Search

For the selection of stock portfolios, the optimiastion models or expert systems are used separately. To take advantage of both approaches, the integration of optimisation model and expert system is attempted. The proposed architecture-ISPMS (Intelligent Stock Portfolio Management System)-accomplishes the integration by interpreting the knowledge as a part of the formulation of optimisation model. The other generic issues in ISPMS are the integration of personal preference with the expert knowledge and knowledge acquisition by machine learning. This paper thus describes the representation and the inference of an expert knowledge, the representation of personal preference and its integration with the expert knowledge, the interpretation of the knowledge and preference to associate with optimisation model, and the solution algorithm of the modified optimisation model accordingly. [ABSTRACT FROM AUTHOR]

A new method for spot matching in two-dimensional gel electrophoresis images using a cumulative frequency matrix is proposed. The method improves on the weak points of the previous method called 'spot matching by topological patterns of neighbour spots'. It accumulates the frequencies of neighbour spot pairs produced through the entire matching process and determines spot pairs one by one in order of higher frequency. Spot matching by frequencies of neighbour spot pairs shows a fairly better performance. However, it can give researchers a hint for whether the matching results can be trustworthy or not, which can save researchers a lot of effort for verification of the results.