CN111814035B - Information recommendation method, electronic equipment and storage medium - Google Patents

Abstract

The invention provides an information recommendation method, electronic equipment and a storage medium, wherein the method comprises the following steps: responding to the login information of the current terminal under the current task, and acquiring the current terminal identification; acquiring historical operation data under all task types corresponding to the current terminal identification; and generating a recommendation information list according to the type of the current task and the historical operation data and controlling the current terminal to display recommendation information according to the sequence of the recommendation information list. According to the scheme provided by the invention, after the current terminal logs in under the current task, the historical operation of the current terminal under the current task is referred, and the recommendation information is generated according to the historical operation of the current terminal under all types of tasks, so that the accuracy of the recommendation information is greatly improved compared with the recommendation mode in the prior art.

Description

Information recommendation method, electronic equipment and storage medium

Technical Field

The present invention relates to the field of information processing technologies, and in particular, to an information recommendation method, an electronic device, and a storage medium.

Background

At present, information recommendation applications are involved in many application scenarios, and the recommended information is used for selection by a user terminal, so that the user terminal is prevented from inputting related content. For example, when a user logs in to the network taxi-taking platform and clicks on an online customer service to provide help, the online customer service often automatically recommends relevant questions for the user to select. At present, the information recommendation implementation modes of the method adopt the problems proposed by the user terminal at the last few times or recommend the information recommendation implementation modes under the task type of the user terminal at present. For example, as shown in fig. 1, since the previous problem posed by the client is related to fee deduction, refund, etc., when the client clicks on the online customer service, the recommendation problem is related to fee. For another example, the task type the user is currently on is a taxi service, in which case the recommended questions of online customer service are all related to the taxi service, such as different charging standards for different types of vehicles, what differences are among different types of vehicles, how to complain if the taxi experience is not satisfied, etc. However, the accuracy of the information recommended in the two ways is limited, and the search items are required to be input again in most cases. Therefore, a more accurate information recommendation method is needed.

Disclosure of Invention

The embodiment of the invention aims to provide an information recommendation method, electronic equipment and a storage medium, so as to solve the technical problem of low accuracy of recommended information in the prior art.

For this purpose, the invention provides an information recommendation method, comprising the following steps:

responding to the login information of the current terminal under the current task, and acquiring the current terminal identification;

acquiring historical operation data under all task types corresponding to the current terminal identification;

and generating a recommendation information list according to the type of the current task and the historical operation data and controlling the current terminal to display recommendation information according to the sequence of the recommendation information list.

Optionally, the information recommending method includes, before the step of generating the recommended information list according to the type of the current task and the historical operation data:

Acquiring record data of the terminal selection recommendation information under each task type, wherein the record data comprises: the terminal has completed operation data under all task types; the terminal finally selects recommended information or information identification of autonomous input information;

Obtaining the corresponding relation between the completed operation data of the terminal under all task types and the information identification of the recommendation information or the autonomous input information finally selected by the terminal according to the recorded data;

in the step of generating a recommendation information list according to the type of the current task and the historical operation data: and obtaining the recommendation information list according to the corresponding relation, the type of the current task and the historical operation data.

Optionally, before the step of generating the recommended information list according to the type of the current task and the historical operation data, the information recommending method further includes:

Analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking information identification of recommendation information or autonomous input information finally selected by a terminal as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model;

In the step of generating a recommendation information list according to the type of the current task and the historical operation data: and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model.

Optionally, in the information recommending method, the step of generating a recommended information list according to the type of the current task and the historical operation data and controlling the current terminal to display recommended information according to the order of the recommended information list further includes:

And acquiring the type of the task to which each piece of recommended information belongs, and controlling the current terminal to display each piece of recommended information and the type of the task to which each piece of recommended information belongs in an associated mode.

Optionally, before the step of generating the recommended information list according to the type of the current task and the historical operation data, the information recommending method further includes:

Analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking an information identifier of recommendation information or autonomous input information finally selected by a terminal and a type of a task to which the recommendation information belongs as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model;

In the step of generating a recommendation information list according to the type of the current task and the historical operation data: and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model.

