Training pattern replication and weighted class allocation in artificial neural network classification

In some image classifications the importance of classes varies, and it is desirable to weight allocation to selected classes. Often the desire is to weight allocation in favour of classes that are abundant in the area represented by an image at the expense of the less abundant classes. If there is prior knowledge on the distribution of class occurrence, this weighting can be achieved with widely used statistical classifiers by setting appropriate a prioriprobabilities of class membership. With an artificial neural network, the incorporation of prior knowledge is more problematic. An approach to weight class allocation in an artificial neural network classification by replicating selected training patterns is presented. This investigation focuses on a series of classifications in which some classes were more abundant than others, but the same number of training cases were available for each class. By replicating the training patterns of abundant classes the representation of the abundant classes in the training set is increased, reflecting more closely the relative abundance of the classes in an image. Significant increases in classification accuracy were obtained by replicating the training patterns of abundant classes. Furthermore, in comparison against a discriminant analysis for the classification of synthetic aperture radar imagery, the results showed that training pattern replication could be used to weight class allocation with an effect similar to that of incorporating a prioriprobabilities of class membership into the discriminant analysis, and resulted in a significant 20.88%, increase in classification accuracy. This increase in classification accuracy was obtained without any new information, but was the result of making fuller use of what was available.     

Characterizing spatial patterns of invasive species using sub-pixel classifications

Invasive species disrupt landscape patterns and compromise the functionality of ecosystem processes. Non-native saltcedar poses significant threats to native vegetation and groundwater resources in the southwestern U.S. and Mexico, and quantifying spatial and temporal distribution patterns is essential for monitoring its spread. Considerable research focuses on determining the accuracy of various remote sensing techniques for distinguishing saltcedar from native woody riparian vegetation through sub-pixel, or soft classifications. However, there is a lack of research quantifying spatial distribution patterns from these classifications, mainly because landscape metrics, which are commonly used to statistically assess these patterns, require bounded classes and cannot be applied directly to soft classifications. This study tests a new method for discretizing sub-pixel data to generate landscape metrics using a continuum of fractional cover thresholds. The developed approach transforms sub-pixel classifications into discrete maps compliant with metric terms and computes and interprets metric results in the context of the region to explain patterns in the extent, distribution, and connectivity of saltcedar in the Rio Grande basin. Results indicate that landscape metrics are sensitive to sub-pixel values and can vary greatly with fractional cover. Therefore spectral unmixing should be performed prior to metric calculations. Analysis of metric trends provides evidence that saltcedar has expanded away from the immediate riparian zones and is displacing native vegetation. This information, coupled with control management strategies, can be used to target remediation activities along the Rio Grande.     

An evaluation of fuzzy and texture-based classification approaches for mapping regenerating tropical forest classes from Landsat-TM data

Two classification approaches were investigated for the mapping of tropical forests from Landsat-TM data of a region north of Manaus in the Brazilian state of Amazonas. These incorporated textural information and made use of fuzzy approaches to classification. In eleven class classifications the texture-based classifiers (based on a Markov random field model) consistently provided higher classification accuracies than conventional per-pixel maximum likelihood and minimum distance classifications, indicating that they are more able to characterize accurately several regenerating forest classes. Measures of the strength of class memberships derived from three classification algorithms (based on the probability density function, a posteriori probability and the Mahalanobis distance) could be used to derive fuzzy image classifications and be used in post-classification processing. The latter, involving either the summation of class memberships over a local neighbourhood or the application of homogeneity measures, were found to increase classification accuracy by some 10 per cent in comparison with a conventional maximum likelihood classification, a result of comparable accuracy to that derived from the texture-based classifications.     

