Information processing device, information processing method, and information processing program

An information processing method configured to predict a menstruation date of a target woman, the information processing method performed by a computer, the information processing method comprising:

• collecting a plurality of menstruation candidate parameter information associated with a plurality of women, wherein the plurality of women are users of an information system S, wherein the information system S comprises the information processing device, a network, a plurality of user terminals, and a plurality of thermometers associated with respective user terminals and associated with respective women;
• organizing the plurality of menstruation information in a database according to one or more of the following characteristics of the respective women:
• age, season or months when the respective women entered respective data, body temperature change tendency, length of low-temperature phases, length of high temperature phase;
• determining, based on the plurality of menstruation candidate parameter information from the database for women similar to the target woman a parameter combination, the parameter combination including:
• i) a first threshold in units of days for removing outliers based on a positional relationship between a frequency distribution and the first threshold, ii) a second threshold expressed in percentile for winsorizing outliers, and iii) a representative value type (PR),
• modifying a plurality of past menstruation information of the target woman based on the first threshold of the parameter combination and the second threshold of the parameter combination to obtain a plurality of modified past menstrual menstruation information for the target woman,
• determining, using the representative value type (PR) of the parameter combination and the plurality of modified past menstrual information, a prediction parameter P indicating a predicted length of a current menstrual cycle of the target woman, and
• determining, based on the prediction parameter P and a last menstruation date input by the target woman, a predicted menstruation date for the target woman;
• and
• sending the predicted menstruation date to a user terminal of the target woman.

A non-transitory computer readable medium storing thereon an information processing program configured to predict a menstruation date of a target woman, the information processing program causing a computer to:

• collect a plurality of menstruation candidate parameter information associated with a plurality of women, wherein the plurality of women are users of an information system S, wherein the information system S comprises the information processing device, a network, a plurality of user terminals, and a plurality of thermometers associated with respective user terminals and associated with respective women;
• organize the plurality of menstruation information in a database according to one or more of the following characteristics of the respective women:
• age, season or months when the respective women entered respective data, body temperature change tendency, length of low-temperature phases, length of high temperature phase;
• determine, based on the plurality of menstruation candidate parameter information from the database for women similar to the target woman a parameter combination, the parameter combination including:
• i) a first threshold in units of days for removing outliers based on a positional relationship between a frequency distribution and the first threshold, ii) a second threshold expressed in percentile for winsorizing outliers, and iii) a representative value type (PR),
• modify a plurality of past menstruation information of the target woman based on the first threshold of the parameter combination and the second threshold of the parameter combination to obtain a plurality of modified past menstrual menstruation information for the target woman,
• determine, using the representative value type (PR) of the parameter combination and the plurality of modified past menstrual information, a prediction parameter P indicating a predicted length of a current menstrual cycle of the target woman, and
• determine, based on the prediction parameter P and a last menstruation date input by the target woman, a predicted menstruation date for the target woman;
• and
• send the predicted menstruation date to a user terminal of the target woman.