CN111861540A - Information pushing method and device, computer equipment and storage medium - Google Patents
Information pushing method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN111861540A CN111861540A CN202010516358.5A CN202010516358A CN111861540A CN 111861540 A CN111861540 A CN 111861540A CN 202010516358 A CN202010516358 A CN 202010516358A CN 111861540 A CN111861540 A CN 111861540A
- Authority
- CN
- China
- Prior art keywords
- user
- information
- characteristic information
- electronic
- shared vehicle
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000012549 training Methods 0.000 claims description 58
- 230000006399 behavior Effects 0.000 claims description 33
- 238000004590 computer program Methods 0.000 claims description 31
- 238000013135 deep learning Methods 0.000 claims description 22
- 238000003066 decision tree Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 7
- 238000010586 diagram Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000001960 triggered effect Effects 0.000 description 5
- 238000007418 data mining Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 241001122767 Theaceae Species 0.000 description 1
- 235000021152 breakfast Nutrition 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000013410 fast food Nutrition 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0261—Targeted advertisements based on user location
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to an information pushing method, an information pushing device, computer equipment and a storage medium. According to the method, the locking information of the shared vehicle is received, the position information of the shared vehicle and the user characteristic information corresponding to the current locked user account are obtained according to the locking information, the information to be pushed in a set range is obtained according to the position information of the shared vehicle, the use probabilities of a plurality of electronic certificates in the information to be pushed are predicted according to the user characteristic information corresponding to the current locked user account, and the electronic certificates with the target number of higher use probabilities are pushed to the current locked user account.
Description
Technical Field
The present application relates to the field of information processing technologies, and in particular, to an information pushing method and apparatus, a computer device, and a storage medium.
Background
With the development of the sharing economy, sharing vehicles are distributed all over the country and become an important travel mode, but at present, the use mode is single, and the sharing vehicles can only be used for pure transportation instead of providing more extended services.
Since the shared vehicle reaches millions or even tens of millions, the shared vehicle is gradually used for information dissemination, and some merchants try to push their products in a form of posting advertisements on the shared vehicle, but because the pushing is random and the position of the shared vehicle changes continuously, it is impossible to know which users are interested in the pushed products and which users are not interested, so that the pushing success rate is not high. And the form of advertisement posting can also affect the appearance of the market. Therefore, how to fully utilize the sharing bicycle to accurately push information to the user becomes a problem to be solved urgently at present.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a method, an apparatus, a computer device, and a storage medium for pushing information with full advantage of a shared bicycle, in order to solve the above-mentioned problem that the success rate of pushing by a form of posting an advertisement on the shared bicycle is not high.
An information pushing method, the method comprising:
receiving locking information of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked according to the locking information, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
Acquiring information to be pushed in a set range according to the position information of the shared vehicle, wherein the information to be pushed comprises a plurality of electronic certificates which are matched with the current time and used for exchanging resources;
predicting the use probabilities of a plurality of electronic certificates according to user characteristic information corresponding to the current locked user account;
and pushing the electronic certificates with the target number and higher use probability to the current locked user account.
In one embodiment, the use probability prediction model of deep learning is adopted, and the use probabilities of a plurality of electronic certificates are predicted according to user characteristic information corresponding to the user account currently locked, wherein the generation method of the use probability prediction model comprises the following steps: the method comprises the steps of obtaining a training data set, wherein the training data set is obtained by pushing electronic certificates to a plurality of sample user accounts using a shared vehicle, the training data set comprises user characteristic information corresponding to each sample user account, characteristic information of the electronic certificates pushed to the sample user accounts and mark information of whether the electronic certificates are used or not, and the user characteristic information comprises user basic information and user behavior characteristic information; and training the gradient boosting iterative decision tree network through a training data set to obtain a use probability prediction model.
In one embodiment, acquiring information to be pushed within a set range according to the position information of the shared vehicle includes: acquiring a plurality of electronic certificates and characteristic information of each electronic certificate within a set range by calling a data transmission interface, wherein the set range takes position information of a shared vehicle as a center, the characteristic information of each electronic certificate comprises the type of the electronic certificate, attribute information of resources corresponding to the electronic certificates and the distance between the using place of the electronic certificate and the position of the shared vehicle, and determining the electronic certificates as information to be pushed; the method for predicting the use probabilities of the electronic certificates by adopting the deep learning use probability prediction model according to the user characteristic information corresponding to the user account currently locked comprises the following steps: preprocessing user characteristic information corresponding to a user account currently locked and characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate; and respectively inputting the data to be predicted corresponding to each electronic certificate into a deep learning use probability prediction model to respectively obtain the predicted use probability of each electronic certificate.
In one embodiment, the pre-processing the user characteristic information corresponding to the currently locked user account and the characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate includes: and splicing the user characteristic information corresponding to the user account currently locked and the characteristic information of each electronic certificate according to a set data splicing rule to obtain spliced data to be predicted corresponding to each electronic certificate.
In one embodiment, after the target number of electronic certificates with higher usage probability are pushed to the currently locked user account, the method further includes: obtaining the use result of the electronic certificates of the target number pushed by the user account; and training the use probability prediction model according to the user characteristic information corresponding to the user account, the characteristic information of each pushed electronic certificate and the corresponding use result to obtain the re-trained use probability prediction model.
In one embodiment, the pushing of the target number of electronic certificates with higher usage probability to the currently locked user account includes: ordering the plurality of electronic certificates according to the predicted using probability of the plurality of electronic certificates; and determining the electronic certificates with the target number ranked at the top in the ranking result as the electronic certificates with higher similarity.
An information pushing method, the method comprising:
responding to a locking triggering instruction of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
sending the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account to the server to indicate the server to acquire the electronic certificates with the target number and high use probability according to the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account;
And receiving the target number of electronic certificates with higher use probability returned by the server.
In one embodiment, the method further comprises: and responding to the use triggering instruction of the electronic certificate, and sending a use result of the electronic certificate to the server.
An information push apparatus, the apparatus comprising:
the locking information receiving module is used for receiving locking information of the shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked according to the locking information, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
the system comprises an information to be pushed acquisition module, a resource exchange module and a resource management module, wherein the information to be pushed acquisition module is used for acquiring information to be pushed in a set range according to position information of a shared vehicle, and the information to be pushed comprises a plurality of electronic certificates which are matched with the current time and used for exchanging resources;
the usage probability prediction module is used for predicting the usage probabilities of the electronic certificates according to the user characteristic information corresponding to the user account currently locked;
and the pushing module is used for pushing the electronic certificates with the target number and higher use probability to the current locked user account.
