| 摘要 |
<p>PROBLEM TO BE SOLVED: To solve the problem that, in a conventional machine learning device, information of features, not included in machine learning, included in data for classification cannot be classified using information obtained by machine learning.SOLUTION: In a machine learning device 1, correct answer data (p) included in a correct answer data set P are handled as data of a positive example, unknown data in which correct answer data p' included in a correct answer data set P' and incorrect answer data (n) included in an incorrect answer data set N are mixed are handled as data of a negative example, and ranking learning is performed using learning data included in a learning data set in which the number of data of positive examples is less than the number of data of negative examples. Even data of a classification object including features not included in data used for machine learning can be more appropriately classified using information obtained by machine learning by the machine learning device 1 which prepares learning result information by performing predetermined adjustment so as to be closer to a negative example in the case of non-learnt element containing data.</p> |