Interactive foundations of computing

The claim that interactive systems have richer behavior than algorithms is surprisingly easy to prove. Turing machines cannot model interaction machines (which extend Turing machines with interactive input/output) because interaction is not expressible by a finite initial input string. Interaction machines extend the Chomsky hierarchy, are modeled by interaction grammars, and precisely capture fuzzy concepts like open systems and empirical computer science. Computable functions cannot model real-world behavior because functions are too strong an abstraction, sacrificing the ability to model time and other real-world properties to realize formal tractability.Part I of this paper examines extensions to interactive models for algorithms, machines, grammars, and semantics, while Part II considers the expressiveness of different forms of interaction. Interactive identity machines are already more powerful than Turing machines, while noninteractive parallelism and distribution are algorithmic. The extension of Turing to interaction machines parallels that of the lambda to the pi calculus. Asynchronous and nonserializable interaction are shown to be more expressive than sequential interaction (multiple streams are more expressive than a single stream).In Part III, it is shown that interaction machines cannot be described by sound and complete first-order logics (a form of Godel incompleteness), and that incompleteness is inherently necessary to realize greater expressiveness. In the final section the robustness of interactive models in expressing open systems, programming in the large, graphical user interfaces, and agent-oriented artificial intelligence is compared to the robustness of Turing machines. Less technical discussion of these ideas may be found in [25–27]. Applications of interactive models to coordination, objects and components, patterns and frameworks, software engineering, and AI are examined elsewhere [28,29].The propositions P1-P36 embody the principal claims, while observations 01 through 040 provide additional insights.     

Use of Adaptive Networks to Define Highly Predictable Protein Secondary–Structure Classes

We present an adaptive, neural network method that determinesnew classes of protein secondary structure that are significantly more predictable from local amino-acid sequence than conventional classifications. Accurate prediction of the conventional secondary-structure classes, alpha-helix, beta-strand, and coil, from primary sequence has long been an important problem in computational molecular biology, with many ramifications, including multiple-sequence alignment, prediction of functionally important regions of proteins, and prediction of tertiary structure from primary sequence. The algorithm presented here uses adaptive networks to simultaneously examine both sequence and structure data, as available from, for example, the Brookhaven Protein Database, and to determine new secondary-structure classes that can be predicted from sequence with high accuracy. These new classes have both similarities to, and differences from, conventional secondary-structure classes. They represent a new, nontrivial classification of protein secondary structure that is predictable from primary sequence.     

Classification of Symbol Sequences over Their Frequency Dictionaries: Towards the Connection between Structure and Natural Taxonomy

The classifications of bacterial 16S RNA sequences developed over the real and transformed frequency dictionaries have been studied. Two sequences are considered to be close, when their frequency dictionaries are close in Euclidean metrics. A procedure to transform a dictionary is proposed that makes clear some features of the information pattern of a symbol sequence. A comparative study of classifications developed over real frequency dictionaries vs. the transformed ones has been carried out. A correlation between information patterns of nucleotide sequences and taxonomy of the bearer of the sequence was found. The sites with high information value are found to be the main factors of the difference between the classes in a classification. The classification of nucleotide sequences developed over real frequency dictionaries of thickness 3 reveals the best correlation to a gender of bacteria. A set of sequences of the same gender is included entirely into one class, as a rule, and exclusions occur rarely. A hierarchical classification yields one or two taxonomy groups on each level of classification. An unexpectedly often, or unexpectedly rare occurrence of some sites within a sequence makes a basic difference between the structure patterns of the classes yielded; a number of those sites is not to large. Further investigations are necessary in order to campare the sites revealed with those determined due to other methodology.     

Spatiotemporal evolution analysis of time-series land use change using self-organizing map to examine the zoning and scale effects

In this study, we applied a self-organizing map (SOM) neural network method to analyze the spatiotemporal evolution of land-use in Beijing using five time-period classification data from 2005 to 2013. We conducted a spatiotemporal integrated expression and a comparative analysis of the time-series of land use data at 5 km grid level. The experiments at the township level and three different grid levels (20 km, 10 km and 1 km) were simultaneously conducted as the comparison study to analysis the modifiable areal unit problem (MAUP). The land use structure data of analysis unit over 5 years were used as input data for SOM. After training the SOM network, the aggregation modes for different land use types were identified on the output plane. Then, the second-step cluster of the output neurons of the SOM was analyzed to construct a series of land use change trajectories that enabled us to get the spatiotemporal patterns of land use change. The results showed five spatial aggregation patterns and three spatiotemporal change patterns of land use 2005 to 2013. The three patterns of spatiotemporal change represent (1) the expansion of urban areas onto farmland in the southeast plains, (2) the development of forest land in the northwest mountainous areas, and (3) the development of piedmont mixed type land use structures. The results of the comparison experiments showed the zoning effect and the scale effect of MAUP, which were: the 5 km grid-based analysis could provide more precise spatiotemporal evolution patterns in the mountainous area, whereas the township level analysis was more appropriate in the plain area; the pattern of forest land development could be better revealed on 20 km and 10 km grid level, while the pattern of built-up land development could be better revealed on 5 km and 1 km grid level.     

