CN114330885A

CN114330885A - Method, device and equipment for determining target state and storage medium

Info

Publication number: CN114330885A
Application number: CN202111642784.4A
Authority: CN
Inventors: 陈思翰; 王硕佳; 杨梦祺
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2021-12-29
Filing date: 2021-12-29
Publication date: 2022-04-12

Abstract

The application discloses a method, a device, equipment and a storage medium for determining a target state, and belongs to the technical field of computers. In the embodiment of the application, when the state of the target object is predicted, the target object information of the target object and the object information of the first objects are combined to determine the first similarities, and since the object information includes the attribute information and the state information, the first similarities can comprehensively represent the similarities between the target object and the first objects from the aspects of both the attribute and the state, and the accuracy of the first similarities is high. And then, a plurality of first probabilities can be accurately determined based on the occurrence frequency of the plurality of candidate states in the plurality of first objects and the plurality of first similarities, and a final target candidate state is determined through the plurality of first probabilities. The above process is independent of the judgment condition, can automatically adapt to the target object or the first object when the object information of the target object or the first object is changed, and has high accuracy of state determination.

Description

Method, device and equipment for determining target state and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a storage medium for determining a target state.

Background

With the development of computer technology, the functions provided by computer equipment are more and more abundant. In some cases, the computer device can predict the future state of the target object according to the historical state of the target object, for example, the computer device can predict the future possible faults of the machine according to the historical faults of the machine, the computer device can also predict the future possible diseases of the patient according to the historical diseases of the patient, and the like.

In the related art, when predicting the future state of the target object based on the historical state of the target object, the prediction is usually implemented based on some preset judgment conditions, that is, some judgment conditions are set according to the possible states, and in the prediction process, the historical state of the target object is compared with the set judgment conditions, so that the future state of the target object is finally obtained.

However, the accuracy of the state obtained in this way is closely related to the set judgment condition, and the judgment condition needs to be updated and increased frequently when the situation is complicated. Since the addition and update of the determination condition are often delayed, the accuracy of the state determined based on this method is low.

Disclosure of Invention

The embodiment of the application provides a method, a device, equipment and a storage medium for determining a target state, and can improve the accuracy of determining the target state. The technical scheme is as follows:

in one aspect, a method for determining a target state is provided, where the method includes:

acquiring object information of a plurality of first objects, wherein the object information comprises attribute information and state information;

determining a plurality of first similarities of a target object based on target object information of the target object and object information of the plurality of first objects, the first similarities representing similarities between the target object and the first objects;

determining a plurality of first probabilities of the target object based on frequency of occurrence of a plurality of candidate states in object information of the plurality of first objects and the plurality of first similarities, the first probabilities being probabilities that the target object will be in the candidate states;

and determining a target candidate state from the candidate states based on the first probabilities of the target object, wherein the target candidate state is a candidate state with a first probability meeting a target probability condition.

In one possible embodiment, the determining a plurality of first similarities of the target object based on the plurality of attribute similarities and the plurality of state similarities includes:

for any first object in the plurality of first objects, adopting a first weight and a second weight to fuse the attribute similarity and the state similarity between the target object and the first object to obtain a first similarity between the target object and the first object, wherein the first weight and the second weight are weights with cross loss meeting a target condition.

In a possible implementation, the determining, based on the frequency of occurrence of the candidate state in the plurality of first objects, the plurality of first similarities and the plurality of second similarities, a first probability that the target object will be in the candidate state includes:

dividing a first fusion similarity and a second fusion similarity to obtain a third fusion similarity, wherein the first fusion similarity is the sum of the second similarities, and the second fusion similarity is the sum of the first similarities;

and fusing the third fusion similarity and the occurrence frequency of the candidate state in the plurality of first objects to obtain a first probability that the target object is to be in the candidate state.

In a possible implementation, the determining a target candidate state from the plurality of candidate states based on the plurality of first probabilities of the target object includes any one of:

determining a candidate state with a highest first probability of previous target digits in the plurality of candidate states as the target candidate state;

determining a candidate state of the plurality of candidate states having a first probability greater than or equal to a probability threshold as the target candidate state.

In a possible implementation, after determining a target candidate state from the plurality of candidate states based on the plurality of first probabilities of the target object, the method further includes:

and sending alarm information under the condition that a target state exists in the target candidate states, wherein the alarm information is used for indicating that the target object is to be in the target state.

In a possible implementation, before the obtaining the object information of the plurality of first objects, the method further includes:

comparing the target state information with state information of a plurality of candidate objects;

and acquiring the candidate object as the first object under the condition that the target state information is matched with the state information of any candidate object.

