The present invention provides a method of extracting a document item indicating a trend different from a particular core technology or the number of documents of all patent documents in a particular field of a particular applicant and a time series trend. Kind Code: A1 A specific document set analysis method includes time series document number data of patent document items (independent document items) of a specific document set and patent document items (dependent document items) included in the document set. A data table for patent independent and dependent document item analysis is created with the number of documents time series data, and the time series trend of the number of documents has a gradient other than zero for the section number of analysis target section is 3 or more. Setting a dummy value of the dependent analysis item to be a straight line, a correlation coefficient between the dummy of the dependent document item and the number of documents in the document item in the time series direction, and a time series trend of the three or more documents The gradient of the approximate straight line is calculated, and the dependent document item that matches the extraction condition combining the parameters such as the time series trend gradient ratio of the document item of the adjacent analysis target section is extracted. [Selected figure] Figure 2
展开▼