CN110751576B - Passenger journey determining method, device and server - Google Patents
Passenger journey determining method, device and server Download PDFInfo
- Publication number
- CN110751576B CN110751576B CN201911001970.2A CN201911001970A CN110751576B CN 110751576 B CN110751576 B CN 110751576B CN 201911001970 A CN201911001970 A CN 201911001970A CN 110751576 B CN110751576 B CN 110751576B
- Authority
- CN
- China
- Prior art keywords
- travel
- city
- sequence
- passenger
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/14—Travel agencies
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Primary Health Care (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The method comprises the steps of firstly obtaining flight segment data of a target passenger in a first preset time period, then analyzing the obtained flight segment data to determine a travel city sequence of the target passenger, wherein the travel city sequence comprises a plurality of travel cities, the travel cities are arranged according to the sequence of travel time, a pre-trained travel city sequence marked with a destination city is called as a training sample, a travel determination model obtained by a multi-layer neural network is trained, and finally the travel city sequence of the target passenger is input into the travel determination model to determine the target city of the target passenger. The method for determining the travel distance of the passengers can determine the target cities which are possibly traveled by the target passengers in the state that the travel distance is not scheduled to be completed, reduce the range of cities which need to push related information, reduce the data quantity which needs to be processed by a server, and reduce the load of the server.
Description
Technical Field
The invention belongs to the technical field of information processing, and particularly relates to a passenger journey determining method, device and server.
Background
In the civil aviation field, the time period of the whole travel process of the passengers can be divided into three stages of before travel, during travel and after travel. In the two phases of travel and in-transit, a class of passengers whose journey is not scheduled to be completed, specifically passengers who do not schedule all air tickets of the current complete journey.
Related research data show that a significant part of passengers exist in passengers whose travel is not scheduled to be completed, the scheduled time of the air tickets of two adjacent travel sections can be at least one day apart, accordingly, between two scheduled actions of the air tickets, many passengers can form new or temporary travel demands, and in this case, airlines can push related information, such as air ticket information, hotel information and the like of different cities, to the passengers.
However, in the pushing manner in the prior art, only relevant information of which cities are pushed to the passengers is determined according to the historical data, and because the cardinalities of the passengers and the relevant cities are large, when a pushing service is initiated, a large amount of data is processed by the server, and the load of the server is heavy, and even the normal operation of the server is affected.
Therefore, how to determine the subsequent destination of the passenger in the state that the journey is not scheduled to be completed reduces the range of cities in which related information needs to be pushed, reduces the data volume required to be processed by the server, reduces the load of the server, and becomes one of technical problems required to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the present invention aims to provide a method, an apparatus and a server for determining a passenger journey, which determine a subsequent destination of a passenger in a state where the journey is not scheduled to be completed, reduce the range of cities in which related information needs to be pushed, reduce the amount of data that the server needs to process, and reduce the load of the server, and specifically, the present invention provides the following scheme:
in a first aspect, the present invention provides a method for determining a passenger journey, comprising:
acquiring leg data of a target passenger in a first preset time period;
analyzing the navigation segment data, and determining a travel city sequence of the target passenger, wherein the travel city sequence comprises a plurality of travel cities, and the travel cities are arranged according to the sequence of travel time;
invoking a pre-trained journey determination model, wherein the journey determination model is obtained by taking a travel city sequence marked with a destination city as a training sample and training a multi-layer neural network;
and inputting the travel city sequence of the target passenger into the journey determination model to determine the target city of the target passenger.
Optionally, the acquiring the leg data of the target passenger in the first preset time period includes:
acquiring passenger seat reservation record PNR information of a target passenger in a first preset time period;
dividing the PNR information by taking the air range as a unit to obtain air range data of the target passenger in the first preset time range.
Optionally, the analyzing the leg data to determine the travel city sequence of the target passenger includes:
acquiring the flight time in each leg data;
and arranging the air segment data according to the sequence of the flight time, and extracting the travel city information in each air segment data to obtain the travel city sequence of the target passenger.