In one embodiment, the using a probabilistic prediction module is implemented by a deep-learning using a probabilistic prediction model, the using the probabilistic prediction model comprising: the training data set acquisition unit is used for acquiring a training data set, wherein the training data set is obtained by pushing electronic certificates to a plurality of sample user accounts using a shared vehicle, the training data set comprises user characteristic information corresponding to each sample user account, characteristic information of the electronic certificates pushed to the sample user accounts and marking information of whether the electronic certificates are used, and the user characteristic information comprises user basic information and user behavior characteristic information; and the model generation unit is used for training the gradient lifting iterative decision tree network through a training data set to obtain the use probability prediction model.
An information push apparatus, the apparatus comprising:
the instruction response module is used for responding to a locking triggering instruction of the shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
the sending module is used for sending the position information of the shared vehicle and the user characteristic information corresponding to the user account locked currently to the server so as to indicate the server to obtain the electronic certificates with the target number and high use probability according to the position information of the shared vehicle and the user characteristic information corresponding to the user account locked currently;
and the receiving module is used for receiving the electronic certificates with the target number and higher using probability returned by the server.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method as described above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method as set forth above.
According to the information pushing method, the information pushing device, the computer equipment and the storage medium, the locking information of the shared vehicle is received, the position information of the shared vehicle and the user characteristic information corresponding to the current locked user account are obtained according to the locking information, the information to be pushed in the set range is obtained according to the position information of the shared vehicle, the use probability of a plurality of electronic certificates in the information to be pushed is predicted according to the user characteristic information corresponding to the current locked user account, and the electronic certificates with the target number and high use probability are pushed to the current locked user account.
Drawings
FIG. 1 is a diagram of an exemplary information push method;
FIG. 2 is a flowchart illustrating an information pushing method according to an embodiment;
FIG. 3 is a schematic flow chart diagram illustrating the model generation step in one embodiment;
FIG. 4 is a schematic flow chart of the step of using probabilities by model prediction in one embodiment;
FIG. 5 is a schematic flow chart diagram illustrating the model optimization step in one embodiment;
FIG. 6 is a flowchart illustrating an information pushing method according to an embodiment;
Fig. 7 is a diagram illustrating a specific application of the information push method in an embodiment;
FIG. 8 is a block diagram of an information pushing apparatus according to an embodiment;
FIG. 9 is a block diagram showing the construction of an information pushing apparatus according to still another embodiment;
FIG. 10 is a diagram showing an internal structure of a computer device in one embodiment;
fig. 11 is an internal structural view of a computer device in still another embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The information pushing method provided by the application can be applied to the application environment shown in fig. 1. The terminal 102 and the vehicle lock device 104 on the shared vehicle communicate with the server 106 through the network, respectively. Specifically, the terminal 102 may be, but is not limited to, various smart phones, tablet computers, and portable wearable devices, an application program that can obtain the shared vehicle permission is installed in the terminal 102, the server 106 may be implemented by an independent server or a server cluster formed by a plurality of servers, the vehicle locking device 104 is a lock with communication and positioning functions, it can locate its own location when the status (i.e., unlocked or locked) changes, and communicate with the server 106 to report status and location information, meanwhile, the server 106 is also a background server of the application program installed in the terminal 102, the instructions of the application are verified by the server 106 and the lock device 104 on the shared vehicle is controlled, so that the user can obtain the usage right of the shared vehicle through the application installed on the terminal 102.
In this embodiment, when the server 106 receives the locking information of the shared vehicle, the location information of the shared vehicle and the user feature information corresponding to the user account currently locked are obtained according to the locking information, and the information to be pushed in the set range is obtained according to the location information of the shared vehicle, where the information to be pushed includes a plurality of electronic certificates used for exchanging resources and matched with the current time, and then the usage probabilities of the plurality of electronic certificates are predicted according to the user feature information corresponding to the user account currently locked, so that the electronic certificates with the target number of higher usage probabilities are pushed to the user account currently locked, thereby implementing accurate pushing based on data mining and machine learning of the shared vehicle, not only improving the success rate of information pushing, but also avoiding the problem that advertisement is posted on the shared vehicle to affect the market.
In one embodiment, as shown in fig. 2, an information pushing method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:
The locking information of the shared vehicle refers to information that the locking device is triggered to lock after the travel of the shared vehicle is finished. The location information of the shared vehicle refers to a location of the shared vehicle that is currently triggered to be in an off-lock state, and may be, for example, longitude and latitude information of the shared vehicle obtained through GPS (Global Positioning System) Positioning, or specific street information obtained by converting the longitude and latitude information. The current locked user account refers to the user account of the user who obtains the use right of the shared vehicle and uses the shared vehicle in the corresponding application program. After the shared vehicle is used, the user can change the state of the shared vehicle through a lock device on the shared vehicle, namely, the unlocking state is adjusted to the locking state, so that the journey is finished. Because the state of the shared vehicle is communicated with the server and the state and the position information of the shared vehicle are reported when the state of the shared vehicle changes, when the server receives the locking information of the shared vehicle, the position information of the shared vehicle can be acquired, meanwhile, in order to accurately push the information to the user, the server can also acquire user characteristic information corresponding to the currently locked user account, specifically, the user characteristic information includes user basic information and user behavior characteristic information, for example, the user basic information includes but is not limited to the age, the gender, the occupation and the like of the user, and the user behavior characteristic information includes but is not limited to the historical riding path of the user and the record of the historical use of the electronic certificate of the user.
And 204, acquiring the information to be pushed in the set range according to the position information of the shared vehicle.
The set range may be a set distance range, specifically, a distance range with respect to the position of the shared vehicle, and the distance range may be set according to actual needs, for example, the distance range may be within 1 km, within 3 km, or within 5 km near the position of the shared vehicle. The information to be pushed refers to information to be pushed to the user, and specifically, the information to be pushed includes a plurality of electronic certificates used for exchanging resources, which are matched with the current time. It can be understood that the electronic certificate generally has timeliness and regionality, and the timeliness is represented by that the electronic certificate has a corresponding validity period, and the matching with the current time means that the electronic certificate is valid at the current time; the regional performance is that the electronic certificate has a corresponding use range. In this embodiment, after the server receives the locking information of the shared vehicle, the server acquires the information to be pushed within the set range according to the position information of the shared vehicle, that is, acquires a plurality of electronic certificates which are matched with the current time and the position of the shared vehicle. If the set range is 3 kilometers, if the position of the shared vehicle is in a certain building, the server acquires all electronic certificates which are currently valid and within the range of 3 kilometers of the building and used for exchanging resources, and the electronic certificates are used as information to be pushed.