Optionally, in the information recommendation method, the machine learning model is a MAML element learning mathematical model, and the training step includes:

Setting parameters K, N and S in the MAML element learning mathematical model, wherein K represents the number of task types processed by the model at one time, N represents sample data under each task type, and S represents the total number of tasks processed by the model at each time;

Dividing a characteristic variable obtained after the analysis of the operation data is completed into a plurality of sample sets according to the type of the task;

And selecting the corresponding characteristic variable in the sample set as an input variable of the MAML element learning mathematical model according to parameters K, N and S in the MAML element learning mathematical model, and training the MAML element learning mathematical model.

The invention also provides an information recommendation electronic device, which comprises at least one processor and a memory in communication connection with at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to:

responding to the login information of the current terminal under the current task, and acquiring the current terminal identification;

acquiring historical operation data under all task types corresponding to the current terminal identification;

and generating a recommendation information list according to the type of the current task and the historical operation data and controlling the current terminal to display recommendation information according to the sequence of the recommendation information list.

Optionally, in the above information recommendation electronic device, the processor is further capable of:

Acquiring record data of the terminal selection recommendation information under each task type, wherein the record data comprises: the terminal has completed operation data under all task types; the terminal finally selects recommended information or information identification of autonomous input information;

Obtaining the corresponding relation between the completed operation data of the terminal under all task types and the information identification of the recommendation information or the autonomous input information finally selected by the terminal according to the recorded data;

And obtaining the recommendation information list according to the corresponding relation, the type of the current task and the historical operation data.

Optionally, in the above information recommendation electronic device, the processor is further capable of:

Analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking information identification of recommendation information or autonomous input information finally selected by a terminal as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model;

And extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model.

Optionally, in the above information recommendation electronic device, the processor is further capable of:

And acquiring the type of the task to which each piece of recommended information belongs, and controlling the current terminal to display each piece of recommended information and the type of the task to which each piece of recommended information belongs in an associated mode.

Optionally, in the above information recommendation electronic device, the processor is further capable of:

Analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking an information identifier of recommendation information or autonomous input information finally selected by a terminal and a type of a task to which the recommendation information belongs as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model;

And extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model.

Optionally, in the above information recommendation electronic device, the machine learning model is a MAML element learning mathematical model, and the processor is further capable of:

Setting parameters K, N and S in the MAML element learning mathematical model, wherein K represents the number of task types processed by the model at one time, N represents sample data under each task type, and S represents the total number of tasks processed by the model at each time;

Dividing a characteristic variable obtained after the analysis of the operation data is completed into a plurality of sample sets according to the type of the task;

And selecting the corresponding characteristic variable in the sample set as an input variable of the MAML element learning mathematical model according to parameters K, N and S in the MAML element learning mathematical model, and training the MAML element learning mathematical model.

The invention also provides a storage medium which can be read and written by a computer, wherein the storage medium stores program instructions, and the computer executes the information recommendation method according to any one of the above schemes after reading the program instructions.

Compared with the prior art, the technical scheme provided by the embodiment of the invention has at least the following beneficial effects:

The embodiment of the invention provides an information recommendation method, electronic equipment and a storage medium, wherein the method comprises the following steps: responding to the login information of the current terminal under the current task, and acquiring the current terminal identification; acquiring historical operation data under all task types corresponding to the current terminal identification; and generating a recommendation information list according to the type of the current task and the historical operation data and controlling the current terminal to display recommendation information according to the sequence of the recommendation information list. According to the scheme, after the current terminal logs in under the current task, the historical operation of the current terminal under the current task is referred, and the recommendation information is generated according to the historical operation of the current terminal under all types of tasks, so that the accuracy of the recommendation information is improved.

Drawings

FIG. 1 is a schematic diagram of a recommendation result for an "online customer service consultation problem" in an online customer service display interface;

FIG. 2 is a flowchart of an information recommendation method according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a recommendation result for an "online customer service consultation problem" in an online customer service display interface according to an embodiment of the present invention;

FIG. 4 is a flowchart of an information recommendation method according to another embodiment of the present invention;

Fig. 5 is a schematic diagram of a hardware connection relationship of an electronic device for implementing the information recommendation method according to an embodiment of the present invention.

Detailed Description

Embodiments of the present invention will be further described below with reference to the accompanying drawings. In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of description of the present invention, and are not to indicate or imply that the apparatus or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Wherein the terms "first position" and "second position" are two different positions.