In a possible implementation manner, the determining, based on the first probability and the plurality of second probabilities corresponding to the candidate state, the target probability that the target object is to be in the candidate state includes:

determining a highest probability of the first probability and the plurality of second probabilities as the target probability.

In one aspect, an apparatus for determining a target state is provided, the apparatus comprising:

the object information acquisition module is used for acquiring object information of a plurality of first objects, and the object information comprises attribute information and state information;

a first similarity determination module, configured to determine, based on target object information of a target object and object information of the plurality of first objects, a plurality of first similarities of the target object, where the first similarities are used to represent similarities between the target object and the first objects;

a first probability determination module, configured to determine a plurality of first probabilities of the target object based on frequency of occurrence of a plurality of candidate states in object information of the plurality of first objects and the plurality of first similarities, the first probability being a probability that the target object will be in the candidate states;

and the target candidate state determining module is used for determining a target candidate state from the candidate states based on a plurality of first probabilities of the target object, wherein the target candidate state is a candidate state with a first probability meeting a target probability condition.

In a possible implementation manner, the first similarity determining module is configured to determine a plurality of attribute similarities of the target object based on target attribute information in the target object information and the attribute information, where the attribute similarities are attribute similarities between the target object and the first object; determining a plurality of state similarities of the target object based on target state information and the state information in the target object information, wherein the state similarities are state similarities between the target object and the first object; determining a plurality of first similarities of the target object based on the plurality of attribute similarities and the plurality of state similarities.

In a possible implementation manner, the first similarity determining module is configured to determine, for any one of the plurality of first objects, a plurality of reference attribute similarities between the target object and the first object, where the reference attribute similarities are similarities between a plurality of attributes in the target attribute information and corresponding attributes in the attribute information of the first object; and fusing the multiple reference attribute similarities to obtain multiple attribute similarities of the target object.

In a possible implementation manner, the first similarity determination module is configured to generate a state matrix based on the target state information and the state information, where the state matrix is used to represent the target object and correspondence between the plurality of first objects and a plurality of states; based on the state matrix, a plurality of state similarities of the target object are determined.

In a possible implementation, the first similarity determination module is configured to generate an initial state matrix based on the target state information and the state information, columns of the initial state matrix are used for representing states, and rows of the initial state matrix are used for representing objects; and filling a first numerical value, a second numerical value and a third numerical value in the initial state matrix based on the target state information and the state information to obtain the state matrix, wherein the first numerical value is used for representing that the object and the state have a primary corresponding relationship, the second numerical value is used for representing that the object and the state have a secondary corresponding relationship, and the third numerical value is used for representing that the object and the state have a tertiary corresponding relationship.

In a possible implementation manner, the first similarity determining module is configured to determine, for any position in the initial state matrix, an object and a state corresponding to the position; in the case that the state information of the object indicates that the object has a primary correspondence with the state, filling the position with the first numerical value; filling the second numerical value in the position under the condition that the state information of the object indicates that the object and the state have a three-level corresponding relation and the object and a target state have a two-level corresponding relation, wherein the type of the target state is the same as that of the state; and filling the third numerical value in the position in the case that the state information of the object indicates that the object has a three-level correspondence with the state and the object also has a three-level correspondence with the target state.

In a possible implementation manner, the first similarity determining module is configured to determine, for any first object in the plurality of first objects, a first inverse frequency of each state in the state information of the first object in the state matrix, where the first inverse frequency is associated with the number of occurrences of each state in the state matrix and a total number of occurrences of all states in the state matrix; and determining the state similarity between the target object and the first object based on the state matrix, the first inverse frequency and a second inverse frequency, wherein the second inverse frequency is the inverse frequency of each state corresponding to the target object in the state matrix.

In a possible implementation manner, the first similarity determining module is configured to determine, for any state in the state information of the first object, a first number of occurrences of the state in the state matrix and a total number of occurrences of all states in the state matrix; dividing the total occurrence number by the first occurrence number to obtain an initial inverse frequency; and carrying out logarithmic operation on the initial inverse frequency to obtain the first inverse frequency.

In a possible implementation manner, the first similarity determining module is configured to obtain, from the state matrix, a numerical value corresponding to each state of the target object and a numerical value corresponding to each state of the first object; and fusing the first inverse frequency, the second inverse frequency, the numerical values corresponding to the states of the target object and the numerical values corresponding to the states of the first object to obtain the state similarity between the target object and the first object.

In a possible implementation manner, the first similarity determining module is configured to, for any first object in the plurality of first objects, fuse the attribute similarity and the state similarity between the target object and the first object by using a first weight and a second weight to obtain the first similarity between the target object and the first object, where the first weight and the second weight are weights with cross loss meeting a target condition.