And step 206, predicting the use probabilities of the electronic certificates according to the user characteristic information corresponding to the current locked user account.
The use probability of the electronic certificate refers to the probability that the electronic certificate will be used by the user corresponding to the currently locked user account. Specifically, the usage probability of the electronic certificate is obtained based on data mining and machine learning prediction.
And step 208, pushing the electronic certificates with the target number and higher use probability to the current locked user account.
The target number refers to the number of the electronic certificates which are predefined and pushed to the user. Because the obtained electronic certificates to be pushed are all the electronic certificates matched with the positions of the shared vehicles and the current time, if all the matched electronic certificates are pushed to the users, more traffic is occupied, and most electronic certificate users may not need the electronic certificates, thereby causing resource waste. Therefore, in this embodiment, the usage probability of each electronic certificate is predicted based on the user characteristic information corresponding to the currently locked user account, and then the electronic certificates with the target number of higher usage probability are pushed to the user, that is, only the electronic certificates with higher interest degree are pushed to the user, so that traffic is saved, and accurate pushing is realized.
According to the information pushing method, the locking information of the shared vehicle is received, the position information of the shared vehicle and the user characteristic information corresponding to the current locked user account are obtained according to the locking information, the information to be pushed in the set range is obtained according to the position information of the shared vehicle, the use probabilities of a plurality of electronic certificates in the information to be pushed are predicted according to the user characteristic information corresponding to the current locked user account, and the electronic certificates with the target number of high use probabilities are pushed to the current locked user account.
In an embodiment, the use probabilities of a plurality of electronic certificates are predicted according to user characteristic information corresponding to a user account currently locked, which may be specifically implemented by using a deep learning use probability prediction model, and specifically, as shown in fig. 3, the generation method of the use probability prediction model includes the following steps:
The training data set is obtained by pushing electronic certificates to a plurality of sample user accounts using the shared vehicle. Specifically, the training data set includes user characteristic information corresponding to each sample user account, characteristic information of an electronic certificate pushed to the sample user account, and flag information of whether the electronic certificate is used, in this embodiment, the user characteristic information includes user basic information and user behavior characteristic information, the user basic information includes, but is not limited to, an age, a gender, an occupation, and the like of the user, and the user behavior characteristic information includes, but is not limited to, a historical riding path of the user, a record of historical use of the electronic certificate by the user, and the like. The characteristic information of the electronic voucher comprises the type (such as catering, shopping and the like) of each electronic voucher, attribute information (such as the merchant qualification, the merchant score and the like for exchanging the resources) of the corresponding resource of the electronic voucher, the distance between the use place of the electronic voucher and the position of the shared vehicle and the like. Specifically, the electronic voucher may be a coupon, voucher, or the like for exchanging resources.
For example, the training data set may be obtained by first randomly pushing information to a large number (e.g., 10000) of users using the shared vehicle, such as pushing valid electronic certificates within a set range (e.g., within 3 km) near the locked location of the shared vehicle to the users after the users use the shared vehicle and lock the shared vehicle, and recovering the result of whether the pushed electronic certificates are used after a set time. If the electronic certificate is pushed for 1 hour, acquiring a result of whether the electronic certificate is used, and marking the electronic certificate through marking information, for example, if a certain electronic certificate is used, marking the electronic certificate as 1, acquiring user characteristic information corresponding to a sample user account using the electronic certificate and characteristic information of the electronic certificate at the same time, and taking the user characteristic information and the characteristic information as training data to be included in a positive sample; if a certain electronic certificate is not used, marking the electronic certificate as 0, simultaneously acquiring user characteristic information corresponding to a sample user account receiving the electronic certificate and the characteristic information of the electronic certificate, and taking the user characteristic information and the characteristic information as training data to be included in a negative sample. Thereby, obtain the training data of a plurality of electronic certificates that are pushed to 10000 users, that is, training data set.
And step 304, training the gradient boosting iterative decision tree network through the training data set to obtain a use probability prediction model.
The usage probability prediction model is a deep learning machine model for predicting the usage probability of the electronic certificate, and specifically, the gradient boosting iterative Decision tree gbdt (gradient boosting Decision tree) is trained by the obtained training data set, so that the usage probability prediction model capable of predicting the usage probability of the electronic certificate can be obtained by learning the relevant information reflecting time, place and user characteristics from the characteristic information, the marking information and the user characteristic information of the electronic certificate in the training data.
Specifically, the GBDT training process is as follows: by fitting the training data set and newly generating a decision tree at each step of the fitting, the first and second derivatives of the loss function on each sample (i.e. each training data) need to be calculated, i.e. g, before fitting the treeiAnd hi. After the tree is generated through the greedy strategy, G of each leaf node is calculatedjAnd HjAnd using the equationAnd calculating a predicted value w. Further generating a new decision tree ft(x) Adding intoThus, a probabilistic predictive model is used, wherein epsilon is the learning rate, mainly for suppressing the overfitting of the model.
In one embodiment, obtaining user characteristic information corresponding to a currently locked user account according to the locking information includes: and acquiring user characteristic information such as user basic information and user behavior characteristic information corresponding to the currently locked user account from a distributed database according to the locking information, wherein the user basic information includes but is not limited to the age, sex, occupation and the like of the user, and the user behavior characteristic information includes but is not limited to the historical riding path of the user and the record of the historical use of the electronic voucher by the user. The distributed database can be realized by adopting HBase, and the data of T +1 (namely the previous day) which is stable and cannot change in real time along with time and place is stored in the distributed database, so that stable user characteristic information, such as user basic information and user behavior characteristic information, can be obtained by directly accessing the HBase.
Further, acquiring information to be pushed in a set range according to the position information of the shared vehicle includes: and acquiring a plurality of electronic certificates and characteristic information of each electronic certificate within a set range by calling a data transmission interface and taking the position information of the shared vehicle as the center. The characteristic information of the electronic certificate comprises the type of the electronic certificate, attribute information of resources corresponding to the electronic certificate, the distance between the using place of the electronic certificate and the position of the shared vehicle and the like. Specifically, the type of the electronic certificate is the type of the electronic certificate used for exchanging resources, including but not limited to catering, shopping and the like; the attribute information of the resource corresponding to the electronic certificate includes, but is not limited to, the merchant qualification, the merchant score, etc. of the electronic certificate for exchanging the resource. Since the plurality of electronic certificates are acquired based on the position information of the shared vehicle, when the positions of the shared vehicle are different, the plurality of acquired electronic certificates and the corresponding feature information are also different. Based on this, the part of information which changes in real time can be acquired from the corresponding external system in real time by calling the data transmission interface with the outside.