In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between the two components. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.

Example 1

The embodiment provides an information recommendation method, which can be applied to a cloud server, as shown in fig. 2, and includes the following steps:

s101: responding to the login information of the current terminal under the current task, and acquiring the current terminal identification; taking the car-restraining platform as an example for explanation, when a passenger uses the car-restraining platform to carry out car-restraining, the car-restraining application program is downloaded and registered through the mobile phone end, and when the passenger logs in the application program, login information is sent to the cloud server, so that the cloud server responds to the login information.

S102: and acquiring historical operation data under all task types corresponding to the current terminal identification. When the vehicle-restraining application program is logged in through the mobile phone, the application program can acquire information filled in during registration and send the information to the cloud server, and the cloud server can acquire and store terminal identifiers of all registered passengers and operation data executed during each login. Therefore, after the passenger logs in, the cloud server can call the history information corresponding to the current terminal.

S103: and generating a recommendation information list according to the type of the current task and the historical operation data and controlling the current terminal to display recommendation information according to the sequence of the recommendation information list. That is, regardless of which task type the current terminal enters, when providing recommendation information for the current terminal, a recommendation information list is generated based on operations it has performed under all task types. Still taking the car-holding platform as an example, as shown in fig. 3, when the current terminal clicks on the online customer service, the cloud server can retrieve all the history data corresponding to the current terminal, for example, according to the history data, the cloud server can find out that a driver notes that a left-over object is on the car for the current terminal, and even if the current terminal is currently the online customer service clicked under the car-holding task, the related recommendation of the lost object of the passenger can appear in the recommended problem of 'guessing you want to ask'.

Obviously, by the scheme, after the current terminal logs in under the current task, the recommendation information can be generated according to the historical operation of the current terminal under all types of tasks besides referencing the historical operation of the current terminal under the current task, so that the accuracy of the recommendation information is improved.

Example 2

The embodiment provides an information recommendation method, as shown in fig. 4, including:

S201: acquiring record data of the terminal selection recommendation information under each task type, wherein the record data comprises: the terminal has completed operation data under all task types; and finally selecting recommended information or information identification of autonomous input information by the terminal. The autonomous input information refers to information that no corresponding related information exists in the current recommended information list, and the terminal receives autonomous input information of the passenger finally. There will be a corresponding identification for each information under each task type to distinguish between different information. For example, the problem: how are the allied car platforms? And the problems: how do complaints drivers? Which necessarily have different identifications.

S202: obtaining the corresponding relation between the completed operation data of the terminal under all task types and the information identification of the recommendation information or the autonomous input information finally selected by the terminal according to the recorded data; for each terminal, there may be an operation of selecting recommended information in a history period under various tasks, and since different selections may be made for history data that is not used when the terminal performs the corresponding operation, the records of all terminals in the history of the terminal are used as sample data to obtain a correspondence between a desired history operation record and the recommended information selection.

S203: and responding to the login information of the current terminal under the current task, and acquiring the current terminal identification. When a passenger logs in an application program, login information is sent to the cloud server, and is responded to by the cloud server.

S204: and acquiring historical operation data under all task types corresponding to the current terminal identification. After the passenger logs in, the cloud server can call the history information corresponding to the current terminal.

S205: and obtaining the recommendation information list according to the corresponding relation, the type of the current task and the historical operation data.

The corresponding relation between the historical operation and the recommended information selection is obtained through the steps, so that the current terminal can determine which type of information is the recommended information with the highest matching degree with the current terminal according to the corresponding relation after the historical operation data of the current terminal is obtained. The above process can be realized by a big data analysis method, and specifically, the scheme is realized by the following steps:

s301: analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking information identification of recommendation information or autonomous input information finally selected by a terminal as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model;

s302: and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model.

In the above scheme, the feature variables selected should be the same when training the data model and when finally obtaining the recommended information list. For example, in the vehicle-restraining platform, the vehicle type, the on-board point, the off-board point, the drunk, the evaluation, the payment timeliness and the like can be used.