Optionally, the passenger journey determining method provided in any one of the first aspects of the present invention further includes:
and pushing preset pushing information of the target city to the target passenger.
Optionally, the process of training the journey-determining model includes:
acquiring travel city sequence samples of a plurality of passengers in a second preset time period, wherein the travel city sequence samples take destination cities in the samples as labels;
respectively determining the errors between the output results of the multi-layer neural network to the travel city sequence samples and the corresponding labels to obtain the corresponding errors of the travel city sequence samples;
and adjusting parameters of the multi-layer neural network by taking errors corresponding to the travel city sequence samples as training targets within a preset range to obtain a journey determination model.
Optionally, the determining, respectively, an error between the output result of the multi-layer neural network to each travel city sequence sample and the corresponding label, to obtain the corresponding error of each travel city sequence sample includes:
determining the sequence length of each travel city sequence sample, wherein the sequence length is used for representing the number of travel cities contained in the travel city sequence samples;
dividing all the travel city sequence samples into a plurality of sample sets according to the sequence length of each travel city sequence sample and a preset dividing rule;
and respectively determining errors between output results of the multi-layer neural network on each travel city sequence sample in each sample set and the corresponding label to obtain errors corresponding to each travel city sequence sample in each sample set.
Optionally, the passenger journey determining method provided in the first aspect of the present invention further includes:
counting the iteration times of training the multi-layer neural network;
and if the iteration times are greater than a preset training threshold, taking a model obtained by the last training as a travel determination model.
In a second aspect, the present invention provides a passenger journey determination device comprising:
the acquisition unit is used for acquiring the leg data of the target passenger in the first preset time period;
the first determining unit is used for analyzing the navigation segment data and determining a travel city sequence of the target passenger, wherein the travel city sequence comprises a plurality of travel cities, and the travel cities are arranged according to the sequence of travel time;
the invoking unit is used for invoking a pre-trained journey determination model, wherein the journey determination model is obtained by taking a travel city sequence marked with a destination city as a training sample and training a multi-layer neural network;
and the second determining unit is used for inputting the travel city sequence of the target passenger into the journey determining model and determining the target city of the target passenger.
Optionally, the acquiring unit is configured to, when acquiring leg data in a first preset time period of the target passenger, specifically include:
acquiring passenger seat reservation record PNR information of a target passenger in a first preset time period;
dividing the PNR information by taking the air range as a unit to obtain air range data of the target passenger in the first preset time range.
In a third aspect, the present invention provides a server comprising: a memory and a processor; the memory stores a program adapted to be executed by the processor to implement the passenger trip determination method according to any one of the first aspects of the present invention.
Based on the passenger journey determination method provided by the invention, firstly, the flight segment data of the target passenger in the first preset time period are acquired, then the acquired flight segment data are analyzed to determine the travel city sequence of the target passenger, the travel city sequence comprises a plurality of travel cities, the travel cities are arranged according to the sequence of travel time, the pre-trained travel city sequence marked with the destination city is called as a training sample, the journey determination model obtained by the multi-layer neural network is trained, and finally, the travel city sequence of the target passenger is input into the journey determination model to determine the target city of the target passenger. According to the passenger journey determining method provided by the invention, the journey determining model trained in advance is called by taking the journey city sequence of the target passenger as an input parameter, so that the target city in which the target passenger is likely to travel in a journey undetermined finished state is determined, the range of cities needing to push related information is reduced, the data volume to be processed by a server is reduced, and the load of the server is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for determining a passenger journey according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a multi-layer neural network employed in an embodiment of the present application;
fig. 3 is a block diagram of a passenger journey determining device according to an embodiment of the present invention;
fig. 4 is a block diagram of a server according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, fig. 1 is a flowchart of a passenger trip determining method provided by an embodiment of the present invention, where the method may be applied to an electronic device, and the electronic device may be an electronic device with data processing capability, such as a notebook computer, a smart phone, a PC (personal computer), etc., and obviously, the electronic device may also be implemented by a server on a network side in some cases; referring to fig. 1, a method for determining a passenger journey provided by an embodiment of the present invention may include:
and step S100, acquiring the leg data of the target passenger in the first preset time period.
