Abstract The overall goal of this research is to study reasoning under uncertainty by combining B... more Abstract The overall goal of this research is to study reasoning under uncertainty by combining Bayesian Networks and Case-Based Reasoning through constructing an experimental decision support system for classification of cancer pain. We have experimentally analysed a medical dataset in order to reveal properties of the data with respect to properties of the two reasoning methods.
In this paper we propose a framework, called mixtures of truncated basis functions (MoTBFs), for ... more In this paper we propose a framework, called mixtures of truncated basis functions (MoTBFs), for representing general hybrid Bayesian networks. The proposed framework generalizes both the mixture of truncated exponentials (MTEs) framework and the Mixture of Polynomials (MoPs) framework.
Abstract In this paper we will present the basic properties of Bayesian network models, and discu... more Abstract In this paper we will present the basic properties of Bayesian network models, and discuss why this modelling framework is well suited for an application in collaborative filtering. We will then describe a new collaborative filtering model, which is built using a Bayesian network. By examining how the model operates on the well-known MovieLens-dataset, we can inspect its merits both qualitatively and quantitatively.
The second symposium of the Norwegian AI Society (NAIS) was held in Gjøvik November 22nd, 2010. N... more The second symposium of the Norwegian AI Society (NAIS) was held in Gjøvik November 22nd, 2010. NAIS was established in the early 1980s, but has been dormant for approximately a decade. This symposium is the second after NAIS has been reinitiated. The symposium functions as a scientific meeting-place for practitioners and theoreticians working in artificial intelligence in Norway.
When constructing a Bayesian network, it can be advantageous to employ structural learning algori... more When constructing a Bayesian network, it can be advantageous to employ structural learning algorithms to combine knowledge captured in databases with prior information provided by domain experts. Unfortunately, conventional learning algorithms do not easily incorporate prior information, if this information is too vague to be encoded as properties that are local to families of variables. For instance, conventional algorithms do not exploit prior information about repetitive structures, which are often found in object oriented domains such as computer networks, large pedigrees and genetic analysis.
Mathematical and statistical methods …, Jan 1, 2003
We investigate the mathematical modelling of maintenance and repair of components that can fail d... more We investigate the mathematical modelling of maintenance and repair of components that can fail due to a variety of failure mechanisms. Our motivation is to build a model, which can be used to unveil aspects of the quality of the maintenance performed. The model we propose is motivated by imperfect repair models, but extended to model preventive maintenance as one of several "competing risks". This helps us to avoid problems of identifiability previously reported in connection with imperfect repair models. Parameter estimation in the model is based on maximum likelihood calculations. The model is tested using real data from the OREDA database, and the results are compared to results from standard repair models.
Annals of Mathematics and Artificial Intelligence, Jan 1, 2001
This paper describes a method for parameter learning in Object-Oriented Bayesian Networks (OOBNs)... more This paper describes a method for parameter learning in Object-Oriented Bayesian Networks (OOBNs). We propose a methodology for learning parameters in OOBNs, and prove that maintaining the object orientation imposed by the prior model will increase the learning speed in object-oriented domains. We also propose a method to efficiently estimate the probability parameters in domains that are not strictly object oriented. Finally, we attack type uncertainty, a special case of model uncertainty typical to object-oriented domains.
Classification problems have a long history in the machine learning literature. One of the simple... more Classification problems have a long history in the machine learning literature. One of the simplest, and yet most consistently well-performing set of classifiers is the Naïve Bayes models. However, an inherent problem with these classifiers is the assumption that all attributes used to describe an instance are conditionally independent given the class of that instance. When this assumption is violated (which is often the case in practice) it can reduce classification accuracy due to "information double-counting" and interaction omission.