In a possible implementation, the first probability determination module is configured to, for any one of the candidate states, obtain a plurality of second similarities from the plurality of first similarities, where the plurality of second similarities are similarities between the target object and a plurality of second objects, and the second objects are first objects in or once in the candidate state from among the plurality of first objects;

determining a first probability that the target object will be in the candidate state based on the frequency of occurrence of the candidate state in the plurality of first objects, the plurality of first similarities, and the plurality of second similarities.

In a possible implementation manner, the first probability determination module is configured to divide a first fused similarity and a second fused similarity to obtain a third fused similarity, where the first fused similarity is a sum of the plurality of second similarities, and the second fused similarity is a sum of the plurality of first similarities; and fusing the third fusion similarity and the occurrence frequency of the candidate state in the plurality of first objects to obtain a first probability that the target object is to be in the candidate state.

In one possible implementation, the target candidate state determination module is configured to perform any one of:

In a possible implementation manner, the apparatus further includes an alarm information sending module, configured to send alarm information in a case where a target state exists in the target candidate states, where the alarm information is used to indicate that the target object is to be in the target state.

In a possible implementation, the apparatus further comprises a first object determination module for comparing the target state information with state information of a plurality of candidate objects; and acquiring the candidate object as the first object under the condition that the target state information is matched with the state information of any candidate object.

In a possible implementation, the plurality of first objects belong to a first object group, the state information and target state information in the target object information both have a first state, the target candidate state determination module is configured to determine, for any candidate state, a target probability that the target object will be in the candidate state based on a first probability and a plurality of second probabilities corresponding to the candidate state, the plurality of second probabilities being probabilities that the target object will be in the candidate state determined based on a second object group corresponding to different states in the target state information; determining the target candidate state from the plurality of candidate states based on a plurality of target probabilities of the target object.

In one possible implementation, the target candidate state determination module determines the highest probability of the first probability and the plurality of second probabilities as the target probability.

In one aspect, a computer device is provided, the computer device comprising one or more processors and one or more memories having stored therein at least one computer program, the computer program being loaded and executed by the one or more processors to implement the method of determining the target state.

In one aspect, a computer-readable storage medium is provided, in which at least one computer program is stored, which is loaded and executed by a processor to implement the method of determining the target state.

In one aspect, a computer program product or a computer program is provided, which includes program code stored in a computer-readable storage medium, which is read by a processor of a computer device from the computer-readable storage medium, and which is executed by the processor to cause the computer device to execute the method for determining the target state.

According to the technical scheme provided by the embodiment of the application, when the state of the target object is predicted, the target object information of the target object and the object information of the first objects are combined to determine the first similarities, and the object information comprises the attribute information and the state information, so that the first similarities can comprehensively reflect the similarity between the target object and the first objects from the two aspects of attribute and state, and the accuracy of the first similarities is high. And then, a plurality of first probabilities can be accurately determined based on the occurrence frequency of the plurality of candidate states in the plurality of first objects and the plurality of first similarities, and a final target candidate state is determined through the plurality of first probabilities. The above process is independent of the judgment condition, can automatically adapt to the target object or the first object when the object information of the target object or the first object is changed, and has high accuracy of state determination.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a schematic diagram of an implementation environment of a method for determining a target state according to an embodiment of the present application;

fig. 2 is a flowchart of a method for determining a target state according to an embodiment of the present application;

FIG. 3 is a flow chart of a method for determining a target state according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram of a page provided by an embodiment of the present application;

FIG. 5 is a flow chart of a method for determining a target state according to an embodiment of the present disclosure;

FIG. 6 is a schematic structural diagram of a state prediction model provided in an embodiment of the present application;

fig. 7 is a schematic structural diagram of a target state determination device according to an embodiment of the present disclosure;

fig. 8 is a schematic structural diagram of a terminal according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a server according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

The terms "first," "second," and the like in this application are used for distinguishing between similar items and items that have substantially the same function or similar functionality, and it should be understood that "first," "second," and "nth" do not have any logical or temporal dependency or limitation on the number or order of execution.

The term "at least one" in this application means one or more, and the meaning of "a plurality" means two or more.

Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

Machine Learning (ML) is a multi-domain cross discipline, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge submodel to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning.

It is understood that in the specific implementation of the present application, related data such as attribute information and status information are involved, when the above embodiments of the present application are applied to specific products or technologies, user permission or consent needs to be obtained, and the collection, use and processing of related data need to comply with related laws and regulations and standards of related countries and regions.

Fig. 1 is a schematic diagram of an implementation environment of a method for determining a target state according to an embodiment of the present application, and referring to fig. 1, the implementation environment may include a terminal 110 and a server 140.