To determine other cities to which a target passenger is likely to travel for a short time interval between the aforementioned two airline reservation actions, it is first necessary to acquire leg data for a first preset period of time. Specifically, for the selection of the first preset time period, the design requirements of the stroke determination model described later and the data processing capability of the computer or the server can be combined for determination, and of course, the selection of the first preset time period can also be simply selected according to human experience, and the selection of the first preset time period is not limited in the embodiment of the invention. It should be noted that, the method for determining the journey of the passenger provided by the embodiment of the invention can be also suitable for predicting cities to which other passengers who are not in a journey undetermined completion state are likely to go.
Optionally, to obtain the leg data of the target passenger in the first preset time period, all PNR (Passenger Name Record, passenger seat reservation record) information of the target passenger in the first preset time period may be first obtained, and then all PNR information obtained by dividing the obtained PNR information by taking the leg as a unit, so as to obtain all the leg data of the target passenger in the first preset time period. Specifically, the leg data at least includes information such as passenger ID, flight departure date, origin city, arrival city, ticket booking time, etc. According to the information included in the leg data, one leg data corresponds to one historical leg record, and one record represents one complete leg of the target passenger.
And S110, analyzing the leg data and determining a travel city sequence of the target passenger.
As described above, each leg data includes a flight time, so after all the leg data of the target passenger in the first preset time period are obtained, each leg data is analyzed, the flight time in each leg data is obtained, the leg data are arranged according to the sequence of the flight time, and after the arrangement is completed, the travel city information in each piece of leg data is extracted, so that the travel city sequence of the target passenger can be obtained. According to the obtaining process of the travel city sequence, the travel city sequence should include a plurality of travel cities, and each travel city is arranged according to the sequence of travel time.
Specifically, for each piece of the sorted leg data, connecting an arrival city in each leg data with an originating city in the next adjacent leg data, and if the arrival city is the same as the originating city, only one of the arrival city and the originating city is reserved. For example: the two adjacent legs of the target passenger are Beijing-Shanghai and Shanghai-hong Kong, and the corresponding city sequence should include Beijing, shanghai and hong Kong.
Step S120, invoking a pre-trained journey determination model.
After the travel city sequence of the target passenger is obtained, a journey determination model which is obtained through pre-training can be called. The journey determination model is obtained by training a multi-layer neural network by taking a travel city sequence marked with a destination city as a training sample. The training process of the journey determination model will be described below and will not be described in detail here.
And step S130, inputting the travel city sequence of the target passenger into a journey determination model to determine the target city of the target passenger.
After the journey determination model is called, the travel city sequence of the target passenger can be input into the journey determination model, and the output result of the model is the target city to which the target passenger is likely to go.
Optionally, after determining that the target city to which the target passenger is likely to travel is obtained, preset push information related to the target city may be pushed to the target passenger. Such as air ticket information, hotel information, tourist attraction information, etc.
In summary, by the passenger journey determining method provided by the embodiment of the invention, the target city to which the target passenger is likely to go can be determined according to the travel city sequence formed by the target passenger, so that the range of cities needing to push related information is reduced, the data amount needed to be processed by the server is reduced, and the load of the server is reduced.
Furthermore, the method can help an airline company to better judge the follow-up schedule of the travel incomplete passengers, more reasonably recommend related civil aviation products for the passengers, and be helpful for carrying out accurate and personalized recommendation service on the next travel of the passengers. Meanwhile, the method is beneficial to reducing the loss of passengers caused by unnecessary information inquiry and improving the product benefit of airlines and the comfort of the passengers.