Specifically, as shown in fig. 4, the method for predicting the use probabilities of a plurality of electronic certificates according to the user characteristic information corresponding to the currently locked user account by using the deep learning use probability prediction model includes the following steps:
step 402, preprocessing the user characteristic information corresponding to the currently locked user account and the characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate.
The data to be predicted refers to data required for predicting the use probability of the electronic certificate. The preprocessing is processing such as rearranging, superimposing, or editing data before prediction of the usage probability. Specifically, in this embodiment, the preprocessing refers to splicing the user characteristic information corresponding to the currently locked user account and the characteristic information of each electronic certificate according to a set data splicing rule, so as to obtain spliced data to be predicted corresponding to each electronic certificate, where the user characteristic information includes user basic information and user behavior characteristic information.
And step 404, respectively inputting the data to be predicted corresponding to each electronic certificate into the deep learning use probability prediction model to respectively obtain the predicted use probability of each electronic certificate.
Specifically, the use probability for each electronic certificate obtained by the deep learning use probability prediction model is a value between 0 and 1.
In this embodiment, in order to improve the success rate of information pushing and achieve accurate pushing, so as to obtain user characteristic information, such as user basic information and user behavior characteristic information, corresponding to a current locked user account from a distributed database according to the locking information, obtain a plurality of electronic certificates and characteristic information of each electronic certificate within a set range with the position information of a shared vehicle as the center by calling a data transmission interface, further pre-process the user basic information, the user behavior characteristic information and the characteristic information of each electronic certificate corresponding to the current locked user account, obtain data to be predicted corresponding to each electronic certificate, and predict the use probability of each electronic certificate through a deep-learning use probability prediction model, where the use probability sufficiently reflects the relationship between time, place and user characteristics, electronic credentials of interest may be pushed to the user based on the probability of use of the electronic credentials.
In an embodiment, when the electronic certificates with the target number of higher usage probability are pushed to the user account currently locked, the electronic certificates may be specifically sorted according to the predicted size of the usage probability of the electronic certificates, for example, the electronic certificates may be sorted according to the order from the largest usage probability to the smallest usage probability, so as to obtain a sorting result, and then the electronic certificates with the target number of higher usage probability in the sorting result are determined as the electronic certificates with the higher usage probability, so as to push the electronic certificates with the target number of higher usage probability to the user account currently locked, so as to implement accurate pushing.
In one embodiment, in order to further optimize the deep learning usage probability prediction model, as shown in fig. 5, after pushing a target number of electronic certificates with higher usage probability to a currently locked user account, the method further includes the following steps:
The use result comprises a used result or an unused result of each electronic certificate pushed by the user account. The use result may be obtained after the electronic certificate is pushed to the user account for a period of time, for example, after 1 hour, 2 hours, or one day.
And step 504, training a use probability prediction model according to the user characteristic information corresponding to the user account, the characteristic information of each pushed electronic certificate and the corresponding use result.
Specifically, after the electronic certificates with the target number of higher use probabilities are pushed to the current locked user account, the use results of the pushed electronic certificates are further counted, and a use probability prediction model is trained based on the use results of each electronic certificate, the feature information of the electronic certificate and the corresponding user feature information (including user basic information and user behavior feature information), so that the re-trained use probability prediction model is obtained, the prediction accuracy of the use probability prediction model is improved, and the use probability prediction model is optimized.
In an embodiment, as shown in fig. 6, an information pushing method is provided, which is described by taking the method as an example applied to the terminal in fig. 1, and includes the following steps:
step 602, in response to a locking triggering instruction for the shared vehicle, acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked.
The locking triggering instruction of the shared vehicle is an operation or instruction for triggering the vehicle locking device to lock after the travel of the shared vehicle is finished. The position information of the shared vehicle refers to the position of the shared vehicle currently triggered to the off-lock state. The current locked user account refers to the user account of the user who obtains the use right of the shared vehicle and uses the shared vehicle in the corresponding application program. After the shared vehicle is used, the user can change the state of the shared vehicle through a lock device on the shared vehicle, namely, the unlocking state is adjusted to the locking state, so that the journey is finished. Since the state information of the shared vehicle is synchronized to the terminal corresponding to the currently operated user account when the state of the shared vehicle changes, when the terminal detects a lock-off trigger instruction for the shared vehicle, the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account can be acquired, specifically, the user characteristic information includes user basic information and user behavior characteristic information, for example, the user basic information includes, but is not limited to, the age, sex, and occupation of the user, and the user behavior characteristic information includes, but is not limited to, the historical riding path of the user and the record of the user's historical use of the electronic credential.
And step 604, sending the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account to the server.
Specifically, when the terminal detects a lock closing triggering instruction for the shared vehicle, and after position information of the shared vehicle and user characteristic information corresponding to a currently locked user account are obtained, the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account are sent to the server, so that the server is instructed to obtain a target number of electronic certificates with a high use probability according to the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account, wherein the electronic certificates can be used for exchanging resources.
And 606, receiving the target number of electronic certificates with higher use probability returned by the server.
The usage probability of the electronic certificate refers to the probability that the electronic certificate will be used by the user corresponding to the currently locked user account. Specifically, the usage probability of the electronic certificate is obtained by the server side based on data mining and machine learning prediction.
According to the information pushing method, the position information of the shared vehicle and the user characteristic information corresponding to the current locked user account are obtained by responding to the locking triggering instruction of the shared vehicle, the position information of the shared vehicle and the user characteristic information corresponding to the current locked user account are sent to the server, and the electronic certificates with the target number and high use probability returned by the server are received. Thereby realized the accurate propelling movement of electron voucher, and can improve the success rate of information propelling movement, still avoided pasting the problem that advertisement influences the appearance of the market on the shared vehicle moreover.