The terminal 110 is connected to the server 140 through a wireless network or a wired network. Optionally, the terminal 110 is a smartphone, a tablet, a laptop, a desktop computer, a smart watch, a vehicle-mounted terminal, etc., but is not limited thereto. The terminal 110 is installed and operated with an application program supporting the state determination.

The physical server of the server 140 may be a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, a cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Delivery Network (CDN), a big data and artificial intelligence platform, and the like. In some embodiments, server 140 provides background services for applications running on terminal 110.

Optionally, the terminal 110 generally refers to one of a plurality of terminals, and the embodiment of the present application is illustrated by the terminal 110.

Those skilled in the art will appreciate that the number of terminals described above may be greater or fewer. For example, the number of the terminal is only one, or several tens or hundreds, or more, and in this case, other terminals are also included in the implementation environment. The number of terminals and the type of the device are not limited in the embodiments of the present application.

After the description of the implementation environment of the embodiment of the present application, an application scenario of the embodiment of the present application will be described below with reference to the implementation environment, in the following description, a terminal is also referred to as the terminal 110, and a server is also referred to as the server 140.

The method for determining the target state provided by the embodiment of the application can be applied to various scenes in which the state is determined, for example, the method can be applied to a scene in which the disease of a patient is predicted, and the disease of the patient is also the state of the patient; or the method is applied to the scene of predicting the fault of the production machine, and the fault of the production machine is also the state of the production machine; or the method is applied to the scene of predicting the vehicle fault, namely the state of the vehicle.

Under the scene of predicting the disease of the patient, the target object is the patient, and the patient can upload object information through the terminal, wherein the object information is related information of the patient. In this application scenario, the attribute includes information such as the age, the location, and the occupation of the patient, and it should be noted that the acquisition and the use of the information such as the age, the location, and the occupation of the patient all need to be authorized by the patient. The status information is used to indicate the status of the patient, which in this application scenario includes the disease that the patient is currently suffering or has suffered from. After the terminal acquires the object information of the patient, object information of a plurality of first objects can be acquired, wherein the first objects are candidate patients related to the patient. The terminal determines a plurality of first similarities of the patient according to the object information of the patient and the object information of a plurality of candidate patients, wherein the first similarities are used for representing the similarities between the patient and the candidate patients. The terminal determines a plurality of first probabilities of the patient according to the frequency of occurrence of a plurality of candidate states in object information of a plurality of candidate patients and the plurality of first similarities, wherein the candidate states are candidate diseases, and the first probabilities are probabilities corresponding to the candidate diseases to which the patient will suffer. The terminal determines a target candidate state from the plurality of candidate states according to the plurality of first probabilities of the patient, that is, determines a target candidate disease from the plurality of candidate diseases, and the target candidate disease is a disease which the patient may have. In the above description, the terminal executes the method for determining the target state as an example, but in another possible embodiment, the method for determining the target state may be executed by the server. That is, after the terminal acquires the object information of the patient, the object information is sent to the server, and the server executes subsequent processing procedures, so as to determine the target candidate disease of the patient. In addition, the scenario of predicting patient disease can be further extended to two sub-scenarios of targeted physical examination or disease prevention and chronic complication prevention and control. In the targeted physical examination or disease prevention, a user can input basic information (attribute) and disease information (state) of the user through a terminal, the susceptible diseases can be obtained through the technical scheme provided by the embodiment of the application, the user can reserve physical examination items in a targeted mode according to the susceptible diseases, or the user can prevent the susceptible diseases in a targeted mode according to the output susceptible diseases. In the prevention and control of diseases and chronic complications, basic information (attributes) and disease information (states) can be well applied to the precise prevention and control of chronic diseases or chronic complications.

Under the scene of predicting the fault of the production machine, the target object is the production machine, and the technical personnel can upload object information through the terminal, wherein the object information is also related information of the production machine. The object information includes attribute information and state information of the production machine, the attribute information is used for indicating attributes of the production machine, and in this application scenario, the attributes include information such as usage duration, location, and machine type of the production machine. The status information is used to indicate the status of the production machine, and in this application scenario, the status includes the fault that the production machine currently exists or has existed. After acquiring the object information of the production machine, the terminal can acquire the object information of a plurality of first objects, wherein the first objects are candidate production machines related to the production machine. The terminal determines a plurality of first similarities of the production machine according to the object information of the production machine and the object information of a plurality of candidate production machines, wherein the first similarities are used for representing the similarities between the production machine and the candidate production machines. The terminal determines a plurality of first probabilities of the production machine according to the occurrence frequency of a plurality of candidate states in object information of a plurality of candidate production machines and the plurality of first similarities, wherein the candidate states are candidate faults, and the first probabilities are probabilities corresponding to the candidate faults to occur on the production machine. The terminal determines a target candidate state from the candidate states according to the first probabilities of the production machine, namely determines a target candidate fault from the candidate faults, wherein the target candidate fault is a fault to be generated by the production machine. In the above description, the terminal executes the method for determining the target state as an example, but in another possible embodiment, the method for determining the target state may be executed by the server. That is, after acquiring the object information of the production machine, the terminal sends the object information to the server, and the server executes subsequent processing procedures, thereby determining the target candidate fault of the production machine.