Alternatively, referring to fig. 2, fig. 2 is a schematic structural diagram of a multi-layer neural network according to an embodiment of the present invention, and the process of training the journey determination model is described below with reference to the structural block diagram of the multi-layer neural network shown in fig. 2.
Firstly, obtaining travel city sequence samples of a plurality of passengers within a second preset time period. Similar to the travel city sequence described in the embodiment shown in fig. 1, the travel city sequence adopted in the training of the journey determination model also includes a plurality of travel cities, and the travel cities are arranged according to the sequence of travel time, except that the travel city sequence samples use destination cities in the samples as labels, and specifically, for each travel city sequence sample, the destination city is the travel city with the last time obtained by sequencing according to the time sequence.
It should be noted that the second preset time period described in this embodiment may be selected as the same time period as the first preset time period, or may be selected as a different time period.
Optionally, for the travel city sequence sample, the sample data meeting the above requirements after being processed by third party software or equipment may be directly used when the multi-layer neural network is trained to obtain the journey determination model.
Of course, the processing procedure of the sample data may also be integrated into the training method provided by the embodiment of the present invention. Specifically, similar to the above-mentioned urban travel sequence for obtaining the target passengers, PNR information of a plurality of passengers in a second preset time period is collected first, and then all PNR information obtained by dividing the PNR information by taking the air range as a unit is obtained, so that all air range data of all passengers in the second preset time period are obtained. Further, analyzing the obtained navigation segment data, determining a travel city sequence in a second preset time period, and taking a terminal city as a label to obtain a plurality of travel city sequence samples.
Optionally, the travel city time series sample corresponding to any passenger is set as ID n Then the sequence samples can be expressed as < id 1 ,id 2 ,id 3 ,……id n >. The length of each travel city sequence sample is N, which indicates the number of all cities travelled by the passenger in a second preset time period.
And then, respectively determining the output results of the multi-layer neural network on the travel city sequence samples and the errors between the corresponding labels to obtain the corresponding errors of the travel city sequence samples.
Optionally, in order to solve the problem of sparsity of travel records of civil aviation passengers, data of travel city time sequences of passengers selected as samples may be processed, a sequence length of each travel city sequence sample is determined, and all travel city sequence samples are divided into a plurality of sample sets according to the sequence length of each travel city sequence sample and a preset division rule. For example, repartition into Itny based on sequence length 1 、Itny 5 And Itny 10 And three groups of sample sets respectively represent sample sets with sequence lengths of more than 1, more than 5 and more than 10, wherein the sequence lengths are used for representing the number of travel cities contained in the travel city sequence samples.
After a plurality of sample sets are obtained, the output result of the multi-layer neural network on each travel city sequence sample in each sample set can be determined respectively, and the error between the corresponding labels is reached, so that the error corresponding to each travel city sequence sample in each sample set is obtained. And adjusting parameters of the multi-layer neural network by taking errors corresponding to the travel city sequence samples within a preset range as training targets to obtain a journey determination model.
Specifically, a travel city time sequence < id formed by the first N-1 travel cities is input to each training sample with the length of N in the sample set 1 ,id 2 ,id 3 ,……id n-1 The method comprises the steps of inputting the vector into a multi-layer neural network, firstly constructing a corresponding directed graph through a graph construction module according to an input travel city sequence, and simultaneously endowing a vector corresponding to each id in the sequence to serve as an initial representation < x 'of the travel city for the travel city corresponding to the id' 1 ,x' 2 ,x' 3 ……x' n-1 Each travel city in the travel city sequence corresponds to a node in the constructed directed graph, respectively.