Motif discoveries in unaligned molecular sequences using self-organizing neural networks

In this paper, we study the problem of motif discoveries in unaligned DNA and protein sequences. The problem of motif identification in DNA and protein sequences has been studied for many years in the literature. Major hurdles at this point include computational complexity and reliability of the search algorithms. We propose a self-organizing neural network structure for solving the problem of motif identification in DNA and protein sequences. Our network contains several layers, with each layer performing classifications at different levels. The top layer divides the input space into a small number of regions and the bottom layer classifies all input patterns into motifs and nonmotif patterns. Depending on the number of input patterns to be classified, several layers between the top layer and the bottom layer are needed to perform intermediate classifications. We maintain a low computational complexity through the use of the layered structure so that each pattern's classification is performed with respect to a small subspace of the whole input space. Our self-organizing neural network will grow as needed (e.g., when more motif patterns are classified). It will give the same amount of attention to each input pattern and will not omit any potential motif patterns. Finally, simulation results show that our algorithm outperforms existing algorithms in certain aspects. In particular, simulation results show that our algorithm can identify motifs with more mutations than existing algorithms. Our algorithm works well for long DNA sequences as well.     

Ecosystem mapping at the African continent scale using a hybrid clustering approach based on 1-km resolution multi-annual data from SPOT/VEGETATION

The goal of this study is to propose a new classification of African ecosystems based on an 8-year analysis of Normalized Difference Vegetation Index (NDVI) data sets from SPOT/VEGETATION. We develop two methods of classification. The first method is obtained from a k-nearest neighbour (k-NN) classifier, which represents a simple machine learning algorithm in pattern recognition. The second method is hybrid in that it combines k-NN clustering, hierarchical principles and the Fast Fourier Transform (FFT). The nomenclature of the two classifications relies on three levels of vegetation structural categories based on the Land Cover Classification System (LCCS). The two main outcomes are: (i) The delineation of the spatial distribution of ecosystems into five bioclimatic ecoregions at the African continental scale; (ii) Two ecosystem maps were made sequentially: an initial map with 92 ecosystems from the k-NN, plus a deduced hybrid classification with 73 classes, which better reflects the bio-geographical patterns. The inclusion of bioclimatic information and successive k-NN clustering elements helps to enhance the discrimination of ecosystems. Adopting this hybrid approach makes the ecosystem identification and labelling more flexible and more accurate in comparison to straightforward methods of classification. The validation of the hybrid classification, conducted by crossing-comparisons with validated continental maps, displayed a mapping accuracy of 54% to 61%.     

A Comparison of the Rough Sets and Recursive Partitioning Induction Approaches: An Application to Commercial Loans

Credit scoring is the term used to describe methods utilized for classifying applicants for credit into classes of risk. This paper evaluates two induction approaches, rough sets and decision trees, as techniques for classifying credit (business) applicants. Inductive learning methods, like rough sets and decision trees, have a better knowledge representational structure than neural networks or statistical procedures because they can be used to derive production rules. If decision trees have already been used for credit granting, the rough sets approach is rarely utilized in this domain. In this paper, we use production rules obtained on a sample of 1102 business loans in order to compare the classification abilities of the two techniques. We show that decision trees obtain better results with 87.5% of good classifications with a pruned tree, against 76.7% for rough sets. However, decision trees make more type–II errors than rough sets, but fewer type–I errors.     

Missing values: how many can they be to preserve classification reliability?