Under the scene of predicting vehicle faults, a target object is a vehicle, and technicians can upload object information through a terminal, wherein the object information is related information of the vehicle. The object information includes attribute information and state information of the vehicle, the attribute information is used for indicating attributes of the vehicle, and in the application scenario, the attributes include information such as the use duration, the region where the vehicle is located, and the type of the vehicle. The status information is used to indicate the status of the vehicle, and in this application scenario, the status includes a fault that the vehicle currently exists or has existed once. After the terminal acquires the object information of the vehicle, object information of a plurality of first objects, which are candidate vehicles related to the vehicle, can be acquired. The terminal determines a plurality of first similarities of the vehicle according to the object information of the vehicle and the object information of a plurality of candidate vehicles, wherein the first similarities are used for representing the similarities between the vehicle and the candidate vehicles. The terminal determines a plurality of first probabilities of the vehicle according to the frequency of occurrence of a plurality of candidate states in object information of a plurality of candidate vehicles and the plurality of first similarities, wherein the candidate states are candidate faults, and the first probabilities are probabilities corresponding to the candidate faults to occur in the vehicle. The terminal determines a target candidate state from the candidate states according to the first probabilities of the vehicle, namely determines a target candidate fault from the candidate faults, wherein the target candidate fault is a fault to be generated by the vehicle. In the above description, the terminal executes the method for determining the target state as an example, but in another possible embodiment, the method for determining the target state may be executed by the server. That is, after acquiring the object information of the vehicle, the terminal sends the object information to the server, and the server executes subsequent processing procedures, so as to determine the target candidate fault of the vehicle.

In addition, the above description has been given by taking three scenarios of predicting a patient disease, predicting a production machine failure, and predicting a vehicle failure as an example, and in other possible embodiments, the method for determining a target state provided in the embodiment of the present application can be applied to a scenario of predicting a state of another object, which is not limited in the embodiment of the present application.

After the implementation environment and the application scenario of the technical solution provided by the embodiment of the present application are introduced, the technical solution provided by the embodiment of the present application is introduced below with reference to the implementation environment and the application scenario, and with reference to fig. 2, taking an execution subject as an example, the method includes:

201. the terminal acquires object information of a plurality of first objects, the object information including attribute information and status information.

The first object is a candidate object, the object information is used for representing the characteristics of the first object, the attribute information is used for representing the attribute of the first object, and the state information is used for representing the state of the first object.

202. The terminal determines a plurality of first similarities of the target object based on the target object information of the target object and the object information of the plurality of first objects, the first similarities being used to represent the similarities between the target object and the first objects.

The target object is an object to be subjected to state determination, and has different meanings in different application scenes. Under the scene of predicting the disease of the patient, the target object is the patient to be subjected to disease prediction; under the scene of predicting the fault of the production machine, the target object is the production machine to be subjected to fault prediction; under the scene of predicting the vehicle fault, the target object is a vehicle to be subjected to fault prediction. The target object information is object information of the target object, and the target object information includes target attribute information and target state information of the target object. The first similarity determined based on the target object information and the object information of the plurality of first objects can comprehensively reflect the similarity of the target object and the first objects in the attributes and states.

203. The terminal determines a plurality of first probabilities of the target object based on the frequency of occurrence of a plurality of candidate states in the object information of the plurality of first objects and the plurality of first similarities, wherein the first probabilities are probabilities that the target object is to be in the candidate states.

Wherein the candidate states also have different meanings in different application scenarios. In the context of predicting a disease in a patient, the candidate state is a candidate disease; in the scene of predicting the fault of the production machine, the candidate state is the candidate fault of the production machine; in the scenario of predicting a vehicle failure, the candidate state is a candidate failure of the vehicle. The frequency of occurrence of the candidate state in the object information of the first object refers to the frequency of occurrence of the candidate state in the state information of the first object. In some embodiments, for any one of the plurality of candidate states, the frequency of occurrence of the candidate state is determined by: the number of first objects in the object information of the plurality of first objects in which the candidate state exists is divided by the number of the plurality of first objects. The first probability is a probability that the target object will be in the candidate state, i.e., a probability that the target object may be in the candidate state.

204. The terminal determines a target candidate state from the candidate states based on the first probabilities of the target object, wherein the target candidate state is a candidate state with the first probability meeting the target probability condition.

