Methods for recommending parking lots in smart cities, internet of things systems, and storage medium thereof

An Internet of Things system for recommending a parking lot in a smart city, including a management platform, wherein the management platform is configured to:

• determine an arrival duration of a user reaching at least one candidate parking lot and a prediction time period corresponding to each candidate parking lot in at least one candidate parking lot based on a user request;
• determine an occupation rate of the candidate parking lot during the prediction time period;
• and
• determine a recommended parking lot at least based on the occupation rate of the candidate parking lot during the prediction time period and parking lot information of the candidate parking lot;
• wherein the parking lot information further includes at least one of a historical occupation rate of the parking lot, a parking lot capacity, and surrounding event holding information;
• and
• the management platform is further configured to:
• construct a parking lot feature vector corresponding to the candidate parking lot based on the parking lot capacity of the candidate parking lot, an internal structure of the parking lot, and surrounding event holding information during the prediction time period;
• obtain a reference historical vector through retrieving in a vector database based on the parking lot feature vector;
• and
• use the historical occupation rate of the parking lot corresponding to the reference historical vector as the occupation rate of the candidate parking lot during the prediction time period.

A non-transitory computer readable storage medium, wherein the storage medium stores computer instructions, and when a computer reads the computer instructions, the computer executes a method for recommending a parking lot in a smart city, including:

• determining an arrival duration of a user reaching at least one candidate parking lot and a prediction time period corresponding to each candidate parking lot in the at least one candidate parking lot based on a user request;
• determining an occupation rate of the candidate parking lot during the prediction time period;
• and
• determining a recommended parking lot at least based on the occupation rate of the candidate parking lot during the prediction time period and parking lot information of the candidate parking lot;
• wherein the parking lot information further includes at least one of a historical occupation rate of the parking lot, a parking lot capacity, and surrounding event holding information;
• and
• the determining the occupation rate of the candidate parking lot during the prediction time period includes:
• constructing a parking lot feature vector corresponding to the candidate parking lot based on the parking lot capacity of the candidate parking lot, an internal structure of the parking lot, and the surrounding event holding information during the prediction time period;
• obtaining a reference historical vector through retrieving in a vector database based on the parking lot feature vector;
• and
• using the historical occupation rate of the parking lot corresponding to the reference historical vector as the occupation rate of the candidate parking lot during the prediction time period.