In one embodiment, the method further comprises: and responding to the use triggering instruction of the electronic certificate, and sending a use result of the electronic certificate to the server. Specifically, when a user needs to use the electronic certificate in the terminal, the electronic certificate in the terminal can be checked and sold in a corresponding merchant, the terminal responds to the checking and selling operation of the electronic certificate, namely responds to the use triggering instruction of the electronic certificate, and sends the use result of the electronic certificate to the server, so that the server can further optimize according to the use result of the electronic certificate, the accuracy of information pushing is improved, and the accuracy of the use probability of the electronic certificate returned subsequently is improved.
In one embodiment, as shown in fig. 7, the method of the present application is further described below by a specific embodiment, which specifically includes the following steps:
a) the method comprises the steps of collecting user data, wherein the user data comprise historical order information of a shared vehicle used by a user, such as order generation time, use duration, a departure point, a destination and the like, so that information of a historical riding path, a riding distance, riding habits and the like of the user can be obtained according to the historical order information, and basic information of the user, such as the age, the sex, the city, a mobile phone system and the like of the user, can be obtained.
b) And acquiring feedback data of the user on the historical pushed electronic certificate, specifically including information such as whether the user uses the pushed electronic certificate, the type of the used or unused electronic certificate, the distance between the electronic certificate and the parking position of the vehicle in the order corresponding to the shared vehicle and the like.
c) Collecting information of offline merchants using the electronic voucher for exchanging resources: including business type, area, brand, etc. of the merchant.
d) Model training: and taking the data collected in the three steps as training data, taking the data corresponding to the electronic certificate which is used by the user historically as positive sample data and marking the data as 1, and taking the data corresponding to the electronic certificate which is not used by the user historically as negative sample data and marking the data as 0. Training a gradient boosting iterative decision tree network (GBDT) through the marked training data so as to obtain a prediction model of the use probability of the electronic certificate, and predicting the acceptance probability of other users using the shared vehicle and locking the electronic certificate of different merchants at different time and different places through the model, namely the probability of using the electronic certificate by the user.
e) Application of the model: in particular, the model service is triggered in real time by using a lock-off event of the shared vehicle. When a lock closing event of the shared vehicle is detected, the server acquires the position of the shared vehicle according to the lock closing event, recalls valid electronic certificates within a set range (for example, within 3 kilometers around the center) by taking the position of the shared vehicle as a center, predicts the use probability of each electronic certificate through the trained model, and pushes 3 electronic certificates with the highest probability (the number can be configured differently according to actual needs) to a user account of the current lock closing.
For example, if a white-collar morning at 10 am uses a shared vehicle and locks, the server recalls the electronic certificates provided by the merchants around the locked location under the trigger of the locking event, which includes: and the electronic certificates provided by merchants such as an A coffee shop, a B fast food shop, a C convenience store and the like are used for exchanging resources. Through model prediction, since the C convenience store provides breakfast and the model prediction shows that the electronic certificate provided by the C convenience store has the highest use probability of 90%, the electronic certificate of the C convenience store with the highest use probability is returned to the locked user. Similarly, if it is during the afternoon tea, when the locking event is detected, the use probability of the electronic certificate of the coffee shop a is obtained through model prediction, and therefore the electronic certificate of the coffee shop a is pushed to the user.
f) Model optimization: and collecting data generated in the model application process, wherein the data comprises a used result or an unused result of each electronic certificate pushed by the user account, user characteristic information corresponding to the user account and characteristic information of each pushed electronic certificate, and continuously training the model through the collected data to obtain a retrained use probability prediction model so as to improve the prediction accuracy of the use probability prediction model and realize the optimization of the use probability prediction model.
According to the flow, the prediction model is obtained based on data mining and machine learning, the model service is triggered through the locking event of the shared vehicle, so that a closed-loop information pushing method is formed, the appropriate electronic certificate is accurately pushed to the user at the appropriate time and place, not only is accurate pushing realized, but also the success rate of information pushing can be improved, and the problem that the appearance of the market is influenced by posting advertisements on the shared vehicle is avoided.
It should be understood that although the various steps in the flow charts of fig. 1-7 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1-7 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.
In one embodiment, as shown in fig. 8, there is provided an information pushing apparatus including: a locking information receiving module 801, an information to be pushed acquiring module 802, a usage probability predicting module 803, and a pushing module 804, wherein:
the locking information receiving module 801 is configured to receive locking information of a shared vehicle, and obtain position information of the shared vehicle and user characteristic information corresponding to a user account currently locked according to the locking information, where the user characteristic information includes user basic information and user behavior characteristic information;
the information to be pushed acquiring module 802 is configured to acquire information to be pushed within a set range according to the position information of the shared vehicle, where the information to be pushed includes a plurality of electronic certificates used for exchanging resources and matched with current time;
a usage probability prediction module 803, configured to predict usage probabilities of the multiple electronic certificates according to user characteristic information corresponding to a user account currently locked;
the pushing module 804 is configured to push the electronic certificates with the target number and higher usage probability to the user account currently locked.
In one embodiment, the usage probability prediction module 803 is specifically implemented by a deep learning usage probability prediction model, which then includes: the training data set acquisition unit is used for acquiring a training data set, wherein the training data set is obtained by pushing electronic certificates to a plurality of sample user accounts using a shared vehicle, the training data set comprises user characteristic information corresponding to each sample user account, characteristic information of the electronic certificates pushed to the sample user accounts and marking information of whether the electronic certificates are used, and the user characteristic information comprises user basic information and user behavior characteristic information; and the model generation unit is used for training the gradient lifting iterative decision tree network through a training data set to obtain the use probability prediction model.
In an embodiment, the to-be-pushed information obtaining module 802 is specifically configured to: acquiring a plurality of electronic certificates and characteristic information of each electronic certificate within a set range by calling a data transmission interface, wherein the set range takes position information of a shared vehicle as a center, the characteristic information of each electronic certificate comprises the type of the electronic certificate, attribute information of resources corresponding to the electronic certificates and the distance between the using place of the electronic certificate and the position of the shared vehicle, and determining the electronic certificates as information to be pushed; the use probability prediction module 703 includes: the preprocessing unit is used for preprocessing the user characteristic information corresponding to the currently locked user account and the characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate; and the prediction unit is used for respectively inputting the data to be predicted corresponding to each electronic certificate into the deep-learning use probability prediction model to respectively obtain the predicted use probability of each electronic certificate.
In one embodiment, the preprocessing unit is specifically configured to: and splicing the user characteristic information corresponding to the user account currently locked and the characteristic information of each electronic certificate according to a set data splicing rule to obtain spliced data to be predicted corresponding to each electronic certificate.