Then will be to the resulting cityVector < x' 1 ,x' 2 ,x' 3 ……x' n-1 The > and constructed directed graph is input into a multi-layer neural network, as shown in FIG. 2, where the input < x 1 ,x 2 ,x 3 ,……x n Feature vector for each node, < h 1 ,h 2 ,h 3 ,……x m Is a hidden layer vector, and the output vector is less than x' 1 ,x' 2 ,x' 3 ……x' n > as an embedding obtained after each node passes through the multi-layer neural network. Based on the structural information of the directed graph, a node v is selected, a vector obtained by fusing the city vector of the node v and the city vector of the neighboring neighbor is input into a neural network with the length of 1, and the output result is used as a new representation < x', of the node 1 ,x″ 2 ,x″ 3 ……x″ n-1 ＞。
The sequence of the new representation obtained by each node is < x 1 ,x″ 2 ,x″ 3 ……x″ n-1 The probability distribution of each city selected in the candidate set is finally output through the multi-layer neural network structure < q 1 ,q 2 ,q 3 ,……q M >, and < p with the true label of the sample 1 ,p 2 ,p 3 ,……p M The cross entropy is calculated by > (i.e. the end point city in each travel city sequence sample): h (p, q) = Σ i∈M p i *log(q i ) Wherein p is the true probability distribution of each label city in the candidate set, M is the set of the needed predicted cities, namely all travel city sets, and q represents the probability distribution of each city in the candidate set predicted by the model. The loss function is the average value of cross entropy generated by taking all sample sequences in the current batch sample set as input during batch training, and the loss function is calculated by the following steps:where X and X represent the training set and the corresponding samples therein, respectively.
After the parameters of the multi-layer neural network are adjusted for a plurality of times, the errors meet the preset deviation range requirements, and the training process of the model can be stopped, so that the stroke determination model is obtained.
In some cases, for example, the sample data volume required to be trained is huge, the data processing capacity of the server is limited, or the neural network is still difficult to meet the preset precision requirement after multiple times of training, other methods can be adopted to end the training process.
Optionally, in the model training process, the iteration times of training the multi-layer neural network are counted. If the iteration times are greater than a preset training threshold, the iteration can be stopped, and the model obtained by the last training is used as a travel determination model, so that the training process can be finished within a limited time.
Optionally, after the journey determination model is obtained, to verify the processing capability of the model, other city sequences for travel may be used to test the model.
Alternatively, when the model is tested, a separately prepared test sample may be used, or when a training sample used for training the model is prepared, a part of sample data in the separately prepared test sample may be separately stored as a test sample for subsequent use, which is optional.
During specific test, the first N-1 city sequences < id in any one test sample with the sequence length of N in the test sample set is input 1 ,id 2 ,id 3 ,……id n-1 >, probability of each travel city selected in candidate set < q 1 ,q 2 ,q 3 ,……q M > as the final output of the model. The final output is the probability distribution of the candidate set and the cities in the candidate set to be selected as the next arrival place, the top k cities with the highest probability are taken as the recommended city set, the labels of the test set are compared, whether the cities represented by the labels are in the recommended city set or not is judged, the occurrence is hit, the non-occurrence is miss, and finally the sum is obtained by calculating the number of hits and the ranking of the top k candidate sets of each hit, wherein k represents the size of the recommended city set formed by the top k cities, whereinWherein G is a real label set, rank represents ranking in the corresponding top k-bit prediction set where the label is located, and obviously, the higher the ranking is, the better the model prediction effect is. Finally, k prediction places can be output as recommended objects, namely target cities.
The following describes a passenger trip determining apparatus provided by the embodiment of the present invention, where the passenger trip determining apparatus described below may be regarded as a functional module architecture to be set in a central device for implementing the passenger trip determining method provided by the embodiment of the present invention; the following description may be referred to with respect to the above.
Fig. 3 is a block diagram of a passenger trip determining apparatus according to an embodiment of the present invention, and referring to fig. 3, the apparatus may include:
an acquiring unit 10, configured to acquire leg data of a target passenger within a first preset time period;
the first determining unit 20 is configured to parse the leg data and determine a travel city sequence of the target passenger, where the travel city sequence includes a plurality of travel cities, and each travel city is arranged according to a sequence of travel time;
the invoking unit 30 is configured to invoke a pre-trained trip determination model, where the trip determination model uses a trip city sequence marked with a destination city as a training sample, and trains a multi-layer neural network to obtain;
a second determining unit 40, configured to input the travel city sequence of the target passenger into the journey determination model, and determine a target city of the target passenger.