The target candidate state is also the predicted state.

The above step 201-204 is a simple introduction to the technical solution provided in the embodiment of the present application, and the technical solution provided in the embodiment of the present application will be more clearly described below with reference to some examples, and referring to fig. 3, taking an execution main body as an example, the method includes:

301. the terminal acquires target state information of a target object.

The target objects are all objects to be subjected to state determination, the target state information comprises target attribute information and target state information of the target objects, and the target objects are patients to be subjected to disease prediction in a scene of predicting patient diseases. The target attribute information includes identification information of the target object, for example, the target attribute information includes information such as the age, sex, region, and type of work of the patient, and the target attribute information is also referred to as basic information or personal information of the target object. The target state information includes historical state information and current state information of a target subject, and in the case where the target subject is a target patient, the target state information includes historical disease information and current disease information of the target patient, wherein diseases include physical diseases and mental diseases, and in this case, the target state information is also referred to as disease information of the target subject. Under the scene of predicting the vehicle fault, the target object is a vehicle to be subjected to fault prediction. The target attribute information includes identification information of a target object, for example, information including the age of the vehicle, the type of vehicle, the operation area, and the type of fuel used, and is also referred to as vehicle information. The target state information includes history state information and current state information of a target object, and in the case where the target object is a vehicle, the target state information includes history failure information and current failure information of the vehicle. In this case, the target state information is also referred to as failure information of the target object.

Of course, in other application scenarios, the target object and the target state information have different meanings, and in the following description, the application scenario will be taken as an example of predicting a disease of a patient, and the target object will be taken as a patient.

In one possible implementation, the terminal displays a status information acquisition page for acquiring target status information of the target object. In response to an operation on the state information acquisition page, the terminal acquires target state information of a target object.

In this embodiment, when the technical scheme provided by the embodiment of the application is applied, the terminal can quickly acquire the target state information of the target object through the state information acquisition page, and the efficiency of human-computer interaction is high.

The above embodiments will be explained below by two examples.

Example 1, a terminal runs a target application program, a user account of the target object is logged in the target application program, and the target application program is an application program with a determined support state. The terminal displays the main page of the target application program. In response to the operation on the main page, the terminal displays a state information acquisition page, wherein the state information acquisition page comprises an attribute information acquisition area, a state information acquisition area and an information confirmation control, the attribute information acquisition area is used for acquiring target attribute information of a target object, the state information acquisition area is used for acquiring target state information of the target object, and the information confirmation control is used for confirming the attribute information acquisition information and information in the state information acquisition area. And responding to the clicking operation of the information confirmation control, and acquiring target object information of a target object by the terminal, wherein the target object information comprises target attribute information in the attribute information acquisition area and target state information in the state information acquisition area.

That is to say, in this embodiment, when a user wants to perform disease prediction by using the technical solution provided in the embodiment of the present application, the target application can be started, and target object information is filled in through the information acquisition page of the target application, so that the terminal can perform subsequent steps based on the target object information.

For example, the terminal is a user device such as a mobile phone and a tablet used by a user, and the user may be a target object or a relative of the target object. And the terminal runs a target application program, and the user account of the target object is logged in the target application program. And the terminal displays a main page of the target application program, wherein the main page is used for introducing the function and the using method of the state determination, and the main page comprises an information acquisition control which is used for acquiring the target object information. And responding to the clicking operation of the information acquisition control, and switching the main page into an information acquisition page by the terminal, wherein the state information acquisition page comprises an attribute information acquisition area, a state information acquisition area and an information confirmation control. And responding to the clicking operation of the information confirmation control, and acquiring target object information of a target object by the terminal, wherein the target object information comprises target attribute information in the attribute information acquisition area and target state information in the state information acquisition area.

Referring to fig. 4, the terminal displays a home page 401 of the target application, and an information acquisition control 402 is displayed in the home page 401. In response to the click operation on the information acquisition control 402, the terminal switches the main page 401 to an information acquisition page 403, and the state information acquisition page 403 includes an attribute information acquisition area 404, a state information acquisition area 405, and an information confirmation control 406. In response to the click operation on the information confirmation control 406, the terminal acquires target object information of a target object, the target object information including target attribute information in the attribute information acquisition area 404 and target status information in the status information acquisition area 405.

And 2, the terminal runs a target application program, the user account of the target object is logged in the target application program, and the target application program is an application program with a determined support state. The terminal displays the main page of the target application program. And responding to the operation on the main page, and displaying a state information acquisition page by the terminal, wherein the state information acquisition page comprises an object information selection control. And responding to the clicking operation of the object information selection control, and displaying an object information display popup by the terminal, wherein a plurality of candidate object information are displayed in the object information display popup. In response to that the target object information in the candidate object information is selected, the terminal determines the candidate object information as the target object information, and the terminal acquires the target object information.