In one embodiment, the device further comprises a model optimization module, configured to obtain a result of using the pushed target number of electronic certificates by the user account; and training the use probability prediction model according to the user characteristic information corresponding to the user account, the characteristic information of each pushed electronic certificate and the corresponding use result to obtain the re-trained use probability prediction model.
In one embodiment, the pushing module 804 is specifically configured to: ordering the plurality of electronic certificates according to the predicted using probability of the plurality of electronic certificates; and determining the electronic certificates with the target number ranked at the top in the ranking result as the electronic certificates with higher similarity.
For specific limitations of the information pushing apparatus, reference may be made to the above limitations of fig. 1 to 5 on the information pushing method, which is not described herein again. All or part of the modules in the information pushing device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, as shown in fig. 9, there is provided an information pushing apparatus including: an instruction response module 901, a sending module 902 and a receiving module 903, wherein:
the instruction response module 901 is configured to, in response to a lock-off triggering instruction for a shared vehicle, acquire position information of the shared vehicle and user characteristic information corresponding to a user account currently locked, where the user characteristic information includes user basic information and user behavior characteristic information;
a sending module 902, configured to send, to a server, location information of a shared vehicle and user characteristic information corresponding to a user account currently locked, so as to instruct the server to obtain, according to the location information of the shared vehicle and the user characteristic information corresponding to the user account currently locked, a target number of electronic certificates with a higher usage probability;
and the receiving module 903 is configured to receive the target number of electronic certificates with higher use probability returned by the server.
In one embodiment, the sending module 902 is further configured to send the usage result of the electronic certificate to the server in response to the usage triggering instruction of the electronic certificate.
For specific limitations of the information pushing apparatus, reference may be made to the above limitation of the information pushing method in fig. 6, which is not described herein again. All or part of the modules in the information pushing device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 10. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing the user characteristic information and the characteristic information data of the electronic certificate. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement an information push method.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 11. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement an information push method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the configurations shown in fig. 10 and 11 are block diagrams of only some of the configurations relevant to the present disclosure, and do not constitute a limitation on the computing devices to which the present disclosure may be applied, and that a particular computing device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
receiving locking information of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked according to the locking information, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
acquiring information to be pushed in a set range according to the position information of the shared vehicle, wherein the information to be pushed comprises a plurality of electronic certificates which are matched with the current time and used for exchanging resources;
predicting the use probabilities of a plurality of electronic certificates according to user characteristic information corresponding to the current locked user account;
And pushing the electronic certificates with the target number and higher use probability to the current locked user account.
In one embodiment, by using a deep-learning usage probability prediction model and predicting usage probabilities of a plurality of electronic certificates according to user characteristic information corresponding to a user account currently locked, the processor executes the computer program to further implement the following steps: the method comprises the steps of obtaining a training data set, wherein the training data set is obtained by pushing electronic certificates to a plurality of sample user accounts using a shared vehicle, the training data set comprises user characteristic information corresponding to each sample user account, characteristic information of the electronic certificates pushed to the sample user accounts and mark information of whether the electronic certificates are used or not, and the user characteristic information comprises user basic information and user behavior characteristic information; and training the gradient boosting iterative decision tree network through a training data set to obtain a use probability prediction model.
In one embodiment, the processor, when executing the computer program, further performs the steps of: acquiring a plurality of electronic certificates and characteristic information of each electronic certificate within a set range by calling a data transmission interface, wherein the set range takes position information of a shared vehicle as a center, the characteristic information of each electronic certificate comprises the type of the electronic certificate, attribute information of resources corresponding to the electronic certificates and the distance between the using place of the electronic certificate and the position of the shared vehicle, and determining the electronic certificates as information to be pushed; the method for predicting the use probabilities of the electronic certificates by adopting the deep learning use probability prediction model according to the user characteristic information corresponding to the user account currently locked comprises the following steps: preprocessing user characteristic information corresponding to a user account currently locked and characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate; and respectively inputting the data to be predicted corresponding to each electronic certificate into a deep learning use probability prediction model to respectively obtain the predicted use probability of each electronic certificate.
In one embodiment, the processor, when executing the computer program, further performs the steps of: and splicing the user characteristic information corresponding to the user account currently locked and the characteristic information of each electronic certificate according to a set data splicing rule to obtain spliced data to be predicted corresponding to each electronic certificate.
In one embodiment, the processor, when executing the computer program, further performs the steps of: after electronic certificates with a target number and a higher use probability are pushed to a user account locked currently, obtaining a use result of the user account on the electronic certificates with the target number; and training the use probability prediction model according to the user characteristic information corresponding to the user account, the characteristic information of each pushed electronic certificate and the corresponding use result to obtain the re-trained use probability prediction model.
In one embodiment, the processor, when executing the computer program, further performs the steps of: ordering the plurality of electronic certificates according to the predicted using probability of the plurality of electronic certificates; and determining the electronic certificates with the target number ranked at the top in the ranking result as the electronic certificates with higher similarity.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
Receiving locking information of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked according to the locking information, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
acquiring information to be pushed in a set range according to the position information of the shared vehicle, wherein the information to be pushed comprises a plurality of electronic certificates which are matched with the current time and used for exchanging resources;
predicting the use probabilities of a plurality of electronic certificates according to user characteristic information corresponding to the current locked user account;
and pushing the electronic certificates with the target number and higher use probability to the current locked user account.
In one embodiment, by using a deep-learning usage probability prediction model and predicting usage probabilities of a plurality of electronic certificates according to user characteristic information corresponding to a user account currently locked, the computer program when executed by the processor further implements the following steps: the method comprises the steps of obtaining a training data set, wherein the training data set is obtained by pushing electronic certificates to a plurality of sample user accounts using a shared vehicle, the training data set comprises user characteristic information corresponding to each sample user account, characteristic information of the electronic certificates pushed to the sample user accounts and mark information of whether the electronic certificates are used or not, and the user characteristic information comprises user basic information and user behavior characteristic information; and training the gradient boosting iterative decision tree network through a training data set to obtain a use probability prediction model.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a plurality of electronic certificates and characteristic information of each electronic certificate within a set range by calling a data transmission interface, wherein the set range takes position information of a shared vehicle as a center, the characteristic information of each electronic certificate comprises the type of the electronic certificate, attribute information of resources corresponding to the electronic certificates and the distance between the using place of the electronic certificate and the position of the shared vehicle, and determining the electronic certificates as information to be pushed; the method for predicting the use probabilities of the electronic certificates by adopting the deep learning use probability prediction model according to the user characteristic information corresponding to the user account currently locked comprises the following steps: preprocessing user characteristic information corresponding to a user account currently locked and characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate; and respectively inputting the data to be predicted corresponding to each electronic certificate into a deep learning use probability prediction model to respectively obtain the predicted use probability of each electronic certificate.