Optionally, the acquiring unit 10 is configured to, when acquiring the leg data of the target passenger within the first preset time period, specifically include:
acquiring passenger seat reservation record PNR information of a target passenger in a first preset time period;
dividing the PNR information by taking the air range as a unit to obtain air range data of the target passenger in the first preset time range.
Fig. 4 is a block diagram of a server according to an embodiment of the present invention, and referring to fig. 4, may include: at least one processor 100, at least one communication interface 200, at least one memory 300, and at least one communication bus 400;
in the embodiment of the present invention, the number of the processor 100, the communication interface 200, the memory 300 and the communication bus 400 is at least one, and the processor 100, the communication interface 200 and the memory 300 complete the communication with each other through the communication bus 400; it will be apparent that the communication connection schematic shown in the processor 100, the communication interface 200, the memory 300 and the communication bus 400 shown in fig. 4 is only optional;
alternatively, the communication interface 200 may be an interface of a communication module, such as an interface of a GSM module;
the processor 100 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention.
The memory 300, which stores application programs, may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
The processor 100 is specifically configured to execute an application program in the memory to implement any of the embodiments of the passenger trip determination method described above.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method of determining a travel distance of a passenger, comprising:
acquiring flight segment data of a target passenger in a first preset time period, wherein the target passenger is in a state that the journey is not scheduled to be completed;
analyzing the navigation segment data, and determining a travel city sequence of the target passenger, wherein the travel city sequence comprises a plurality of travel cities, and the travel cities are arranged according to the sequence of travel time;
invoking a pre-trained journey determination model, wherein the journey determination model is obtained by taking a travel city sequence marked with a destination city as a training sample and training a multi-layer neural network;
and inputting the travel city sequence of the target passenger into the journey determination model to determine the target city of the target passenger.
2. The method of claim 1, wherein the obtaining leg data for the target passenger for the first predetermined period of time comprises:
acquiring passenger seat reservation record PNR information of a target passenger in a first preset time period;
dividing the PNR information by taking the air range as a unit to obtain air range data of the target passenger in the first preset time range.
3. The method of claim 1, wherein the parsing the leg data to determine the city sequence for travel of the target passenger comprises:
acquiring the flight time in each leg data;
and arranging the air segment data according to the sequence of the flight time, and extracting the travel city information in each air segment data to obtain the travel city sequence of the target passenger.
4. A method of passenger travel determination according to any of claims 1-3, further comprising:
and pushing preset pushing information of the target city to the target passenger.
5. The method of claim 1, wherein training the journey determination model comprises:
acquiring travel city sequence samples of a plurality of passengers in a second preset time period, wherein the travel city sequence samples take destination cities in the samples as labels;
respectively determining the errors between the output results of the multi-layer neural network to the travel city sequence samples and the corresponding labels to obtain the corresponding errors of the travel city sequence samples;
and adjusting parameters of the multi-layer neural network by taking errors corresponding to the travel city sequence samples as training targets within a preset range to obtain a journey determination model.
6. The method according to claim 5, wherein the determining the error between the output result of the multi-layer neural network to each travel city sequence sample and the corresponding label to obtain the corresponding error of each travel city sequence sample includes:
determining the sequence length of each travel city sequence sample, wherein the sequence length is used for representing the number of travel cities contained in the travel city sequence samples;
dividing all the travel city sequence samples into a plurality of sample sets according to the sequence length of each travel city sequence sample and a preset dividing rule;
and respectively determining errors between output results of the multi-layer neural network on each travel city sequence sample in each sample set and the corresponding label to obtain errors corresponding to each travel city sequence sample in each sample set.