In other words, in this embodiment, the target application program is stored with a plurality of pieces of candidate object information in advance, and when the state determination is performed by using the technical solution provided by the embodiment of the present application, the user can quickly screen out the target object information from the plurality of pieces of candidate object information without inputting the target object information, so that the efficiency of human-computer interaction is high.

For example, the terminal is a user device such as a mobile phone and a tablet used by a user, and the user may be a target object or a relative of the target object. And the terminal runs a target application program, and the user account of the target object is logged in the target application program. And the terminal displays a main page of the target application program, wherein the main page is used for introducing the function and the using method of the state determination, and the main page comprises an information acquisition control which is used for acquiring the target object information. And responding to the clicking operation of the information acquisition control, and switching the main page into an information acquisition page by the terminal, wherein the state information acquisition page comprises an object information selection control. And responding to the clicking operation of the object information selection control, and displaying an object information display popup by the terminal, wherein a plurality of candidate object information are displayed in the object information display popup. In response to the click operation on any one of the candidate object information, the terminal determines the candidate object information as the target object information, and the terminal acquires the target object information.

In the process of describing step 301, the execution subject is taken as an example of a terminal, and in the scenario where step 301 is executed by the server, the server can obtain target object information of the target object from the terminal, that is, after the terminal obtains the target object information according to the above-described embodiment, the server sends the target object information to the server, and the server obtains the target object information. Or, the server may also directly obtain the target object information from a correspondingly maintained database, which is not limited in this embodiment of the present application.

302. The terminal compares the target state information of the target object with the state information of a plurality of candidate objects, and acquires the candidate object as the first object when the target state information is matched with the state information of any one candidate object.

In the scene of predicting the disease of the patient, the candidate object is a candidate patient, and in some embodiments, the candidate patient is a patient who has undergone state determination, that is, a patient who has used the technical scheme provided in the embodiments of the present application; or for registering the patient on a medical record, of course, registration of patient-related information may be performed after patient consent is obtained. The target state information includes a plurality of states of the target object, which refer to a disease that the patient has or currently has in the context of predicting a disease of the patient. The status information of the candidate includes a plurality of statuses of the candidate, and accordingly, the statuses also refer to diseases that the patient has or currently suffers from. In some embodiments, the status information further includes monitoring information of the candidate, such as heartbeat, blood pressure, blood sugar, and the like of the candidate, which is not limited in this application.

In one possible embodiment, the terminal determines the first state from the target state information. The terminal compares the first state with state information of a plurality of candidate objects. And under the condition that the state information of any candidate object comprises the first state, determining the candidate object as a first object. And repeating the steps to determine a plurality of first objects from the plurality of candidate objects. And the plurality of first objects determined by the terminal form a first object group, and the plurality of first objects in the first object group all correspond to the first state in the target state information.

In some embodiments, the terminal is further capable of determining a second object group, a third object group … …, an nth object group, where N is a positive integer, based on the second state in the target state information, each object group corresponding to a different state in the target state information, at the same time as determining the first object or after determining a plurality of first objects. In this way, the terminal can perform clustering based on the plurality of states in the target state information to obtain a plurality of object groups, and then can perform subsequent steps on the plurality of object groups. For example, assume that the disease set of the target patient a is D_aThe disease set D_aI.e. target status information of the target patient a. Assume that the disease set in the candidate patient is D_iThen for each

Patient clustering

k∈K＝|D_aI, wherein the disease set D_iI.e. status information of a plurality of target patients, d_aIs a set of diseases D_iIs a state in the target state information, C_kThe object group numbered k. After clustering, the terminal can perform subsequent steps on different object groups of the target patient a respectively to obtain a final result.

In this embodiment, the terminal is able to determine the first object based on the first state in the target state information, and the determined state information of the first object includes the first state. The mode is based on the clustering of the first state, so that the targets of subsequent processing are reduced, and the processing efficiency is improved.

In order to more clearly explain the technical solutions provided by the embodiments of the present application, in the following explanation process, the terminal is based on a plurality of first objects, that is, the first object group is processed as an example.

For example, in the context of predicting a patient's disease, the target status information includes a plurality of diseases that the target patient has or currently has, such as dyspepsia, acute bronchitis, emphysema, and caries. The terminal takes the first disease "dyspepsia" in the target state information as a first state, compares the first state "dyspepsia" with the state information of a plurality of candidate patients, and if the first state "dyspepsia" exists in the state information of any candidate patient, the candidate patient is determined as a first object, and the subsequent terminal can process the first object. The above steps are repeated to determine a plurality of first subjects from the plurality of candidate patients, and the state information of the plurality of first subjects comprises the first state of dyspepsia.

