Data processing device, data processing method, and data processing program

A data processing method to be executed by a data processing device, including a computer, the data processing method comprising:

• obtain, for a first cell in a first table, a frequency of appearance of words used in a caption of a row and in a caption of a column wherein the row and the column of the first table intersect with each other at the first cell, each of the caption of the row for the first cell and the caption of the column of first cell include a plurality of words;
• wherein the data processing method sorts the words in the caption of the row and the caption of the column at which the row and the column of the table intersect with each other at the first cell, and deletes one or more overlapping words among the sorted words before obtaining the frequency of appearance of words for the first cell;
• generate a first vector for the first cell that includes the obtained frequency of appearance of words used in the caption of a row and in the caption of the column as a value of an element, and
• store the first vector for the first cell and match content corresponding to the first cell as training data;
• generate a second vector for a second cell of a second table wherein the second vector includes a frequency of appearance of words in a caption of a row and in a caption of a column for the second cell at which the row and the column of the second table intersect with each other at the second cell;
• predict match content corresponding to the second cell based on a result of comparing the first vector, the second vector, and the stored match content;
• display the predicted match content corresponding to the second cell based on the result.

A non-transitory recording medium storing a data processing program, wherein execution of the data processing program causes a computer to perform operations comprising:

• obtaining, for a first cell in a first table, a frequency of appearance of words used in a caption of a row and in a caption of a column wherein the row and the column of the first table intersect with each other at the first cell, each of the caption of the row for the first cell and the caption of the column of first cell include a plurality of words;
• wherein the data processing program sorts the words in the caption of the row and the caption of the column at which the row and the column of the table intersect with each other at the first cell, and deletes one or more overlapping words among the sorted words before obtaining the frequency of appearance of words for the first cell;
• generating a first vector for the first cell that includes the obtained frequency of appearance of words used in the caption of a row and in the caption of the column as a value of an element, and
• store the first vector for the first cell and match content corresponding to the first cell as training data;
• generating a second vector for a second cell of a second table wherein the second vector includes a frequency of appearance of words in a caption of a row and in a caption of a column for the second cell at which the row and the column of the second table intersect with each other at the second cell;
• predicting match content corresponding to the second cell based on a result of comparing the first vector, the second vector, and the stored match content;
• display the predicted match content corresponding to the second cell based on the result.