Reliability Engineering & System Safety, Jan 1, 2003
We present an approach to efficiently generating an inspection strategy for fault diagnosis. We e... more We present an approach to efficiently generating an inspection strategy for fault diagnosis. We extend the traditional troubleshooting framework to model nonperfect repair actions, and we include questions. Questions are troubleshooting steps that do not aim at repairing the device, but merely are performed to capture information about the failed equipment, and thereby ease the identification and repair of the fault. We show how Vesely and Fussell's measure of component importance extends to this situation, and focus on its applicability to compare troubleshooting steps. We give an approximate algorithm for generating a "good" troubleshooting strategy in cases where the assumptions underlying Vesely and Fussell's component importance are violated, and discuss how to incorporate questions into this troubleshooting strategy. Finally, we utilize certain properties of the domain to propose a fast calculation scheme.
AAAI Workshop on Case-Based Reasoning …, Jan 1, 1998
In this paper we propose an approach to knowledge intensive CBR, where explanations are generated... more In this paper we propose an approach to knowledge intensive CBR, where explanations are generated from a domain model consisting partly of a semantic network and partly of a Bayesian network (BN). The BN enables learning within this domain model based on the observed data. The domain model is used to focus the retrieval and reuse of past cases, as well as the indexing when learning a new case. Essentially, the BN-powered submodel works in parallel with the semantic network model to generate a statistically sound contribution to case indexing, retrieval and explanation.
Journal of statistical planning and …, Jan 1, 2006
We consider the competing risks problem for a repairable unit which at each sojourn may be subjec... more We consider the competing risks problem for a repairable unit which at each sojourn may be subject to either a critical failure, or a preventive maintenance (PM) action, where the latter will prevent the failure. It is reasonable to expect a dependence between the failure mechanism and the PM regime. The paper presents a new model, called the repair alert model, for handling such cases. This model is a special case of random signs censoring, which was introduced by Roger Cooke [Statistics and Probability Letters, 18:307-312, 1993]. The pleasant feature of random signs censoring is that the marginal distribution of the failure time is identifiable. The repair alert model introduces the so called repair alert function, which characterizes the "alertness" of the maintenance crew, and which is shown to be uniquely identifiable from field data. Statistical estimation is considered both nonparametrically and parametrically.
Abstract The overall goal of this research is to study reasoning under uncertainty by combining B... more Abstract The overall goal of this research is to study reasoning under uncertainty by combining Bayesian Networks and Case-Based Reasoning through constructing an experimental decision support system for classification of cancer pain. We have experimentally analysed a medical dataset in order to reveal properties of the data with respect to properties of the two reasoning methods.
In this paper we propose a framework, called mixtures of truncated basis functions (MoTBFs), for ... more In this paper we propose a framework, called mixtures of truncated basis functions (MoTBFs), for representing general hybrid Bayesian networks. The proposed framework generalizes both the mixture of truncated exponentials (MTEs) framework and the Mixture of Polynomials (MoPs) framework.
Abstract In this paper we will present the basic properties of Bayesian network models, and discu... more Abstract In this paper we will present the basic properties of Bayesian network models, and discuss why this modelling framework is well suited for an application in collaborative filtering. We will then describe a new collaborative filtering model, which is built using a Bayesian network. By examining how the model operates on the well-known MovieLens-dataset, we can inspect its merits both qualitatively and quantitatively.
The second symposium of the Norwegian AI Society (NAIS) was held in Gjøvik November 22nd, 2010. N... more The second symposium of the Norwegian AI Society (NAIS) was held in Gjøvik November 22nd, 2010. NAIS was established in the early 1980s, but has been dormant for approximately a decade. This symposium is the second after NAIS has been reinitiated. The symposium functions as a scientific meeting-place for practitioners and theoreticians working in artificial intelligence in Norway.
When constructing a Bayesian network, it can be advantageous to employ structural learning algori... more When constructing a Bayesian network, it can be advantageous to employ structural learning algorithms to combine knowledge captured in databases with prior information provided by domain experts. Unfortunately, conventional learning algorithms do not easily incorporate prior information, if this information is too vague to be encoded as properties that are local to families of variables. For instance, conventional algorithms do not exploit prior information about repetitive structures, which are often found in object oriented domains such as computer networks, large pedigrees and genetic analysis.
