Systems and methods for route searching

A non-transitory computer readable medium comprising a set of instructions for searching for a route that, when executed by at least one processor, cause the at least one processor to effectuate a method comprising:

• receiving a service request from a terminal of a service requester or a service provider, wherein the service request includes information relating to a start location and a destination of a first route;
• in response to the service request, obtaining route information of the first route;
• encoding the route information of the first route into a first code based on a target model, wherein:
• the target model is configured to encode a route into a code having a predetermined length of n bytes, n being an integer larger than 1, and encode substantially similar routes into a same code;
• the target model is a recurrent neural network (RNN) model having a hidden layer with at least m neurons;
• and
• to obtain the RNN model, the method further includes:
• setting initial parameters of the RNN model, wherein the initial parameters include a first parameter associated with an input layer of the RNN model and a second parameter associated with the hidden layer of the RNN model;
• obtaining sample route information of a plurality of sample routes, wherein each of the plurality of sample routes includes a plurality of sample waypoints, and the sample route information includes sample location information of the plurality of sample waypoints;
• for each of the plurality of sample routes, determining output information associated with the at least m neurons in the hidden layer based on the initial parameters, a predetermined nonlinear activation function, and the sample route information, wherein for a first neuron in the at least m neurons, an input is sample location information of a first sample waypoint in the plurality of sample waypoints and for an ith neuron in the at least m neurons, the input includes sample location information of an ith waypoint in the plurality of waypoints and an output of an (i−1)th neuron in the at least m neurons;
• and
• determining trained parameters of the RNN model based on the output information;
• accessing a target database in the at least one storage medium, wherein the target database includes a plurality of candidate codes encoded through the target model from a plurality of candidate routes;
• identifying from the plurality of candidate codes, a second code based on the first code, the second code being associated with at least one second route;
• and
• generating on an interface of the terminal a presentation of information associated with the at least one second route.

A system for displaying information of a similar route on art interface of a terminal, comprising:

• at least one storage medium including a set of instructions for searching for a route;
• at least one processor in communication with the at least one storage medium, wherein when executing the set of instructions, the at least one processor is directed to:
• receive a service request from a terminal of a service requestor or a service provider, wherein the service request includes information relating to a start location and a destination of a first route;
• in response to the service request, obtain route information of the first route;
• encode the route information of the first route into a first code based on a target model, wherein:
• the target model is configured to encode a route into a code having a predetermined length of n bytes, n being an integer larger than 1, and encode substantially similar routes into a same code;
• the target model is a recurrent neural network (RNN) model having a hidden layer with at least m neurons;
• and
• to obtain the RNN model, the at least one processor is further directed to:
• set initial parameters of the RNN model, wherein the initial parameters include a first parameter associated with an input layer of the RNN model and a second parameter associated with the hidden layer of the RNN model;
• obtain sample route information of a plurality of sample routes, wherein each of the plurality of sample routes includes a plurality of sample waypoints, and the sample route information includes sample location information of the plurality of sample waypoints;
• for each of the plurality of sample routes, determine output information associated with the at least m neurons in the hidden layer based on the initial parameters, a predetermined nonlinear activation function, and the sample route information, wherein for a first neuron in the at least m neurons, an input is sample location information of a first sample waypoint in the plurality of sample waypoints and for an ith neuron in the at least m neurons, the input includes sample location information of an ith waypoint in the plurality of waypoints and an output of an (i−1)th neuron in the at least m neurons;
• and
• determine trained parameters of the RNN model based on the output information;
• access a target database in the at least one storage medium, wherein the target database includes a plurality of candidate codes encoded through the target model from a plurality of candidate routes;
• identify, from the plurality of candidate codes, a second code based on the first code, the second code being associated with at least one second route;
• and
• generate on an interface of the terminal a presentation of information associated with the at least one second route.