In one embodiment, the computer program when executed by the processor further performs the steps of: and splicing the user characteristic information corresponding to the user account currently locked and the characteristic information of each electronic certificate according to a set data splicing rule to obtain spliced data to be predicted corresponding to each electronic certificate.
In one embodiment, the computer program when executed by the processor further performs the steps of: after electronic certificates with a target number and a higher use probability are pushed to a user account locked currently, obtaining a use result of the user account on the electronic certificates with the target number; and training the use probability prediction model according to the user characteristic information corresponding to the user account, the characteristic information of each pushed electronic certificate and the corresponding use result to obtain the re-trained use probability prediction model.
In one embodiment, the computer program when executed by the processor further performs the steps of: ordering the plurality of electronic certificates according to the predicted using probability of the plurality of electronic certificates; and determining the electronic certificates with the target number ranked at the top in the ranking result as the electronic certificates with higher similarity.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
responding to a locking triggering instruction of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
Sending the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account to the server to indicate the server to acquire the electronic certificates with the target number and high use probability according to the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account;
and receiving the target number of electronic certificates with higher use probability returned by the server.
In one embodiment, the processor, when executing the computer program, performs the steps of: and responding to the use triggering instruction of the electronic certificate, and sending a use result of the electronic certificate to the server.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
responding to a locking triggering instruction of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
sending the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account to the server to indicate the server to acquire the electronic certificates with the target number and high use probability according to the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account;
And receiving the target number of electronic certificates with higher use probability returned by the server.
In one embodiment, the computer program when executed by the processor further performs the steps of: and responding to the use triggering instruction of the electronic certificate, and sending a use result of the electronic certificate to the server.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (12)
1. An information pushing method, characterized in that the method comprises:
receiving locking information of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked according to the locking information, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
acquiring information to be pushed in a set range according to the position information of the shared vehicle, wherein the information to be pushed comprises a plurality of electronic certificates which are matched with the current time and used for exchanging resources;
Predicting the use probabilities of the electronic certificates according to user characteristic information corresponding to the current locked user account;
and pushing the electronic certificates with the target number and higher use probability to the current locked user account.
2. The method according to claim 1, wherein the usage probability prediction model is generated by predicting the usage probabilities of the plurality of electronic certificates according to user characteristic information corresponding to the currently locked user account by using a deep learning usage probability prediction model, and the usage probability prediction model comprises:
the method comprises the steps of obtaining a training data set, wherein the training data set is obtained by pushing electronic certificates to a plurality of sample user accounts using a shared vehicle, the training data set comprises user characteristic information corresponding to each sample user account, characteristic information of the electronic certificates pushed to the sample user accounts and mark information of whether the electronic certificates are used or not, and the user characteristic information comprises user basic information and user behavior characteristic information;
and training a gradient boosting iterative decision tree network through the training data set to obtain the use probability prediction model.
3. The method according to claim 2, wherein the obtaining of the information to be pushed within a set range according to the position information of the sharing vehicle comprises:
Acquiring a plurality of electronic certificates and characteristic information of each electronic certificate within a set range by taking position information of the shared vehicle as a center through calling a data transmission interface, wherein the characteristic information of the electronic certificates comprises types of the electronic certificates, attribute information of resources corresponding to the electronic certificates and a distance between a use place of the electronic certificates and the positions of the shared vehicle, and determining the electronic certificates as information to be pushed;
the method for predicting the use probabilities of the electronic certificates by adopting the deep learning use probability prediction model according to the user characteristic information corresponding to the user account currently locked comprises the following steps:
preprocessing user characteristic information corresponding to a user account currently locked and characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate;
and respectively inputting the data to be predicted corresponding to each electronic certificate into the deep learning use probability prediction model to respectively obtain the predicted use probability of each electronic certificate.
4. The method according to claim 3, wherein the preprocessing the user characteristic information corresponding to the currently locked user account and the characteristic information of each electronic certificate to obtain data to be predicted corresponding to each electronic certificate comprises:
And splicing the user characteristic information corresponding to the user account currently locked and the characteristic information of each electronic certificate according to a set data splicing rule to obtain spliced data to be predicted corresponding to each electronic certificate.
5. The method according to claim 3, wherein after the target number of electronic certificates with higher usage probability are pushed to the currently locked user account, the method further comprises:
obtaining the use result of the electronic certificates of the target number pushed by the user account;
and training the use probability prediction model according to the user characteristic information corresponding to the user account, the characteristic information of each pushed electronic certificate and the corresponding use result to obtain the retrained use probability prediction model.
6. The method according to any one of claims 1 to 5, wherein the pushing of the target number of electronic certificates with higher usage probability to the currently locked user account comprises:
ordering the plurality of electronic vouchers according to the predicted size of the usage probabilities of the plurality of electronic vouchers;
and determining the electronic certificates with the target number ranked at the top in the ranking result as the electronic certificates with higher use probability.
7. An information pushing method, characterized in that the method comprises:
responding to a locking triggering instruction of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
sending the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account to a server to indicate the server to acquire the electronic certificates with the target number and high use probability according to the position information of the shared vehicle and the user characteristic information corresponding to the currently locked user account;
and receiving the target number of electronic certificates with higher use probability returned by the server.
8. The method of claim 7, further comprising:
and responding to a use triggering instruction of the electronic certificate, and sending a use result of the electronic certificate to the server side.
9. An information pushing apparatus, characterized in that the apparatus comprises:
the locking information receiving module is used for receiving locking information of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked according to the locking information, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
The information to be pushed acquiring module is used for acquiring information to be pushed in a set range according to the position information of the shared vehicle, wherein the information to be pushed comprises a plurality of electronic certificates which are matched with the current time and used for exchanging resources;
the usage probability prediction module is used for predicting the usage probabilities of the electronic certificates according to the user characteristic information corresponding to the user account currently locked;
and the pushing module is used for pushing the electronic certificates with the target number and higher use probability to the current locked user account.
