Smart dialing recommendation method, non-transitory computer-readable medium, and mobile device

A non-transitory computer-readable medium, comprising computer program instructions, wherein the computer program instructions, when executed by a processor, cause the processor to perform the following steps:

• accessing a database, and
• establishing the database according to a correspondence between a plurality of different time intervals in a past time range and a plurality of communication numbers by using a machine learning algorithm;
• obtaining a time stamp;
• and
• determining whether the time stamp has expired,
• wherein when it is determined that the time stamp has not expired, according to the correspondence in the database and a current time point,
• a recommended number corresponding to the current time point is retrieved from the plurality of communication numbers;
• and
• when it is determined that the time stamp has expired, the machine learning algorithm is performed again to update the plurality of communication numbers by:
• retrieving call records of a mobile device in a specific time interval, wherein the call records in the specific time interval comprise
• a plurality of phone numbers dialed by using the mobile device in a specific time interval and a plurality of dialing time points;
• predicting communication numbers in each time interval according to a historical data feature;
• updating the plurality of communication numbers in the database;
• and
• resetting the time stamp.

A mobile device, comprising:

• an interface;
• a memory, comprising computer program instructions;
• and
• a processor, coupled to the memory and the interface, and configured to execute the computer program instructions,
• wherein the computer program instructions, when executed by the processor, cause the processor to perform the following steps:
• accessing a database from the memory, and
• establishing the database according to a correspondence between a plurality of different time intervals in a past time range and a plurality of communication numbers by using a machine learning algorithm;
• obtaining a time stamp from the database;
• and
• determining whether the time stamp has expired,
• wherein when it is determined that the time stamp has not expired, according to the correspondence in the database and a current time point,
• a recommended number corresponding to the current time point is retrieved from the plurality of communication numbers;
• and
• when it is determined that the time stamp has expired, the machine learning algorithm is performed again to update the plurality of communication numbers in the database by:
• retrieving call records of a mobile device in a specific time interval, wherein the call records in the specific time interval comprise
• a plurality of phone numbers dialed by using the mobile device in a specific time interval and a plurality of dialing time points;
• predicting communication numbers in each time interval according to a historical data feature;
• updating the plurality of communication numbers in the database;
• and
• resetting the time stamp.