Method and apparatus for data analysis

1. A method of selecting a data analysis method from a plurality of such data analysis methods by applying fuzzy logic to characteristics of those data analysis methods to select the data analysis method from the plurality that most closely matches a fuzzy set of user-defined preferences for such characteristics, said method of selecting comprising:

(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 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) retrieving data identifying a plurality of data analysis methods, each of which has a plurality of data analysis characteristics, wherein, in respect of each said data analysis characteristic, the retrieved data includes a range of values;

(iv) comparing the retrieved data with the data analysis characteristic instance in order to identify a data analysis method that matches the user preference; and

(v) outputting data identifying the data analysis method that matches the user preference.

2. A method as in claim 1 , in which, when a plurality of rules is identified at step (i), the evaluating step (ii) comprises identifying occurrences of each data analysis characteristic, so that, if there is more than one occurrence of any data analysis characteristic, the evaluating step (ii) includes combining instances corresponding thereto so as to generate a combined instance of that data analysis characteristic.

3. A method as in claim 1 , in which the comparing step (iv) comprises correlating the instance identified at step (ii) with the retrieved data.

4. A method as in claim 1 , including the step of specifying a type of data analysis method and performing the said selection of a data analysis method in accordance with the specified type.

5. A method as in claim 4 , including removing retrieved data in respect of any data analysis methods that are not of the specified type.

6. A method as in claim 1 including:

for one or more data analysis methods, modifying the range of values comprising the fuzzy sets corresponding to at least one of the data analysis characteristics in such a manner as to increase the number of data analysis methods that match the user preference.

7. A method as in claim 1 , including:

(i) ranking the identified methods in accordance with their degree of match with the user preference;

(ii) receiving an indicator representative of which of the identified data analysis methods best matches user requirements; and

(iii) modifying the fuzzy set corresponding to the user preference in accordance with at least one aspect of the indicated data analysis method.

8. A method as in claim 7 , in which the step of modifying the fuzzy set includes:

comparing the range of values of the user preference with the range of values in the characteristic of the best match data analysis method in order to evaluate an error value; and

using the error value to modify the range of values in the user preference.

9. A method as in claim 1 , including:

receiving data for use in creating a model corresponding to the identified data analysis method, the model comprising one or more model parameters;

inputting the data to a learning process associated with the model;

invoking the learning process, thereby modifying at least one of the model parameters, and monitoring the same;

comparing each monitored model parameter with a control rule comprising at least one fuzzy set relating to one or more model parameters, so as to identify whether the control rule should be applied, and if so, evaluating the control rule; and

modifying the learning process in accordance with the evaluated rule.

10. A tangible digital storage medium containing a computer program, or a suite of computer programs, comprising a set of instructions to cause a computer, or a suite of computers, to perform the method as in claim 1 .

11. Server apparatus for 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, wherein the server apparatus has access to a first storage device for storing one or more rules, each rule comprising at least one fuzzy set that relates features of data analysis methods to data analysis characteristics, and a second storage device for storing a plurality of data analysis methods, each having a plurality of data analysis characteristics, each of which data analysis characteristics comprising a fuzzy set having a range of values, the server apparatus comprising:

(i) identifying means for identifying a rule, from the first storage device, corresponding to the user preference;

(ii) evaluating means for evaluating the 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) comparing means for comparing data analysis characteristics associated with the data analysis methods stored in the second storage device with the identified data analysis characteristic instance in order to identify a data analysis method that matches the user preference; and

(iv) means for outputting data identifying the data analysis method that matches the user preference.

12. Client apparatus for selecting a data analysis method in accordance with a user preference, including: receiving means for receiving a user preference specifying a feature of the data analysis method, the user preference being represented by a fuzzy set comprising a range of values, wherein the client apparatus co-operates with server apparatus having access to a first storage device for storing one or more rules: each rule comprising at least one fuzzy set that relates features of data analysis methods to data analysis characteristics, and a second storage device for storing a plurality of data analysis methods, each of which has a plurality of data analysis characteristics, each of which data analysis characteristics comprising a fuzzy set having a range of values, the server apparatus comprising:

(i) identifying means for identifying a rule, from the first storage device, corresponding to the user preference;

(ii) evaluating means for evaluating the 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) comparing means for comparing data analysis characteristics associated with the data analysis methods stored in the second storage device with the identified data analysis characteristic instance in order to identify a data analysis method that matches the user preference; and

means for outputting data identifying the identified data analysis method to the client, and wherein the client apparatus is arranged to display said output data.