10. An information pushing apparatus, characterized in that the apparatus comprises:
the instruction response module is used for responding to a locking triggering instruction of a shared vehicle, and acquiring position information of the shared vehicle and user characteristic information corresponding to a user account currently locked, wherein the user characteristic information comprises user basic information and user behavior characteristic information;
the sending module is used for sending the position information of the shared vehicle and the user characteristic information corresponding to the user account locked currently to the server so as to indicate the server to obtain the electronic certificates with the target number and higher use probability according to the position information of the shared vehicle and the user characteristic information corresponding to the user account locked currently;
And the receiving module is used for receiving the electronic certificates with the target number and higher using probability returned by the server.
11. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 8.
12. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010516358.5A CN111861540A (en) | 2020-06-09 | 2020-06-09 | Information pushing method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010516358.5A CN111861540A (en) | 2020-06-09 | 2020-06-09 | Information pushing method and device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111861540A true CN111861540A (en) | 2020-10-30 |
Family
ID=72987336
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010516358.5A Pending CN111861540A (en) | 2020-06-09 | 2020-06-09 | Information pushing method and device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111861540A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113919856A (en) * | 2020-07-09 | 2022-01-11 | 上海钧正网络科技有限公司 | Target user selection method, system, device and storage medium |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170244689A1 (en) * | 2016-02-22 | 2017-08-24 | International Business Machines Corporation | Retrospective discovery of shared credentials |
CN107180366A (en) * | 2017-05-23 | 2017-09-19 | 周雷文 | The information-pushing method and system of a kind of shared bicycle |
CN107391605A (en) * | 2017-06-30 | 2017-11-24 | 北京奇虎科技有限公司 | Information-pushing method, device and mobile terminal based on geographical position |
CN108337316A (en) * | 2018-02-08 | 2018-07-27 | 平安科技(深圳)有限公司 | Information-pushing method, device, computer equipment and storage medium |
CN109167816A (en) * | 2018-08-03 | 2019-01-08 | 广州虎牙信息科技有限公司 | Information-pushing method, device, equipment and storage medium |
CN110415032A (en) * | 2019-07-24 | 2019-11-05 | 深圳乐信软件技术有限公司 | A kind of exposure conversion ratio predictor method, device, equipment and storage medium |
CN110458638A (en) * | 2019-06-26 | 2019-11-15 | 平安科技(深圳)有限公司 | A kind of Method of Commodity Recommendation and device |
CN110766507A (en) * | 2019-02-25 | 2020-02-07 | 北京嘀嘀无限科技发展有限公司 | Resource allocation method and device |
CN111131415A (en) * | 2019-12-12 | 2020-05-08 | 上海钧正网络科技有限公司 | Information pushing method and device, computer equipment and storage medium |
-
2020
- 2020-06-09 CN CN202010516358.5A patent/CN111861540A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170244689A1 (en) * | 2016-02-22 | 2017-08-24 | International Business Machines Corporation | Retrospective discovery of shared credentials |
CN107180366A (en) * | 2017-05-23 | 2017-09-19 | 周雷文 | The information-pushing method and system of a kind of shared bicycle |
CN107391605A (en) * | 2017-06-30 | 2017-11-24 | 北京奇虎科技有限公司 | Information-pushing method, device and mobile terminal based on geographical position |
CN108337316A (en) * | 2018-02-08 | 2018-07-27 | 平安科技(深圳)有限公司 | Information-pushing method, device, computer equipment and storage medium |
CN109167816A (en) * | 2018-08-03 | 2019-01-08 | 广州虎牙信息科技有限公司 | Information-pushing method, device, equipment and storage medium |
CN110766507A (en) * | 2019-02-25 | 2020-02-07 | 北京嘀嘀无限科技发展有限公司 | Resource allocation method and device |
CN110458638A (en) * | 2019-06-26 | 2019-11-15 | 平安科技(深圳)有限公司 | A kind of Method of Commodity Recommendation and device |
CN110415032A (en) * | 2019-07-24 | 2019-11-05 | 深圳乐信软件技术有限公司 | A kind of exposure conversion ratio predictor method, device, equipment and storage medium |
CN111131415A (en) * | 2019-12-12 | 2020-05-08 | 上海钧正网络科技有限公司 | Information pushing method and device, computer equipment and storage medium |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113919856A (en) * | 2020-07-09 | 2022-01-11 | 上海钧正网络科技有限公司 | Target user selection method, system, device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Cottrill | MaaS surveillance: Privacy considerations in mobility as a service | |
Gurumurthy et al. | Analyzing the dynamic ride-sharing potential for shared autonomous vehicle fleets using cellphone data from Orlando, Florida | |
US11669864B2 (en) | Determining locations of interest based on user visits | |
US8401771B2 (en) | Discovering points of interest from users map annotations | |
Bicocchi et al. | Investigating ride sharing opportunities through mobility data analysis | |
Lee et al. | Internet of Vehicles: From intelligent grid to autonomous cars and vehicular fogs | |
Zhang et al. | On reliable task assignment for spatial crowdsourcing | |
Wang et al. | Moving destination prediction using sparse dataset: A mobility gradient descent approach | |
CN111868763A (en) | Optimizing and predicting resource availability in a shared vehicle environment | |
Hachem et al. | Service-oriented middleware for large-scale mobile participatory sensing | |
Zhang et al. | Feeder: supporting last-mile transit with extreme-scale urban infrastructure data | |
Duarte | Disassembling bike-sharing systems: Surveillance, advertising, and the social inequalities of a global technological assemblage | |
Xiao et al. | Multi-user mobile sequential recommendation for route optimization | |
Zhang et al. | Impacts of urban built environment on empty taxi trips using limited geolocation data | |
Tang et al. | Discovering Travel Community for POI Recommendation on Location‐Based Social Networks | |
Liu et al. | Heuristic approach for the multiobjective optimization of the customized bus scheduling problem | |
Petkovics et al. | Crowdsensing solutions in smart cities towards a networked society | |
Bicocchi et al. | On recommending opportunistic rides | |
Li et al. | A location-aware recommender system for Tourism mobile commerce | |
Antoniou et al. | The value of privacy: Evidence from the use of mobile devices for traveler information systems | |
Huang et al. | Multi-head spatio-temporal attention mechanism for urban anomaly event prediction | |
CN111861540A (en) | Information pushing method and device, computer equipment and storage medium | |
Zhang et al. | Last-mile transit service with urban infrastructure data | |
Mbiydzenyuy et al. | Optimization analysis of multiservice architecture concepts in road transport telematics | |
Perra | A framework for the development of sustainable urban mobility applications |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201030 |