A carregar...
4 resultados
Resultados da pesquisa
A mostrar 1 - 4 de 4
- An agent-based approach to support decisions on electronic marketplacesPublication . Viamonte, Maria João; Ramos, Carlos; Rodrigues, Maria de Fátima Coutinho; Cardoso, José; Ramos, Carlos; Vale, ZitaWith the increasing importance of Electronic Commerce across the Internet the need for software agents to support both customers and suppliers in buying and selling goods/services is growing rapidly. It is becoming increasingly evident that in a few years the Internet will host a large number of interacting software agents. Most of them will be economically motivated, and will negotiate a variety of goods and services. It is therefore important to consider the economic incentives and behaviours of economic software agents, and to use all available means to anticipate their collective interactions. This paper addresses this concern by presenting a Market Simulator designed for analysing agent market strategies based on a complete understanding of buyer and seller behaviours, preference models and pricing algorithms. The system includes agents that are capable of increasing their performance with their own experience, by adapting to the market conditions. The results of the negotiations between agents will be analysed by Data mining tools in order to extract rules that will give the agents feedback to improve their strategies. We will describe the characteristics and technologies involved in the architecture we are specifying and developing.
- Characterization of mv consumers using hierarchical clusteringPublication . Figueiredo, V.; Pinheiro, R.; Carvalho Ramos, Sérgio Filipe; Rodrigues, Maria de Fátima Coutinho; Vale, Zita; Ramos, Carlos; Vale, ZitaWith the electricity market liberalization, distribution and retail companies are looking for better market strategies based on adequate information upon the consumption patterns of its electricity customers. A fair insight on the customers’ behaviour will permit the definition of specific contract aspects based on the different, consumption patterns. In this paper, we propose a Knowledge Discovery in Databases project applied to electricity consumption data from a utility client’s database. To form the different customers’ classes, and find a set of representative consumption patterns, we have used the Two-Step algorithm which is a hierarchical clustering algorithm. Each consumer class will be represented by its load profile resulting from the clustering operation. Next, to characterize each consumer class a classification model will be constructed with the C5.0 classification algorithm.
- An architecture to integrate discovered knowledge in a rule based systemPublication . Oliveira, Paulo; Rodrigues, Maria de Fátima Coutinho; Ramos, Carlos; Vale, ZitaThe techniques and tools of Knowledge Discovery in Databases seek to transform data into knowledge in an “intelligent” and semi-automatic way. One of the possible uses to this discovered knowledge consists in its integration or fusion with the knowledge that is in the knowledge base of an Expert System. It thus complements the knowledge . initially given by the expert, which is not always complete, or the most up-to-date. Using an alternative source it is possible to discover knowledge that is implicit in data, and then proceed with its fusion with the one already in the knowledge base. However, this process can result in errors appearing (for example, inconsistencies) in the knowledge base resulting from the fusion. Thus, one of the requirements to fulfil is the consistency and correction of this new knowledge base. A generic and domain independent architecture that allows a rule based knowledge fusion, in the context above described is presented. Consistency and correction are guaranteed through the detection of errors, and by the adoption of an approach based in maximal consistent subsets of rules.
- Prediction of football match results with Machine LearningPublication . Rodrigues, Fátima; Pinto, ÂngeloFootball is one of the most popular sports in the world, so the perception of the game and the prediction of results is of general interest to fans, coaches, media and gamblers. Although predicting football results is a very complex task, the football betting business has grown over time. The unpredictability of football results and the growing betting business justify the development of prediction models to support gamblers. In this article, we develop machine learning methods that take multiple statistics of previous matches and attributes of players from both teams as inputs to predict the outcome of football matches. Several prediction models were tested, with the experimental results showing encouraging performance in terms of the profit margin of football bets.