Improving Remote Sensing Multiple Classification by Data and Ensemble Selection

In this article, margin theory is exploited to design better ensemble classifiers for remote sensing data. A semi-supervised version of the ensemble margin is at the core of this work. Some major challenges in ensemble learning are investigated using this paradigm in the difficult context of land cover classification: selecting the most informative instances to form an appropriate training set, and selecting the best ensemble members. The main contribution of this work lies in the explicit use of the ensemble margin as a decision method to select training data and base classifiers in an ensemble learning framework. The selection of training data is achieved through an innovative iterative guided bagging algorithm exploiting low-margin instances. The overall classification accuracy is improved by up to 3%, with more dramatic improvement in per-class accuracy (up to 12%). The selection of ensemble base classifiers is achieved by an ordering-based ensemble-selection algorithm relying on an original margin-based criterion that also targets low-margin instances. This method reduces the complexity (ensemble size under 30) but maintains performance.

The Transition in Household Energy Use for Cooking in India: Evidence from a Longitudinal Survey

This article seeks to assess the transition in household energy use for cooking in India based on data from two rounds of the India Human Development Survey in 2004–2005 and 2011–2012. In this study, we have used the multinomial logistic regression and Multiple Classification Analysis conversion model to assess the transition in household energy use according to the socio-economic characteristics of households. Our findings suggest that although the transition from solid fuel to clean fuel is universal across households, it is greater among the socio-economically better-off households than their poorer counterparts. The use of solid fuel for cooking was more prevalent among the socio-economically disadvantaged households than among their socio-economically better-off counterparts in both 2004–2005 and 2011–2012. Convergence in clean cooking fuel use across the households can be possible only when socio-economically disadvantaged households progress faster than their already better-off counterparts. JEL Codes: B5, C23, D31, I3, Q5

Evidential classification of incomplete instance based on K-nearest centroid neighbor

Classification of incomplete instance is a challenging problem due to the missing features generally cause uncertainty in the classification result. A new evidential classification method of incomplete instance based on adaptive imputation thanks to the framework of evidence theory. Specifically, the missing values of different incomplete instances in test set are adaptively estimated based on Shannon entropy and K-nearest centroid neighbors (KNCNs) technology. The single or multiple edited instances (with estimations) then are classified by the chosen classifier to get single or multiple classification results for the instances with different discounting (weighting) factors, and a new adaptive global fusion method finally is proposed to unify the different discounted results. The proposed method can well capture the imprecision degree of classification by submitting the instances that are difficult to be classified into a specific class to associate the meta-class and effectively reduce the classification error rates. The effectiveness and robustness of the proposed method has been tested through four experiments with artificial and real datasets.

Research on Ultrasonic Testing with Wavelet Packet Analysis for Shotcrete

Shotcrete structures are widely used in tunnel engineering. Quality inspection is difficult, and the traditional ultrasonic testing (UT) method based on first arrival velocity has limitations. In this paper, shotcrete-rock specimens were made in a laboratory and evaluated using UT. Wavelet packet decomposition is introduced for better frequency analysis of the condition evaluation. Two methods, including calculation of the energy eigenvalues and machine learning, are used to describe the contact quality at the interface between the shotcrete and rock. The relative energy eigenvalue increases with the gradual reduction of contact quality, which can become a quantitative index of the contact quality. Machine learning performed well in the rapid recognition of discontinuities in the multiple-classification models. Both methods based on wavelet packet decomposition achieved good results in identifying discontinuities and have the potential to be used in practical engineering applications.

Correlates and Consequences of Delayed Marriage in Malaysia

This paper aims to examine the correlates of age at first marriage and the consequences of late marriage. Data for this paper were drawn from the 2014 Malaysian Population and Family Survey. Simple cross-tabulation and multiple classification analysis were used for the analysis. Age at marriage of women varied across socio-economic groups. Among the ethnic groups, the Other Bumiputera entered marriage earliest, followed by the Malays, Indians and Chinese. Age at marriage was positively associated with urbanisation, educational level, and women’s autonomy in marriage. The assumption of modern norms and ideas, and escalating cost of marriage are important determinants of marriage postponement. Late marriage has a direct impact on demographic outcomes, resulting in ultra-low fertility for some groups of the population. Marriage postponement has positive socio-economic outcomes for individuals. However, postponing marriage beyond the prime reproductive age may result in some reproductive health problems.

Sustainable Development and Rural Tourism in Depopulated Areas

The high degree of depopulation in certain areas of Spain is a serious threat to the country, and is aggravated by the ongoing loss of population from those areas. Rural tourism is one of the activities that can help prevent this depopulation. However, to successfully promote such tourism, we must consider the elements that have the greatest influence on tourists when they choose one location over another, or one accommodation over another. Extensive data have been collected from 1658 valid surveys of tourists in one of the most depopulated areas of Spain. Several multivariate techniques were then applied to the data, including Principal Component Analysis (PCA) and Multiple Classification Analysis (MCA). Factors were obtained that identified both the different motivations that influence tourists, and the variables that identify the province based on its image. An analysis was then made of how both the variables thus identified the influence of the formation of the image that tourists take away from the visit. Tourists are most strongly motivated by natural landscapes, monuments, or events of cultural interest, i.e., natural and cultural attractions rather than social ones, and the cognitive image has the greatest influence on the formation of the new image. The principal findings of this research are that the future of many of these depopulated areas depends on successfully promoting both their beautiful landscapes and their cultural heritage, as well as developing and improving the areas themselves so that the depopulation is slowed down or even reversed, to the benefit of the local population. This would also benefit the local and regional authorities and the establishments linked to rural tourism in the area, increasing their profits and raising the level of employment in the province.

MULTIPLE CLASSIFICATION ALGORITHMS UNIMODAL AND MULTIMODAL TARGET RECOGNITION SYSTEMS

Target recognition is of great importance in military and civil applications – object detection, security and surveillance, access and border control, etc. In the article the general structure and main components of a target recognition system are presented. The characteristics such as availability, distinctiveness, robustness, and accessibility are described, which influence the reliability of a TRS. The graph presentations and mathematical descriptions of a unimodal and multimodal TRS are given. The mathematical models for a probability of correct target recognition in these systems are presented. To increase the reliability of TRS, a new approach was proposed – to use a set of classification algorithms in the systems. This approach permits the development of new kinds of systems - Multiple Classification Algorithms Unimodal and Multimodal Systems (MAUMS and MAMMS). The graph presentations, mathematical descriptions of the MAUMS and MAMMS are described. The evaluation of the correct target recognition was made for different systems. The conditions of systems' effectiveness were established. The modality of the algorithm's recognition probabilitymaximal value determination for an established threshold level of the system's recognition probability was proposed, which will describe the requirements for the quality and, respectively, the costs of the recognition algorithms. The proposed theory permits the system's design depending on a predetermined recognition probability.