Mathematical and statistical methods …, Jan 1, 2003
We investigate the mathematical modelling of maintenance and repair of components that can fail d... more We investigate the mathematical modelling of maintenance and repair of components that can fail due to a variety of failure mechanisms. Our motivation is to build a model, which can be used to unveil aspects of the quality of the maintenance performed. The model we propose is motivated by imperfect repair models, but extended to model preventive maintenance as one of several "competing risks". This helps us to avoid problems of identifiability previously reported in connection with imperfect repair models. Parameter estimation in the model is based on maximum likelihood calculations. The model is tested using real data from the OREDA database, and the results are compared to results from standard repair models.
Annals of Mathematics and Artificial Intelligence, Jan 1, 2001
This paper describes a method for parameter learning in Object-Oriented Bayesian Networks (OOBNs)... more This paper describes a method for parameter learning in Object-Oriented Bayesian Networks (OOBNs). We propose a methodology for learning parameters in OOBNs, and prove that maintaining the object orientation imposed by the prior model will increase the learning speed in object-oriented domains. We also propose a method to efficiently estimate the probability parameters in domains that are not strictly object oriented. Finally, we attack type uncertainty, a special case of model uncertainty typical to object-oriented domains.
Classification problems have a long history in the machine learning literature. One of the simple... more Classification problems have a long history in the machine learning literature. One of the simplest, and yet most consistently well-performing set of classifiers is the Naïve Bayes models. However, an inherent problem with these classifiers is the assumption that all attributes used to describe an instance are conditionally independent given the class of that instance. When this assumption is violated (which is often the case in practice) it can reduce classification accuracy due to "information double-counting" and interaction omission.
Reliability Engineering & System Safety, Jan 1, 2003
We present an approach to efficiently generating an inspection strategy for fault diagnosis. We e... more We present an approach to efficiently generating an inspection strategy for fault diagnosis. We extend the traditional troubleshooting framework to model nonperfect repair actions, and we include questions. Questions are troubleshooting steps that do not aim at repairing the device, but merely are performed to capture information about the failed equipment, and thereby ease the identification and repair of the fault. We show how Vesely and Fussell's measure of component importance extends to this situation, and focus on its applicability to compare troubleshooting steps. We give an approximate algorithm for generating a "good" troubleshooting strategy in cases where the assumptions underlying Vesely and Fussell's component importance are violated, and discuss how to incorporate questions into this troubleshooting strategy. Finally, we utilize certain properties of the domain to propose a fast calculation scheme.
AAAI Workshop on Case-Based Reasoning …, Jan 1, 1998
In this paper we propose an approach to knowledge intensive CBR, where explanations are generated... more In this paper we propose an approach to knowledge intensive CBR, where explanations are generated from a domain model consisting partly of a semantic network and partly of a Bayesian network (BN). The BN enables learning within this domain model based on the observed data. The domain model is used to focus the retrieval and reuse of past cases, as well as the indexing when learning a new case. Essentially, the BN-powered submodel works in parallel with the semantic network model to generate a statistically sound contribution to case indexing, retrieval and explanation.
Journal of statistical planning and …, Jan 1, 2006
We consider the competing risks problem for a repairable unit which at each sojourn may be subjec... more We consider the competing risks problem for a repairable unit which at each sojourn may be subject to either a critical failure, or a preventive maintenance (PM) action, where the latter will prevent the failure. It is reasonable to expect a dependence between the failure mechanism and the PM regime. The paper presents a new model, called the repair alert model, for handling such cases. This model is a special case of random signs censoring, which was introduced by Roger Cooke [Statistics and Probability Letters, 18:307-312, 1993]. The pleasant feature of random signs censoring is that the marginal distribution of the failure time is identifiable. The repair alert model introduces the so called repair alert function, which characterizes the "alertness" of the maintenance crew, and which is shown to be uniquely identifiable from field data. Statistical estimation is considered both nonparametrically and parametrically.
Uploads
Papers by Helge Langseth