Using five medical datasets we detected the influence of missing values on true positive rates and classification accuracy. We randomly marked more and more values as missing and tested their effects on classification accuracy. The classifications were performed with nearest neighbour searching when none, 10, 20, 30% or more values were missing. We also used discriminant analysis and naïve Bayesian method for the classification. We discovered that for a two-class dataset, despite as high as 20–30% missing values, almost as good results as with no missing value could still be produced. If there are more than two classes, over 10–20% missing values are probably too many, at least for small classes with relatively few cases. The more classes and the more classes of different sizes, a classification task is the more sensitive to missing values. On the other hand, when values are missing on the basis of actual distributions affected by some selection or non-random cause and not fully random, classification can tolerate even high numbers of missing values for some datasets.     

Empirically Measuring Soft Knowledge in Visualization

In this paper, we present an empirical study designed to evaluate the hypothesis that humans’ soft knowledge can enhance the cost‐benefit ratio of a visualization process by reducing the potential distortion. In particular, we focused on the impact of three classes of soft knowledge: (i) knowledge about application contexts, (ii) knowledge about the patterns to be observed (i.e., in relation to visualization task), and (iii) knowledge about statistical measures. We mapped these classes into three control variables, and used real‐world time series data to construct stimuli. The results of the study confirmed the positive contribution of each class of knowledge towards the reduction of the potential distortion, while the knowledge about the patterns prevents distortion more effectively than the other two classes.     

An algebraic theory for design of controllers for linear multivariable systems--Part II: Feedback realizations and feedback design

The study of general linear multivariable systems, with possibly different controlled and measured outputs, is continued in this part of the paper. The structure matrices, defined in Part I, are used to solve the feedback realization problem. Feedback realizable transfer functions are then used to solve problems of "regulator type." It is shown that the solutions to problems like disturbance deeoupling, output regulation, and pole placement are all special cases of the solution to a more general problem. Finally, it is shown how the results of Part I and Part II can be combined to solve problems of "servo type" and of "regulator type" simultaneously.     

Building a Medical Decision Support System for Colon Polyp Screening by Using Fuzzy Classification Trees

To deal with highly uncertain and noisy data, for example, biochemical laboratory examinations, a classifier is required to be able to classify an instance into all possible classes and each class is associated with a degree which shows how possible an instance is in that class. According to these degrees, we can discriminate the more possible classes from the less possible classes. The classifier or an expert can pick the most possible one to be the instance class. However, if their discrimination is not distinguishable, it is better that the classifier should not make any prediction, especially when there is incomplete or inadequate data. A fuzzy classifier is proposed to classify the data with noise and uncertainties. Instead of determining a single class for a given instance, fuzzy classification predicts the degree of possibility for every class.Adenomatous polyps are widely accepted to be precancerous lesions and will degenerate into cancers ultimately. Therefore, it is important to generate a predictive method that can identify the patients who have obtained polyps and remove the lesions of them. Considering the uncertainties and noise in the biochemical laboratory examination data, fuzzy classification trees, which integrate decision tree techniques and fuzzy classifications, provide the efficient way to classify the data in order to generate the model for polyp screening.     

Classification of MODIS EVI time series for crop mapping in the state of Mato Grosso,Brazil

Agriculture in Brazilian Amazonia is going through a period of intensification. Crop mapping is important in understanding the way this intensification is occurring and the impact it is having. Two successive classifications based on MODIS (MODerate Resolution Imaging Spectroradiometer)-TERRA/EVI (Enhanced Vegetation Index) time series are applied (1) to map agricultural areas and (2) to identify five crop classes. These classes represent agricultural practices involving three commercial crops (soybean, maize and cotton) planted in single or double cropping systems. Both classifications are based on five steps: (1) analysis of the MODIS/EVI time series, (2) application of a smoothing algorithm, (3) application of a feature selection/extraction process to reduce the data set dimensionality, (4) application of a classifier and (5) application of a post-classification treatment. The first classification detected 95% of the agricultural areas (5 617 250 ha during the 2006–2007 harvest) and correlation coefficients with agricultural statistics exceeded 0.98 for the three crop classes at municipality level. The second classification (overall accuracy?=?74% and kappa index?=?0.675) allowed us to obtain the spatial variability mapping of agricultural practices in the state of Mato Grosso. A total of 30% of the total planted area was cultivated through double cropping systems, especially along the BR163 highway and in the Parecis plateau region.     