7. The passenger travel determination method of claim 5, further comprising:
counting the iteration times of training the multi-layer neural network;
and if the iteration times are greater than a preset training threshold, taking a model obtained by the last training as a travel determination model.
8. A passenger travel determination apparatus, comprising:
the system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is used for acquiring flight segment data of a target passenger in a first preset time period, wherein the target passenger is a passenger in a state that a journey is not scheduled to be completed;
the first determining unit is used for analyzing the navigation segment data and determining a travel city sequence of the target passenger, wherein the travel city sequence comprises a plurality of travel cities, and the travel cities are arranged according to the sequence of travel time;
the invoking unit is used for invoking a pre-trained journey determination model, wherein the journey determination model is obtained by taking a travel city sequence marked with a destination city as a training sample and training a multi-layer neural network;
and the second determining unit is used for inputting the travel city sequence of the target passenger into the journey determining model and determining the target city of the target passenger.
9. The passenger trip determination device according to claim 8, wherein the acquiring unit is configured to, when acquiring leg data of the target passenger within the first preset time period, specifically include:
acquiring passenger seat reservation record PNR information of a target passenger in a first preset time period;
dividing the PNR information by taking the air range as a unit to obtain air range data of the target passenger in the first preset time range.
10. A server, comprising: a memory and a processor; the memory stores a program adapted to be executed by the processor to implement the passenger trip determination method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911001970.2A CN110751576B (en) | 2019-10-21 | 2019-10-21 | Passenger journey determining method, device and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911001970.2A CN110751576B (en) | 2019-10-21 | 2019-10-21 | Passenger journey determining method, device and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110751576A CN110751576A (en) | 2020-02-04 |
| CN110751576B true CN110751576B (en) | 2023-07-25 |
Family
ID=69279132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911001970.2A Active CN110751576B (en) | 2019-10-21 | 2019-10-21 | Passenger journey determining method, device and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110751576B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512996A (en) * | 2015-12-28 | 2016-04-20 | 中国民航信息网络股份有限公司 | Method and system for determining most common place-of-departure |
| CN108446759A (en) * | 2018-02-11 | 2018-08-24 | 陕西师范大学 | Tourist flows amount prediction technique based on seasonal shot and long term memory network model |
| CN109829590A (en) * | 2019-02-14 | 2019-05-31 | 北京无二之旅科技有限公司 | A kind of travel solutions design method and device based on artificial neural network |
| CN110069715A (en) * | 2019-04-29 | 2019-07-30 | 腾讯科技（深圳）有限公司 | A kind of method of information recommendation model training, the method and device of information recommendation |
| KR20190106919A (en) * | 2019-08-30 | 2019-09-18 | 엘지전자 주식회사 | Artificial device and method for controlling the same |
| CN110288121A (en) * | 2019-05-16 | 2019-09-27 | 北京交通大学 | Flight passenger load factor prediction method based on multi-granularity temporal attention mechanism |
| CN110322064A (en) * | 2019-06-27 | 2019-10-11 | 清华大学 | Urban Traffic needing forecasting method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6640098B1 (en) * | 2000-02-14 | 2003-10-28 | Action Engine Corporation | System for obtaining service-related information for local interactive wireless devices |
| WO2013059688A1 (en) * | 2011-10-21 | 2013-04-25 | Getgoing, Inc. | System and method for facilitating the purchase of a travel itinerary subject to destination uncertainty |
| US20170032480A1 (en) * | 2015-08-02 | 2017-02-02 | Chi Him Wong | Personalized travel planning and guidance system |