The machine learning model can adopt various models which are mature and exist in the prior art, and after sample data are obtained, the sample data are directly utilized to carry out iterative training on the existing model. In this embodiment, preferably, the machine learning model is a MAML primitive learning mathematical model, and MAML (Model Agnostic META LEARNING) is a primitive learning technique, which is a deep learning algorithm, and is mainly used for learning how to learn. In popular terms, meta learning constructs a basic model first, and then designs a learning mechanism to make the basic model learn further on new tasks. Meta-learning is not directed to the results of learning, but rather to the process of learning. It learns not a mathematical model that is directly used for prediction, but "how faster and better to learn a mathematical model". Meta-learning enables classification using learned knowledge to new fields with fewer samples as possible. Embedding (embedded knowledge) technology has been widely used for representation learning in recent years, mainly for feature mining and feature representation of ID-class features, such as generating passenger taxi Embedding according to a passenger taxi type relationship, and generating driver Embedding according to a driver complaint condition. Because different learning tasks can generate different feature representations for ID class features, how to fuse embedding learned in different fields and train on the new task is a problem to be solved by the invention. In the embodiment of the invention, the types of tasks are divided by Embedding, and the association between different information features under different Embedding conditions is mined by MAML, so that the association between the selected recommended information under the current Embedding condition and the historical data of all Embedding conditions can be obtained. In this scheme, the training steps of MAML meta learning mathematical model include:

S401: setting parameters K, N and S in the MAML element learning mathematical model, wherein K represents the number of task types processed by the model at one time, N represents sample data under each task type, and S represents the total number of tasks processed by the model at each time; K. n and S are parameters of the MAML algorithm, and the parameters are set according to the data quantity to be processed in the step.

S402: dividing a characteristic variable obtained after the analysis of the operation data is completed into a plurality of sample sets according to the type of the task; as previously described, using Embedding to divide the types of tasks, we aggregate a total of 7 Embedding vectors including drunk passengers, frequent places of passengers, comments of passengers, complaints of passengers, etc., for the customer service dialog guesses of you want to ask tasks shown in fig. 3.

S403: and selecting the corresponding characteristic variable in the sample set as an input variable of the MAML element learning mathematical model according to parameters K, N and S in the MAML element learning mathematical model, and training the MAML element learning mathematical model.

As shown in FIG. 3, for guessing the task of asking you, the task is generated by training the question data of the current terminal in the customer service domain, the aggregated Embedding vector is used to represent the matching degree of the current terminal to question different topics in the customer service dialogue domain, and is used in the ranking of guessing the questions of you about to ask, according to the matching degree order.

Further, the method may further include: and acquiring the type of the task to which each piece of recommended information belongs, and controlling the current terminal to display each piece of recommended information and the type of the task to which each piece of recommended information belongs in an associated mode. That is, the task types of the recommended information can be displayed simultaneously while the recommended information is displayed, and if no accurate recommended information exists in the current recommended information list, the current terminal can also execute the operation of selecting other task types.

Accordingly, when training the machine learning model, the type of the task to which the recommendation information belongs should also be trained as a sample feature variable. Therefore, the above steps further include:

Analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking an information identifier of recommendation information or autonomous input information finally selected by a terminal and a type of a task to which the recommendation information belongs as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model; in the step of generating a recommendation information list according to the type of the current task and the historical operation data: and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model.

Example 3

Fig. 5 is a schematic hardware structure of an electronic device for performing an information recommendation method according to the present embodiment, where the device includes: one or more processors 501 and a memory 502, one processor 501 being illustrated in fig. 5. The apparatus for performing the information recommendation method may further include: an input device 503 and an output device 504. The processor 501, memory 502, input devices 503 and output devices 504 may be connected by a bus or otherwise, for example in fig. 5.

The memory 502, as a non-volatile computer readable storage medium, may be used to store non-volatile software programs, non-volatile computer executable programs, and modules. The processor 501 is capable of executing non-volatile software programs, instructions and modules stored in the memory 502 by: responding to the login information of the current terminal under the current task, and acquiring the current terminal identification; acquiring historical operation data under all task types corresponding to the current terminal identification; and generating a recommendation information list according to the type of the current task and the historical operation data and controlling the current terminal to display recommendation information according to the sequence of the recommendation information list. By the scheme, after the current terminal logs in under the current task, the historical operation of the current terminal under the current task is referred, and the recommendation information is generated according to the historical operation of the current terminal under all types of tasks, so that the accuracy of the recommendation information is improved.

