1,721,035 research outputs found
Mystery shopping at greek banks : a Bayesian network analysis
No full textThe banking industry is highly competitive, and customer satisfaction plays an essential rule. Hence, methods to evaluate customer satisfaction are im-portant for bank managers. In this work, two instruments for customer satisfaction analysis are combined: Mystery shopping methodology and Bayesian Networks.
Mystery shoppers are used to survey and monitor the quality of customer service and to identify areas requiring enhancement. After each visit they complete a report prepared in advance on their service experience. Bayesian Networks are then used to provide a pictorial representation of the dependence structure between the variables of interest, and are used to study the effect of different improvement strategies. We present a real data analysis concerning customer satisfaction in Greek banks
Ordered response models for cyber risk
Full text linkNegli ultimi anni si `e registrato un interesse crescente da parte degli stu-
diosi riguardo il problema del cyber risk e la sua misurazione. Poich`e i dati quan-
titativi sulle perdite sono raramente disponibili, essi sono spesso rilevati su scala
ordinale e riguardano il livello di gravit`a degli attacchi cibernetici. Risulta pertanto
naturale valutare il cyber risk mediante modelli di risposta ordinale che legano la
variabile gravit`a a variabili esplicative spesso di natura categorica. L’effetto di tali
variabili viene valutato utilizzando l’AME (Average Marginal Effect). Il modello
viene applicato a dati reali sulla gravit`a degli attacchi rilevati nel mondo nel 2018.In the last years there have been a scholars increasing interest in cyber-
security risk measurement, data security, and privacy protection. Since quantitative
loss data are rarely available, we deal with ordinal data representing experts’ eval-
uation of the severity of the attacks. Due to the ordinal nature of the available data,
it turns natural to rely on cumulative link models that allows us to express the cu-
mulative probabilities associated with the different severity levels as a non linear
function of a suitable set of explanatory variables. We evaluate the effect of each
explanatory categorical variable on the risk level using the Average Marginal Effect.
We apply our model to a real data set that includes information on serious cyber
attacks occurred worldwide in 2018
Dependency Networks And Bayesian Networks For Web Mining
Full text linkFollowing the approach described by Heckerman et al. ([5]), we present an application of Dependency Networks and Bayesian Networks to the analysis of a clickstream data set. Our target is to discover which paths are more often followed by
the users. The relation between one web page and another one is represent by a
direct graph. Whereas Bayesian Networks use direct acyclic graphs, Dependency
Networks may contain cyclic structures. The analysis will be performed with the
WinMine Toolkit software
Model Determination for Discrete Graphical Models
No full textRJMCM for Model Determination for Discrete Graphical Models
Enhancing cyber risk assessment: Unfolding ordinal data models for effective analysis
No full textIn today’s increasingly digitalized world, where organizations face the constant impact of technological advancements, the proliferation of cyber
attacks poses a significant threat across various industries. While quantitative loss data is often scarce, experts in the field can provide a qualitative
assessment of cyber attack severity on an ordinal scale. To analyze cyber risk effectively, it is natural to employ order response models. These
models allow for exploring how experts assess the severity of cyberattacks based on a range of explanatory variables that describe the attack’s
characteristics. Additionally, a measure of the diffusion of attack effects is incorporated through a network structure into the model’s explanatory
variables. Apart from describing the methodology behind these models, a comprehensive analysis of a real dataset is presented. This dataset
includes information on serious cyber attacks that have occurred worldwide, offering valuable insights into the practical application of the approach.
By unravelling the complexities of cyber risk assessment and leveraging ordinal data models, the aim is to empower organizations to better
understand and mitigate the potential impact of cyberattacks
Generalized belief change with imprecise probabilities and graphical models
Full text linkWe provide a theoretical investigation of probabilistic belief revision in complex frameworks, under extended conditions of uncertainty, inconsistency and imprecision. We motivate our kinematical approach by specializing our discussion to probabilistic reasoning with graphical models, whose modular representation allows for efficient inference. Most results in this direction are derived from the relevant work of Chan and Darwiche (2005), that first proved the inter-reducibility of virtual and probabilistic evidence. Such forms of information, deeply distinct in their meaning, are extended to the conditional and imprecise frameworks, allowing further generalizations, e.g. to experts' qualitative assessments. Belief aggregation and iterated revision of a rational agent's belief are also explored
Body plethysmographic study of specific airway resistance in a sample of healthy adults
No full textBackground and objective: sRaw (specific airway resistance) is a corrected index (Raw multiplied by thoracic gas volume) that describes airway behaviour regardless of lung volume. Normal values of sRaw in adult subjects have never been formally defined. To establish sRaw interpretation criteria and to define a range of reference values, we evaluated variability, reproducibility and reliability of sRaw measurements in a group of healthy adults. Methods: We analysed 517 subjects of both genders, aged 18-65 (group A), and to assess the reproducibility of the measurements, we investigated intra-individual variation and potential daily and weekly sRaw rhythms in a subgroup of 18 co-operative healthy subjects (group B). Results: In group A, there was no pattern of association between any of the considered anthropometric parameters; mean sRaw was higher in men (6.24 vs 5.95 cmH2O s in females; P = 0.0128), but when the data were stratified by age, gender-related differences were only found in the group aged 46-60 (males 6.45 cmH2O s, females 6.01 cmH 2O s; P = 0.0219). In group B, there was no statistically significant, time-dependent variation during the single tests, nor any circadian or weekly rhythm. Conclusions: sRaw is a reliable parameter; therefore, we propose that the lower and upper 95% confidence limits should be considered as reference values for adults of both genders, regardless of age. The availability of reference values may be useful in clinical practice and research. Normal reference values for specific airway resistance in children and adults have never been formally defined. We studied this parameter in 517 healthy subjects. We now propose that the lower and upper 95% confidence limits should be considered as reference values for both genders, regardless of age
Unfolding models for ordinal data in cyber risk assessment
Full text linkIn an increasingly digitalized world, where organizations are affected by technological evolu-
tion, cyber attacks are multiplying rapidly. They have an impact on every class of business and
no industry can consider itself immune to them. Quantitative loss data are rarely available while
it is possible to obtain a qualitative evaluation, expressed on a rating scale, from experts of the
sector. Hence, we focus on ordinal data models for cyber risk evaluation (rating) with particular
emphasis on a mixture model taking into account the uncertainty in the process of scoring. We
examine a set of data regarding cyber attacks that occurred worldwide before and during the
pandemic due to Covid-19. The aim of our analysis is to investigate if Covid-19 has affected ex-
perts’ uncertainty and assessment, and identify the relevant factors which influence the severity
of an attack
Default probability estimation: bayesian Pair Copula model
No full textIn this paper we present a novel Bayesian methodology for default prob-
ability estimation based on multivariate contingent claim analysis and pair copula
theory. In order to compute the default probability of a firm, we use balance sheet
data as a proxy of the equity value. A pair copula approach is applied to obtain the
firm pricing function, and Monte Carlo simulations are then used to calculate the
distribution of the default probability. The methodology will be illustrated through
an application to real dat
- …