Different leaders: Emergent organizational and intellectual leadership in children’s collaborative learning groups

This paper presents two studies that examine emergent leadership in children’s collaborative learning groups. Building on research that finds that leadership moves are distributed among group members during learning activities, we examined whether there were patterns in the distribution of moves, resulting in different types of emergent leaders in groups. Study one examines individual groups working with a teacher, on the same task either with paper or multi-touch tables. Study two examines groups of students in a multi-touch classroom. Results from study one indicated that the leadership was distributed among the students; the distributions aligned with classifications of intellectual leadership moves and organizational leadership moves for about half of the groups. There were no differences in emergent leadership between the multi-touch and paper conditions. These results were explored in more detail in a multi-touch classroom study, exploring emergent leadership in 22 groups of students across six classes. Again, leadership was distributed among group members, and specific roles of intellectual and organizational leader, taken on by two different students, could be identified in half of the groups. These results suggest that attention should be paid to how students are engaging in collaborative learning tasks to ensure all students participate in the intellectual as well as organizational demands of the task. Additionally, the pattern of the distribution of roles suggests that care should be taken to specify behaviors if the role of leader is assigned to collaborative groups.     

具有少量参与者的分层量子密钥分发协议

分层量子密钥分发在量子通信中有重要作用,除了使用EPR与GHZ态可实现分层量子密钥分发,非对称高维多粒子纠缠也为解决分层量子密钥分发提供了一种新思路,这种方法在量子信道使用次数上比传统的使用二部链路的量子密钥分发更有效.介绍了3用户在同构意义下的5种分层密钥结构,并给出4、5用户的可分区分层密钥结构.然后对于所介绍的各类分层密钥结构,通过将上述两种方法进行对比,得到实现各类密钥结构理想化密钥率最高的方案.当量子网络用户大于3且密钥结构可分区时,证明仅使用EPR与GHZ态就可实现各层理想化密钥率是1,并以4、5用户的可分区分层密钥结构为例展开说明.     

Cover: Declassified CORONA KH-4B satellite photography of remains from Rome's desert frontier

Training patterns vary in their importance in image classification. Consequently, the selection and refinement of training sets can have a major impact on classification accuracy. For classification by a neural network, training patterns that lie close to the location of decision boundaries in feature space may aid the derivation of an accurate classification. The role of such border training patterns and their identification is discussed in relation to a series of crop classifications from airborne Thematic Mapper data. It is shown that a neural network trained with a set of border patterns may have a lower accuracy of learning but a significantly higher accuracy of generalization than one trained with a set of patterns drawn from the cores of the classes. Unfortunately, conventional training pattern selection and refinement procedures tend to favour core training patterns. For classification by a neural network, procedures which encourage the inclusion of border training patterns should be adopted as this may facilitate the production of an accurate classification.     

A fuzzy c-means classification of elevation derivatives to extract the morphometric classification of landforms in Snowdonia, Wales

The increasing global coverage of high resolution/large-scale digital elevation data has allowed the study of geomorphological form to receive renewed attention by providing accessible datasets for the characterisation and quantification of land surfaces. Digital elevation models (DEMs) provide quantitative elevation data, but it is the characterisation and extraction of geomorphologically significant measures (morphometric indices) from these raw data that form more informative and useful datasets. Common to many geographical measures, morphometric measures derived from DEMs are dependent on the scale of observation. This paper reports results of employing a fuzzy c-means classification for a sample DEM from Snowdonia, Wales, with a number of morphometric measures at different resolutions as input, and morphometric classification of landforms at each resolution as output. The classifications reveal that different landscape components or morphometric classes are important at different resolutions, and that morphometric classes exhibit resolution dependency in their geographical extents. Examination of the scale dependency and behaviour of morphometric classifications of landforms at different resolutions provides a fuller and more holistic view of the classes present than a single-scale analysis.