| 摘要 |
With known systems, when performing analysis of data, users typically have a solution-oriented view: there is a problem, and data has been collected with a view to providing some information about the problem. Usually a method is selected, based on vague knowledge about some feature of the problem. This is often sub-optimal, because the user can be assumed to be an expert in the application area, but not in data analysis. Embodiments of the invention provide a means of characterising data analysis methods in terms of tangible and non-tangible parameters such as user-friendliness, accuracy, simplicity, interpretability, speed, cost, comprehensibility, and transparency, and then matching these to some pre-specified requirements. This means is provided by a method of selecting a data analysis method in accordance with a user preference, wherein the user preference relates to a feature of the data analysis method and is represented by a fuzzy set comprising a range of values. The method comprises the following steps: (i) using the user preference to identify one or more rules corresponding to the user preference, each rule comprising at least one fuzzy set that relates features of data analysis methods to data analysis characteristics; (ii) evaluating the or each identified rule, thereby identifying an instance of a data analysis characteristic associated with the identified rule, the instance comprising a fuzzy set having a range of values; (iii) accessing a plurality of data analysis methods, each of which has a plurality of data analysis characteristics, wherein each of which data analysis characteristics is characterised by a fuzzy set having a range of values; and (iv) comparing the data analysis characteristics associated with the accessed data analysis methods with the data analysis characteristic instance in order to identify a data analysis method that matches the user preference.
|