Further, the processor 501 is further capable of: acquiring record data of the terminal selection recommendation information under each task type, wherein the record data comprises: the terminal has completed operation data under all task types; the terminal finally selects recommended information or information identification of autonomous input information; obtaining the corresponding relation between the completed operation data of the terminal under all task types and the information identification of the recommendation information or the autonomous input information finally selected by the terminal according to the recorded data; and obtaining the recommendation information list according to the corresponding relation, the type of the current task and the historical operation data. For each terminal, there may be an operation of selecting recommended information in a history period under various tasks, and since different selections may be made for history data that is not used when the terminal performs the corresponding operation, the records of all terminals in the history of the terminal are used as sample data to obtain a correspondence between a desired history operation record and the recommended information selection.

Preferably, the processor 501 is further capable of: analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking information identification of recommendation information or autonomous input information finally selected by a terminal as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model; and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model. The machine learning model can adopt various models which are mature and exist in the prior art, and after sample data are obtained, the sample data are directly utilized to carry out iterative training on the existing model.

Further, the processor 501 is further capable of: and acquiring the type of the task to which each piece of recommended information belongs, and controlling the current terminal to display each piece of recommended information and the type of the task to which each piece of recommended information belongs in an associated mode. Specifically, this can be achieved by: analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking an information identifier of recommendation information or autonomous input information finally selected by a terminal and a type of a task to which the recommendation information belongs as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model; in the step of generating a recommendation information list according to the type of the current task and the historical operation data: and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model. That is, the task types of the recommended information can be displayed simultaneously while the recommended information is displayed, and if no accurate recommended information exists in the current recommended information list, the current terminal can also execute the operation of selecting other task types.

Preferably, the machine learning model is a MAML element learning mathematical model, and the processor 501 is further capable of: setting parameters K, N and S in the MAML element learning mathematical model, wherein K represents the number of task types processed by the model at one time, N represents sample data under each task type, and S represents the total number of tasks processed by the model at each time; dividing a characteristic variable obtained after the analysis of the operation data is completed into a plurality of sample sets according to the type of the task; and selecting the corresponding characteristic variable in the sample set as an input variable of the MAML element learning mathematical model according to parameters K, N and S in the MAML element learning mathematical model, and training the MAML element learning mathematical model. MAML (Model Agnostic META LEARNING) is a meta learning technique, which is a deep learning algorithm, mainly used for learning how to learn. In popular terms, meta learning constructs a basic model first, and then designs a learning mechanism to make the basic model learn further on new tasks. Meta-learning is not directed to the results of learning, but rather to the process of learning. It learns not a mathematical model that is directly used for prediction, but "how faster and better to learn a mathematical model". Meta-learning enables classification using learned knowledge to new fields with fewer samples as possible. Embedding (embedded knowledge) technology has been widely used for representation learning in recent years, mainly for feature mining and feature representation of ID-class features, such as generating passenger taxi Embedding according to a passenger taxi type relationship, and generating driver Embedding according to a driver complaint condition. Because different learning tasks can generate different feature representations for ID class features, how to fuse embedding learned in different fields and train on the new task is a problem to be solved by the invention. In the embodiment of the invention, the types of tasks are divided by Embedding, and the association between different information features under different Embedding conditions is mined by MAML, so that the association between the selected recommended information under the current Embedding condition and the historical data of all Embedding conditions can be obtained.

Example 4

The present embodiment provides a storage medium readable and writable by a computer, the storage medium storing program instructions, the computer executing the information recommendation method of any one of the above after reading the program instructions.

The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method. Technical details not described in detail in this embodiment may be found in the methods provided in the embodiments of the present application.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (9)

1. An information recommendation method is characterized by comprising the following steps:

responding to the login information of the current terminal under the current task, and acquiring the current terminal identification;

Acquiring historical operation data under all task types corresponding to the current terminal identification, wherein the historical operation data comprises that a driver notes that a legacy article is on a vehicle for the current terminal;

Acquiring record data of the terminal selection recommendation information under each task type, wherein the record data comprises: the terminal has completed the operation data, the terminal finally selects the recommended information or the information identification of the autonomous input information under all task types;

Analyzing the completed operation data of the terminal under all task types to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking information identification of recommendation information or autonomous input information finally selected by the terminal as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model, wherein the characteristic variable comprises: vehicle type, get on the bus, get off the bus, drunk, evaluation and pay timeliness;

and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model, wherein the recommendation information list comprises recommendation information of the lost articles of passengers, and controlling the current terminal to display the recommendation information according to the sequence of the recommendation information list.