| CN105512773A (en) * | 2015-12-25 | 2016-04-20 | 中国民航信息网络股份有限公司 | Passenger travel destination prediction method and device |
| CN110163772A (en) * | 2019-04-19 | 2019-08-23 | 杭州电子科技大学 | A kind of tour schedule route generation method based on BiLSTM-CRF |
-
2019
- 2019-10-21 CN CN201911001970.2A patent/CN110751576B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512996A (en) * | 2015-12-28 | 2016-04-20 | 中国民航信息网络股份有限公司 | Method and system for determining most common place-of-departure |
| CN108446759A (en) * | 2018-02-11 | 2018-08-24 | 陕西师范大学 | Tourist flows amount prediction technique based on seasonal shot and long term memory network model |
| CN109829590A (en) * | 2019-02-14 | 2019-05-31 | 北京无二之旅科技有限公司 | A kind of travel solutions design method and device based on artificial neural network |
| CN110069715A (en) * | 2019-04-29 | 2019-07-30 | 腾讯科技（深圳）有限公司 | A kind of method of information recommendation model training, the method and device of information recommendation |
| CN110288121A (en) * | 2019-05-16 | 2019-09-27 | 北京交通大学 | Flight passenger load factor prediction method based on multi-granularity temporal attention mechanism |
| CN110322064A (en) * | 2019-06-27 | 2019-10-11 | 清华大学 | Urban Traffic needing forecasting method |
| KR20190106919A (en) * | 2019-08-30 | 2019-09-18 | 엘지전자 주식회사 | Artificial device and method for controlling the same |
Non-Patent Citations (3)
| Title |
|---|
| Forecasting Air Travel Demand for Selected Destinations Using Machine Learning Methods;Murat Firat 等;Journal of Universal Computer Science;第27卷(第6期);564-581 * |
| 基于深度时空卷积网络的民航需求预测;林友芳 等;北京交通大学学报;第42卷(第2期);1-8 * |
| 民航旅客后续行程推断方法应用;邹威;;电子技术与软件工程(第05期);208-210 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110751576A (en) | 2020-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210174270A1 (en) | Rideshare vehicle demand forecasting device, method for forecasting rideshare vehicle demand, and storage medium | |
| EP3252704B1 (en) | Information providing method and system for on-demand service | |
| CN102867408B (en) | Method for selecting bus trip route | |
| CN104102719B (en) | The method for pushing and device of a kind of trace information | |
| CN107103754A (en) | A kind of road traffic condition Forecasting Methodology and system | |
| CN114372830B (en) | A method for predicting online car-hailing demand based on spatiotemporal multi-graph neural network | |
| CN110334289A (en) | The determination method on trip purpose ground and the determination method of target user | |
| CN112418696B (en) | Construction method and device of urban traffic dynamic knowledge graph | |
| CN104077417A (en) | Figure tag recommendation method and system in social network | |
| JP2014203272A (en) | New branch store starting candidate place analysis device, method, and program | |
| JP2016540319A5 (en) | ||
| CN112651671A (en) | Flight space adjusting method and related equipment | |
| CN114780871A (en) | Method and device for recommending get-off point, computer equipment and storage medium | |
| CN109242579A (en) | Ticket price prediction technique based on convolutional neural networks | |
| CN116046008A (en) | Situation awareness-based route planning method, system and efficiency evaluation device | |
| CN110751395B (en) | Passenger journey state determining method, device and server | |
| CN104834990A (en) | Passenger informatization coding method and device | |
| CN112784177B (en) | A Spatial Distance Adaptive Next Interest Point Recommendation Method | |
| CN110751576B (en) | Passenger journey determining method, device and server | |
| Kieu et al. | Stochastic collective model of public transport passenger arrival process | |
| CN108234596A (en) | Aviation information-pushing method and device | |
| Yassin et al. | Malaysian tourism interest forecasting using Nonlinear Auto-Regressive (NAR) model | |
| CN116503086A (en) | Method, system and medium for data processing of air ticket freight rate based on machine learning | |
| CN116128160A (en) | Method, system, equipment and medium for predicting peak passenger flow of railway station | |
| CN115994787A (en) | Car pooling demand prediction matching method based on neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |