Catálogo de publicaciones - libros

The article presents a job choice model allowing to measure the importance of items of employer images in a multi-stage decision process. Based on scientific research a model for the multi-stage decision process is presented that contains details on how decisions are made on each stage. A method using logistic regression to empirically validate the model is described and compared to an alternative method. The results of applying the method are presented and discussed.

Kalyanam and Shively (1998) and van Heerde et al. (2001) have proposed semiparametric models to estimate the influence of price promotions on brand sales, and both obtained superior performance for their models compared to strictly parametric modeling. Following these researchers, we suggest another semiparametric framework which is based on penalized B-splines to analyze sales promotion effects flexibly. Unlike these researchers, we introduce a stepwise procedure with simultaneous smoothing parameter choice for variable selection. Applying this stepwise routine enables us to deal with product categories with many competitive items without imposing restrictions on the competitive market structure in advance. We illustrate the new methodology in an empirical application using weekly store-level scanner data.

Mining association rules is an important technique for discovering meaningful patterns in transaction databases. In the current literature, the properties of algorithms to mine association rules are discussed in great detail. We present a simple probabilistic framework for transaction data which can be used to simulate transaction data when no associations are present. We use such data and a real-world grocery database to explore the behavior of confidence and lift, two popular interest measures used for rule mining. The results show that confidence is systematically influenced by the frequency of the items in the left-hand-side of rules and that lift performs poorly to filter random noise in transaction data. The probabilistic data modeling approach presented in this paper not only is a valuable framework to analyze interest measures but also provides a starting point for further research to develop new interest measures which are based on statistical tests and geared towards the specific properties of transaction data.

Model based clustering is common approach used in cluster analysis. Here each cluster is characterized by some kind of model, for example — multivariate distribution, regression, principal component etc. One of the most well known approaches in model based clustering is the one proposed by Banfield and Raftery (1993), where each class is described by multivariate normal distribution. Due to the eigenvalue decomposition, one gets flexibility in modeling size, shape and orientation of the clusters, still assuming general elliptical shape of the set of observations. In the paper we consider the other proposal based on the general stochastic approach in two versions:

We propose the use of the copula approach, by representing the multivariate distribution as the copula function of univariate marginal distributions. We give the theoretical bases for such an approach and the algorithms for practical use.

The discussed methods are illustrated by some simulation studies and real examples using financial data.

One of the key challenges in large information systems such as online shops and digital libraries is to discover the relevant knowledge from the enormous volume of information. Recommender systems can be viewed as a way of reducing large information spaces and to personalize information access by providing recommendations for information items based on prior usage.

Collaborative Filtering, the most commonly-used technique for this task, which applies the nearest-neighbor algorithm, does not make use of object attributes. Several so-called content-based and hybrid recommender systems have been proposed, that aim at improving the recommendation quality by incorporating attributes in a collaborative filtering model.

In this paper, we will present an adapted as well as two novel hybrid techniques for recommending items. To evaluate the performances of our approaches, we have conducted empirical evaluations using a movie dataset. These algorithms have been compared with several collaborative filtering and non-hybrid approaches that do not consider attributes. Our experimental evaluations show that our novel hybrid algorithms outperform state-of-the-art algorithms.

The human signature is still the most widely used and accepted form of personal authorisation and verification applied to documents and transactions. In this paper we describe the design and implementation of a flexible and highly configurable framework for the experimental construction and performance investigation of biometric signature verification systems. We focus on a design approach which encompasses a general process model for automatic signature processing, reference instances relating to specific data parsing, feature extraction and classification algorithms and detail the provision of a framework whereby unique signature systems can be easily constructed, trialed and assessed.

In this paper we present a multimodal biometric authentication system based on features of the human hand. A new biometric approach to biometric authentication based on eigen-coefficients of palm, fingers between first and third phalanx, and finger tips, is described. The system was tested on a database containing 10 grey-level images of the left hand and 10 grey-level images of the right hand of 43 people. Preliminary experimental results showed high accuracy of the system in terms of the correct recognition rate (99.49 %) and the equal error rate (0.025 %).

In this paper two potential applications of data hiding technology in a medical scenario are considered. In particular the first application refers to labelling and the watermarking algorithm provides the possibility of directly embed into a medical image the data of the patient; the second application regards authentication and integrity verification, and data hiding is applied for verifying whether and where the content has been modified or falsified since its distribution. Two algorithms developed for these specific purposes will be presented.

This paper demonstrates the feasibility of a new method of hand-geometry recognition based on parameters derived from the contour of the hand. The contour can be modelled by parameters, or features, that can capture more details of the shape of the hand than what is possible with the standard geometrical features used in hand-geometry recognition. The set of features considered in this paper consists of the spatial coordinates of certain landmarks on the contour. The verification performance obtained with contour-based features is compared with the verification performance of other methods described in the literature.

Today biometric techniques are based either on passive (e.g. IrisScan, Face) or active methods (e.g. voice and handwriting). In our work we focus on evaluation of the latter. These methods, also described as behavioral Biometric, are characterized by a trait that is learnt and acquired over time. Several approaches for user authentication have been published, but today they have not yet been evaluated under cultural aspects such as language, script and personal background of users. Especially for handwriting such cultural aspects can lead to a significant and essential outcome, as different spoken and written languages are being used and also the script used for handwriting is different in nature.