2. The information recommendation method according to claim 1, wherein the step of controlling the current terminal to display the recommendation information in the order of the recommendation information list further comprises:

And acquiring the type of the task to which each piece of recommended information belongs, and controlling the current terminal to display each piece of recommended information and the type of the task to which each piece of recommended information belongs in an associated mode.

3. The information recommendation method according to claim 2, wherein training of the machine learning model further comprises:

analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking an information identifier of recommendation information or autonomous input information finally selected by a terminal and a type of a task to which the information identifier belongs as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model.

4. The information recommendation method according to claim 3, wherein the machine learning model is a MAML element learning mathematical model, and the training step comprises:

Setting parameters K, N and S in the MAML element learning mathematical model, wherein K represents the number of task types processed by the model at one time, N represents sample data under each task type, and S represents the total number of tasks processed by the model at each time;

Dividing a characteristic variable obtained after the analysis of the operation data is completed into a plurality of sample sets according to the type of the task;

And selecting the corresponding characteristic variable in the sample set as an input variable of the MAML element learning mathematical model according to parameters K, N and S in the MAML element learning mathematical model, and training the MAML element learning mathematical model.

5. An information recommendation electronic device, comprising at least one processor and a memory communicatively coupled to at least one of the processors; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to:

responding to the login information of the current terminal under the current task, and acquiring the current terminal identification;

Acquiring historical operation data under all task types corresponding to the current terminal identification, wherein the historical operation data comprises that a driver notes that a legacy article is on a vehicle for the current terminal;

Acquiring record data of the terminal selection recommendation information under each task type, wherein the record data comprises: the terminal has completed the operation data, the terminal finally selects the recommended information or the information identification of the autonomous input information under all task types;

Analyzing the completed operation data of the terminal under all task types to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking information identification of recommendation information or autonomous input information finally selected by the terminal as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model, wherein the characteristic variable comprises: vehicle type, get on the bus, get off the bus, drunk, evaluation and pay timeliness;

and extracting the characteristic variable in the historical operation data as an input variable of the information recommendation model, and obtaining the recommendation information list according to the output of the information recommendation model, wherein the recommendation information list comprises recommendation information of the lost articles of passengers, and controlling the current terminal to display the recommendation information according to the sequence of the recommendation information list.

6. The information recommendation electronic device of claim 5, wherein said processor is further capable of:

And acquiring the type of the task to which each piece of recommended information belongs, and controlling the current terminal to display each piece of recommended information and the type of the task to which each piece of recommended information belongs in an associated mode.

7. The information recommendation electronic device of claim 6, wherein said processor is further capable of:

analyzing the completed operation data to obtain a characteristic variable, taking the characteristic variable as an input variable of a machine learning model, taking an information identifier of recommendation information or autonomous input information finally selected by a terminal and a type of a task to which the information identifier belongs as an output variable of the machine learning model, and training the machine learning model to obtain an information recommendation model.

8. The information recommendation electronic device of claim 7, wherein said machine learning model is a MAML element learning mathematical model, said processor being further capable of:

Setting parameters K, N and S in the MAML element learning mathematical model, wherein K represents the number of task types processed by the model at one time, N represents sample data under each task type, and S represents the total number of tasks processed by the model at each time;

Dividing a characteristic variable obtained after the analysis of the operation data is completed into a plurality of sample sets according to the type of the task;

And selecting the corresponding characteristic variable in the sample set as an input variable of the MAML element learning mathematical model according to parameters K, N and S in the MAML element learning mathematical model, and training the MAML element learning mathematical model.

9. A storage medium readable and writable by a computer, wherein the storage medium has program instructions stored therein, and the computer executes the information recommendation method according to any one of claims 1 to 4 after reading the program instructions.