No. 3 (2002)
Published:
2002-09-30
ARTICLES FROM THIS ISSUE
-
The concept of ambient intelligence and decision support for telecommunications
Abstract
The paper presents first the concept of AmI (ambient intelligence) and AmI scenarios of ISTAG (Information Society Technology Advisory Group of European Commission). The requirements of intelligence versus decision support are then discussed. Resulting challenges for decision support systems (DSS) in telecommunications are then outlined and conclusions presented.
-
Rough set theory and its applications
Abstract
In this paper rudiments of the theory will be outlined, and basic concepts of the theory will be illustrated by a simple tutorial example, concerning churn modeling in telecommunications. Real life applications require more advanced extensions of the theory but we will not discuss these extensions here. Rough set theory has an overlap with many other theories dealing with imperfect knowledge, e.g., evidence theory, fuzzy sets, Bayesian inference and others. Nevertheless, the theory can be regarded as an independent, complementary, not competing, discipline in its own rights.
-
Using reporting and data mining techniques to improve knowledge of subscribers; applications to customer profiling and fraud management
Abstract
One of the main issues of operators today is to be able not only to store and manage the huge amount of data generated by the applications and customer contact points, but also to give value to these data. But this implies using tools for storing the data, to manage it, look at it, understand it, exploit it, generate actions such as marketing campaigns. It is therefore obvious that using one tool for each of these functions will lead to a too big and unusable solution. We will discuss here the technical issues involved and show how we turned them into an easy-to-use solution for business users.
-
Analytical properties of a stochastic teletraffic system with MMPP input and an access function
Abstract
Stochastic modeling of teletraffic systems with restricted availability and correlated input arrival rates is of great interest in GoS (grade of service) analysis and design of certain telecommunication networks. This paper presents some analytical properties of a recursive nature, associated with the infinitesimal generator of the Markov process which describes the state of a teletraffic system with MMPP (Markov modulated Poisson process) input traffic, negative exponentially distributed service times, finite queue and restricted availability defined through a loss function. Also the possible application of the derived properties to a direct method of resolution of the linear system, which gives the stationary probability distribution of the system, will be discussed.
-
Effectiveness of active forgetting in machine learning applied to financial problems
Abstract
One of main features in financial investment problems is that the situation changes very often over time. Under this circumstance, in particular, it has been observed that additional learning plays an effective role. However, since the rule for classification becomes more and more complex with only additional learning, some appropriate forgetting is also necessary. It seems natural that many data are forgotten as the time elapses. On the other hand, it is expected more effective to forget unnecessary data actively. In this paper, several methods for active forgetting are suggested. The effectiveness of active forgetting is shown by examples in stock portfolio problems.
-
A study on fractal dimensions and convergence in fuzzy control systems
Abstract
This paper addresses a problem in the area of intelligent, knowledge-based systems, namely the generation of knowledge, by presenting a proposal for the automation of this task. The proposed approach is limited however by focusing on fuzzy control systems (FCSs). Results obtained from different experimental investigations indicate the potential of the approach.
-
Multi-agent utility theory for ethical conflict resolution
Abstract
In this paper we try to construct a two-attribute group disutility function for two conflicting decision makers, taking into account the property of utility independence and/or convex dependence between them. Two variables in the group utility function are disutility levels of two conflicting decision makers. The disutility level of each decision maker is modeled using multiple attributes, that is disutility function of each decision maker is formulated as a multi-attribute disutility function. By using a~group disutility function for two conflicting agents, we can model the mutual concessions of the two conflicting agents taking into account ethical preference of each decision maker, and hence we can expect fairer multiple agents decision making for realizing better social welfare.
-
On equitable approaches to resource allocation problems: the conditional minimax solutions
Abstract
In this paper we introduce and analyze a solution concept of the conditional minimax as a generalization of the minimax solution concept extended to take into account the number of services (the portion of demand) related to the worst performances. Namely, for a specified portion of demand we take into account the corresponding portion of the maximum results and we consider their average as the worst conditional mean to be minimized. We show that, similar to the standard minimax approach, the minimization of the worst conditional mean can be defined by a linear objective and a number of auxiliary linear inequalities. We report some results of initial computational experience with the new solution concept.
-
A proposition to exploit the partially linear structure of the nonlinear multicommodity flow optimization problem
Abstract
Optimization problems arising in telecommunications are often large-scale nonlinear problems. Usually their big size is generated mainly by their linear parts but the existence of small or medium nonlinear parts prevents us from directly tackling them with linear solvers, which are efficient. Instead, the author has proposed a method to decompose big nonlinear problems into nonlinear and linear parts. Its coordination procedure uses two auxiliary solvers: quadratic and pure nonlinear. The procedure falls in the class of projection methods. Special cuts proposed by the author allow to avoid an excessive zigzagging while not enormously increasing the complexity of both the parts. The validity of these cuts can be analyzed within the framework of obtuse cone model. Here the author summarizes the method and analyses its applicability to nonlinear multicommodity flow problems. The structure and particular sizes of this problem make the method useful. The considerations are illustrated by a numerical example with a multicommodity flow problem.
-
Decision support tool for web cache management
Abstract
Web caching is the subject of intense research and development since it seems to be very promising area. Web caching means storing copies of frequently used objects (documents) geographically close to users requesting them to reduce network load, servers load and user response times. Cache can be situated in different locations between user and servers with original content. It seems that the most significant improvement can be achieved by using the proxy server -- a dedicated web server. Various parameters affect web cache performance: cache size, limitations on document sizes or documents removal policy. Several metrics are used to evaluate this performance: hit rate for example is the ratio of documents obtained by using the cache mechanism comparing to the total number of documents requested. Choosing adequate parameters for interesting traffic patterns to improve cache performance is not a trivial cache management problem. Prototype tool based on multicriteria model analysis and supporting such web cache management is presented. Simple case was examined where small number of sets of cache parameters (variants), miniature traffic representation and only two performance measures are considered.
-
Applications of commercial data analysis software for testing a telecommunications network
Abstract
This paper shows practical aspect in application commercial software to analysis data concerning information from telecommunication network. The following items are discussed: example of network estimation methodology, the main advantages of network testing system delivering information concerning quality of service (QoS) and network performance (NP) from the user`s perspective, general testing system description and data analysis model and example forms of presentation.
