computations, and outperforms component systems that use either gist or texture information alone. Model simulations also show that adjacent textures form higher-order features that are also informative for scene recognition.

Rule Extraction, Fuzzy ARTMAP, and Medical Databases

This paper shows how knowledge, in the form of fuzzy rules, can be derived from a self-organizing supervised learning neural network called fuzzy ARTMAP. Rule extraction proceeds in two stages: pruning removes those recognition nodes whose confidence index falls below a selected threshold; and quantization of continuous learned weights allows the final system state to be translated into a usable set of rules. Simulations on a medical prediction problem, the Pima Indian Diabetes (PID) database, illustrate the method. In the simulations, pruned networks about 1/3 the size of the original actually show improved performance. Quantization yields comprehensible rules with only slight degradation in test set prediction performance.

Accurate localized resolution of identity approach for linear-scaling hybrid density functionals and for many-body perturbation theory

© 2015 IOP Publishing Ltd and Deutsche Physikalische Gesellschaft.A key component in calculations of exchange and correlation energies is the Coulomb operator, which requires the evaluation of two-electron integrals. For localized basis sets, these four-center integrals are most efficiently evaluated with the resolution of identity (RI) technique, which expands basis-function products in an auxiliary basis. In this work we show the practical applicability of a localized RI-variant ('RI-LVL'), which expands products of basis functions only in the subset of those auxiliary basis functions which are located at the same atoms as the basis functions. We demonstrate the accuracy of RI-LVL for Hartree-Fock calculations, for the PBE0 hybrid density functional, as well as for RPA and MP2 perturbation theory. Molecular test sets used include the S22 set of weakly interacting molecules, the G3 test set, as well as the G2-1 and BH76 test sets, and heavy elements including titanium dioxide, copper and gold clusters. Our RI-LVL implementation paves the way for linear-scaling RI-based hybrid functional calculations for large systems and for all-electron many-body perturbation theory with significantly reduced computational and memory cost.

Rapid forensic analysis and identification of "lilac" architectural finishes using Raman spectroscopy

The potential of Raman spectroscopy to discriminate between architectural finishes (household paint) has been investigated using a test set of 51

Hidden Conditional Random Fields for Visual Speech Recognition

In this paper we present the application of Hidden Conditional Random Fields (HCRFs) to modelling speech for visual speech recognition. HCRFs may be easily adapted to model long range dependencies across an observation sequence. As a result visual word recognition performance can be improved as the model is able to take more of a contextual approach to generating state sequences. Results are presented from a speaker-dependent, isolated digit, visual speech recognition task using comparisons with a baseline HMM system. We firstly illustrate that word recognition rates on clean video using HCRFs can be improved by increasing the number of past and future observations being taken into account by each state. Secondly we compare model performances using various levels of video compression on the test set. As far as we are aware this is the first attempted use of HCRFs for visual speech recognition.

P2Y(1) antagonists: Combining receptor-based modeling and QSAR for a quantitative prediction of the biological activity based on consensus scoring

P2Y(1) is an ADP-activated G protein-coupled receptor (GPCR). Its antagonists impede platelet aggregation in vivo and are potential antithrombotic agents. Combining ligand and structure-based modeling we generated a consensus model (LIST-CM) correlating antagonist structures with their potencies. We docked 45 antagonists into our rhodopsin-based human P2Y(1) homology model and calculated docking scores and free binding energies with the Linear Interaction Energy (LIE) method in continuum-solvent. The resulting alignment was also used to build QSAR based on CoMFA, CoMSIA, and molecular descriptors. To benefit from the strength of each technique and compensate for their limitations, we generated our LIST-CM with a PLS regression based on the predictions of each methodology. A test set featuring untested substituents was synthesized and assayed in inhibition of 2-MeSADP-stimulated PLC activity and in radioligand binding. LIST-CM outperformed internal and external predictivity of any individual model to predict accurately the potency of 75% of the test set.

Investigation of the Human Brain Metabolome to Identify Potential Markers for Early Diagnosis and Therapeutic Targets of Alzheimer’s Disease

A study combining high resolution mass spectrometry (liquid chromatography-quadrupole time-of-flight-mass spectrometry, UPLC-QTof-MS) and chemometrics for the analysis of post-mortem brain tissue from subjects with Alzheimer’s disease (AD) (n = 15) and healthy age-matched controls (n = 15) was undertaken. The huge potential of this metabolomics approach for distinguishing AD cases is underlined by the correct prediction of disease status in 94–97% of cases. Predictive power was confirmed in a blind test set of 60 samples, reaching 100% diagnostic accuracy. The approach also indicated compounds significantly altered in concentration following the onset of human AD. Using orthogonal partial least-squares discriminant analysis (OPLS-DA), a multivariate model was created for both modes of acquisition explaining the maximum amount of variation between sample groups (Positive Mode-R2 = 97%; Q2 = 93%; root mean squared error of validation (RMSEV) = 13%; Negative Mode-R2 = 99%; Q2 = 92%; RMSEV = 15%). In brain extracts, 1264 and 1457 ions of interest were detected for the different modes of acquisition (positive and negative, respectively). Incorporation of gender into the model increased predictive accuracy and decreased RMSEV values. High resolution UPLC-QTof-MS has not previously been employed to biochemically profile post-mortem brain tissue, and the novel methods described and validated herein prove its potential for making new discoveries related to the etiology, pathophysiology, and treatment of degenerative brain disorders.

Investigation into DCT Feature Selection for Visual Lip-Based Biometric Authentication

This paper investigated using lip movements as a behavioural biometric for person authentication. The system was trained, evaluated and tested using the XM2VTS dataset, following the Lausanne Protocol configuration II. Features were selected from the DCT coefficients of the greyscale lip image. This paper investigated the number of DCT coefficients selected, the selection process, and static and dynamic feature combinations. Using a Gaussian Mixture Model - Universal Background Model framework an Equal Error Rate of 2.20% was achieved during evaluation and on an unseen test set a False Acceptance Rate of 1.7% and False Rejection Rate of 3.0% was achieved. This compares favourably with face authentication results on the same dataset whilst not being susceptible to spoofing attacks.