| 摘要 |
The present invention relates to a process for the automated classification and analysis of data, said process comprising the steps of i) selecting an analytical method for use in an analytical experiment; ii) determining working parameters and conditions for the selected method, which parameters and conditions have an effect on the results obtained by using the selected method; iii) assigning to a set of data, said set of data including a) the type of the selected method and b) a specific combination of working parameters and conditions used in the analytical experiment a unique analytical method code (Hash value); iv) applying the method characterized by said Hash value to at least one reference sample and at least one or more further samples obtained from a process for preparing said sample, in which process at least one parameter such as temperature, solvent, or ratio of reactants etc. is to be optimized; v) storing the Hash value and the name of the sample as metadata together with the analytical results obtained in step (iv) as an analytical report; vi) automatically extracting, e.g. by using an extraction template or another extraction process, the relevant metadata from the analytical reports and transferring them to an enterprise content management system (ECMS) or other database systems; vii) automatically organizing the results according to the Hash value in the ECMS or any other suitable database system, viii) correlating the sets of data characterized by different Hash values on the basis of the reference sample measured for each Hash value, thereby creating a correlation matrix/table; ix) using the correlation matrix/table for multivariate analysis of the one or more of the parameters of the process resulting in the samples recited in step (iv); x) optionally, visualizing the results as one or more graphic charts; xi) using the results of steps (ix) and, optionally, (x) to optimize said one or more parameters of the process as defined in step (iv); xii) optionally, verifying the process on the laboratory, pilot plant or industrial scale. |
