Some thoughts on the automated generation and selection of hypotheses in the earth sciences

Many major scientific problems in the earth sciences can be expressed in terms of finding a sequence of past events which can explain an observed configuration. The configuration might be anything from distribution of continents to a particular landform pattern. Because the past is unobservable, explanations take the form of one or more subjective hypotheses for a specific configuration origin. The subjective approach allows free rein of the human scientific spirit, but the way also is open for personal bias and construction of needlessly elaborate hypotheses. As an alternative, an objective process of hypothesis generation should be achievable by way of a generalized expert system incorporating all geological environments. The generality is required in order to allow for the possibility of producing surprising hypotheses, which might not have been anticipated in the context of any specific geological environment under study. In selecting from many available hypotheses, the criterion of simplicity is a useful means by which a hypothesis subset can be created and listed. However, no guarantee exists that subset members will better approximate the truth. The rationale is rather in terms of a working rule for avoiding unnecessary complexity in explanations. Creation of a generalized expert system would be a major project involving a team effort. Such a project would have the advantage of raising the scientific profile of mathematical geology, a subdiscipline which at present has something of an image of solving little problems elegantly.     

矿产预测专家系统及其在新疆铜—镍硫化物矿床中的应用

系统地使用综合地质，地球物理、地球化学信息预测深部岩浆铜－镍硫化物矿床的找矿预测专家系统，已在新疆哈密地区香山岩体上就用并取得明显效果，设计的两个验证钻孔都已打到矿体。     

专家系统与智能系统

为使专家系统技术尽早在地学领域取得突破，应该研制智能系统。在人机交换系统中加入智能模块，管理调度各功能模块，即实现人机交互系统到智能系统的升级。将专家系统、骨架系统加以扩充，增加与数据库管理系统、科学计算等功能子系统的接口，即成为一个研制智能系统的工具—智能骨架系统。文末描述了一个智能骨架系统—ＳＧＥＳ的结构和用法     

矿区煤及矸石压占灾害决策树分类提取方法

矿业活动为现代工业提供物质基础的同时,极大扰动矿区原有生态结构。煤与矸石压占是矿区多种地质灾害的一种,煤与矸石压占直接造成矿区珍贵的土地资源的浪费,同时,煤与矸石自燃、淋溶、扬尘以及天然放射性对矿区环境有严重影响。近年来,遥感技术以其大范围、实时性等技术优势,在探索矿业活动与生态保护共同途径方面做了大量有益的尝试,尤其在地质灾害监测预警方面体现了广阔应用前景。本文通过对矿区遥感影像的一系列处理(如图像校正,图像增强,图像分类),探索性地采用基于人工智能的决策树分类方法,有效地排除掉同谱异物的现象的影响,高精度地提取出矿区煤与矸石堆放场,为统计煤与矸石压占,预知煤与矸石自燃,判断煤与矸石淋溶、扬尘以及天然放射性对周围环境的影响,以及为今后煤与矸石场的选址规划,提供第一手定量化的地理属性资料。     

建立气象专家系统的一种有效方法

为适合气象专家系统预报目标多样化的需要以及基层台站用户的实际清况,本文提出了建立气象专家系统的一种有效方法——E-B方法,即利用行编辑程序和BASIC程序来书写系统。本系统对气象电报自动选站、选层、选参数,具有类似“黑板系统”的结构,提供多级推理功能,提供快速推理机制,适于用户理解、使用与扩充。本方法特别适用于那些想根据自己的实际情况建立气象专家系统的基层台站。     

Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances

Region-based classification of PolSAR data can be effectively performed by seeking for the assignment that minimizes a distance between prototypes and segments. Silva et al. [“Classification of segments in PolSAR imagery by minimum stochastic distances between wishart distributions.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 6 (3): 1263–1273] used stochastic distances between complex multivariate Wishart models which, differently from other measures, are computationally tractable. In this work we assess the robustness of such approach with respect to errors in the training stage, and propose an extension that alleviates such problems. We introduce robustness in the process by incorporating a combination of radial basis kernel functions and stochastic distances with Support Vector Machines (SVM). We consider several stochastic distances between Wishart: Bhatacharyya, Kullback-Leibler, Chi-Square, Rényi, and Hellinger. We perform two case studies with PolSAR images, both simulated and from actual sensors, and different classification scenarios to compare the performance of Minimum Distance and SVM classification frameworks. With this, we model the situation of imperfect training samples. We show that SVM with the proposed kernel functions achieves better performance with respect to Minimum Distance, at the expense of more computational resources and the need of parameter tuning. Code and data are provided for reproducibility.     

基坑工程设计中的专家系统

介绍了基坑工程及专家系统技术, 提出用专家系统解决基坑工程中的问题的观点, 并介绍了一个实例, 证明了专家系统作为基坑工程设计的计算机辅助手段是争实可行的。     

Drivers of Change and Sustainability in Linked Social–Ecological Systems: An Analysis in the Volta River Basin of Ghana,West Africa

Understanding the nature and relative importance of various drivers of change is crucial for sustainable management of natural resources and in prioritizing management efforts, allocating limited resources, and understanding cumulative effects. For this article, we employed structured an expert judgments approach to identify, characterize, and assess the relative importance of the key biophysical and socioeconomic drivers of change within the Volta River Basin, West Africa. Precipitation variability, water availability, land use change, drought events, and population growth were perceived as most important, while biodiversity loss, social conflicts, pest and disease occurrence, urbanization, and pollution were viewed as less critical. A majority of these drivers were characterized as “slow” acting processes as compared to rapidly changing drivers. Intra- and interexpert groups agreement were found to be significant and convergent, indicating the reliability of the results. The implications of these results for sustainable water resources management and agricultural production are discussed.     

多分类器实例协同训练遥感图像检索

提出一种基于多分类器协同训练的遥感图像检索方法,该方法在不同特征集上分别建立分类器,利用不同分类器的协同性自动标记未知样本,从而有效解决了小样本问题。通过与相关反馈方法进行实验比较分析,结果表明,这两种方法各有优劣,检索结果基本相当,然而多分类器协同训练方法避免了相关反馈过程中人工的多次反馈,自动化程度更高。     

基岩裂隙水神经网络专家系统实例研究

详细介绍了基岩裂隙水寻找与开发的神经网络专家系统的建立,包括神经网络专家系统的组成、结构和神经网络模型的训练,并将训练后的神经网络模型应用于某裂隙水水源地的勘探工作中,取得了较好的效果.