In the recent years, the ever-increasing diffusion of GIS technology has facilitated the application of quantitative techniques in landslide hazard assessment. Today a wider spectrum of instability causal factors, mainly morphological and geological in nature, can be cost-effectively acquired, stored and analysed in digital form. In particular, by processing elevation data and its derivatives new morphometric parameters can be readily generated over wide regions, and used as predictors of landslide occurrence. Despite the potential of such technological advancements, landslide hazard mapping remains a major, largely unsolved task. The identification and mapping of past and present landslide bodies, which constitute fundamental steps for predicting future slope-failures, remain highly subjective. Likewise, many basic instability determinants cannot be acquired and mapped with adequate accuracy. Most of the current methods for manipulating instability factors and evaluating hazard levels remain error-prone or questionable.

An Introduction to Bayesian Statistical Decision Processes

Gis Technology in Mapping Landslide Hazard

Bayesian Networks are an important area of research and application within the domain of Artificial Intelligence. This paper explores the nature and implications for Bay esian Networks beginning with an overview and comparison of inferential statistics with Bayes’ Theorem. The nature, relevance and applicability of Bayesian Network theory for issues of advanced computability form the core of the current discussion. A number of current applications using Bayesian networks are examined. The paper concludes with a brief discussion of the appropriateness and limitations of Bayesian Networks for human-computer interaction and automated learning.

An Introduction to Bayesian Networks and Their Contemporary Applications

background

Natural patterns of energy dispersal

Recently Active Networks were proposed as the approach that will enable quick introduction of new services in the current telecommunication infrastructure. This approach allows third party executable codes (proxylet) to be deployed into the network, which creates a big security risk. In this paper we present the application of Bayesian technology in the development of an anomaly detection system for proxylets. This system will be incorporated into our Intrusion Detection System (IDS) that will provide runtime security to ensure active platform integrity is maintained while running third party executable codes.

/pdf/active-platform-security-through-intrusion-detection-using-1crrqazsi7.pdf

Active Platform Security through Intrusion Detection Using Naïve Bayesian Network for Anomaly Detection

The aim of this study was the assessment, at basin scale, of the susceptibility to water-induced soil erosion processes (i.e. gully erosion and sheet/rill erosion) using geomorphological analysis, Geographical Information Systems (GIS) and bivariate statistics. The study was carried out in a watershed located in Southern Italy. A detailed analysis of the pre-existing literature led to select lithology, land-use, slope angle and slope aspect as soil-erosion determining factors (DFs), as they are “non-redundant” and affect both the soil-forming processes that control soil erodibility and the erosive power of running waters. Water-produced erosional landforms, such as gullies and areas severely affected by sheet/rill erosion, were surveyed and mapped using classical techniques of geomorphological analysis. The GIS processing of the geomorphological data allowed calculating the areal density of these landforms in each DF class. Weighting values (Wi), corresponding to the susceptibility level of each DF class, were calculated using bivariate statistics. Finally, GIS overlay procedures of the thematic maps, previously reclassified on the basis of the calculated Wi, allowed to produce two Susceptibility Maps (i.e. Gully Erosion and Sheet and/or Rill Erosion Susceptibility Map). The soil-geomorphological coherence of the produced results has been checked and widely discussed in the framework of the pre-existing literature. Both the geomorphological coherence of the calculated Wi and the results of the validation procedure suggested a good reliability of the method, which is also relatively easy to apply and update.

An Introduction to Bayesian Statistical Decision Processes

Citations

Gis Technology in Mapping Landslide Hazard

An Introduction to Bayesian Networks and Their Contemporary Applications

Natural patterns of energy dispersal

Active Platform Security through Intrusion Detection Using Naïve Bayesian Network for Anomaly Detection

Assessing the susceptibility to water-induced soil erosion using a geomorphological, bivariate statistics-based approach

Related Papers (5)

Geographical Information Systems in Assessing Natural Hazards

Gis Technology in Mapping Landslide Hazard

Multivariate models for landslide hazard evaluation

Landslide hazard analysis

Bayesian Analysis A New Approach to Statistical Decision-Making