It should be noted that step 302 is described by taking an execution subject as an example, and in other possible embodiments, step 302 may be executed by a server, which is not limited in the embodiment of the present application.

303. The terminal acquires object information of a plurality of first objects, the object information including attribute information and status information.

The object information of the first object includes attribute information and state information, and in a scenario of predicting a disease of a patient, the attribute information includes identification information of the first object, for example, the attribute information includes information such as an age, a sex, a region where the first object is located, and a work type of the first object. In some embodiments, the attribute information further includes physical information, marital information, economic information, and the like of the first subject. The state information includes historical state information and current state information of the first object, and in the scenario, the state information includes historical disease information and current disease information of the first object.

In one possible implementation, the terminal obtains object information of the plurality of first objects based on the identifications of the plurality of first objects.

For example, the terminal sends an object information obtaining request to the server, where the object information obtaining request carries the identifiers of the plurality of first objects. After receiving the object information acquisition request, the server acquires the identifiers of the plurality of first objects from the object information acquisition request. The server acquires object information of the plurality of first objects based on the identifications of the plurality of first objects. The server sends the object information of the first objects to the terminal, and the terminal acquires the object information of the first objects.

Or, when the plurality of candidate object information are stored in the terminal, the terminal performs a query in the storage space based on the identifiers of the plurality of first objects to obtain the object information of the plurality of first objects.

It should be noted that step 303 is described by taking an execution subject as an example, and in other possible embodiments, step 303 may be executed by a server, which is not limited in the embodiments of the present application.

304. The terminal determines a plurality of attribute similarities of the target object based on the target attribute information and the attribute information in the target object information, wherein the attribute similarities are the attribute similarities between the target object and the first object.

Wherein the attribute information is attribute information of the first object. The attribute similarity refers to similarity between the target object and the first object in the attribute. In the scene of predicting the disease of the patient, the attribute information is the personal information of the patient, and the attribute similarity can reflect the previous similarity of the patient from the perspective of the personal information.

In a possible implementation manner, for any one of the plurality of first objects, the terminal determines a plurality of reference attribute similarities between the target object and the first object, where the reference attribute similarities are similarities between a plurality of attributes in the target attribute information and corresponding attributes in the attribute information of the first object. And the terminal fuses the multiple reference attribute similarities to obtain the multiple attribute similarities of the target object. And the terminal repeatedly executes the steps to obtain the attribute similarity between the target object and the plurality of first objects.

In this embodiment, the terminal can determine the similarity between the target object and the first object in the attribute according to the target attribute information and the attribute information of the first object, and compare each attribute in the target attribute information and the attribute information of the first object in the determination process, thereby obtaining more accurate attribute similarity.

For example, for any attribute in the target attribute information, the terminal obtains a reference attribute similarity between the attribute and a corresponding attribute in the attribute information of the first object. And the terminal repeatedly executes the steps and acquires a plurality of reference similarities. And the terminal performs weighted summation on the multiple similarity degrees to obtain multiple attribute similarity degrees of the target object. For example, the terminal represents the attribute in the form of a vector, and for any attribute vector in the target attribute information, the terminal obtains a cosine similarity between the attribute vector and a corresponding attribute vector in the attribute information of the first object, where the cosine similarity is also a reference attribute similarity. And the terminal performs weighted summation on the multiple similarity degrees to obtain multiple attribute similarity degrees of the target object.

Taking the example that the attribute information includes age (age), sex (sex), region (region), and type of work (job) in the context of predicting a patient disease, the terminal can determine the attribute similarity between the target object and the first object based on the following formula (1).

w_info(a，i)＝μ₁*d(sex_a，sex_i)+μ₂*d(age_a，age_i)+μ_.*d(region_a，region_i)+μ₄*d(job_a，job_i) (1)

Wherein, w_infoRepresenting the similarity of the attributes between the target object and the first object, a representing the target object, representing the first object, sex_aA gender vector, sex, representing the target object_iA gender vector, age, representing the first subject_aAge vector representing target object_iAn age vector, region, representing the first object_aRegion vector, region, representing the target object_iA region vector, joba, representing the first object_aJob type vector, jobA, representing target object_iRepresenting the work type vector of the first object, d (x, y) representing determining the cosine similarity between vector x and vector y, u₁、u₂、u₃And u₄Weights corresponding to different attributes, wherein₁+μ₂+μ₃+μ₄＝1；0≤μ₁，μ₂，μ₃，μ₄≤1，u₁、u₂、u₃And u₄The numerical value of (a) is set by a skilled person under the condition of satisfying the above constraint condition, and the embodiment of the present application does not limit this.

That is, through the above formula 1, the terminal determines the reference attribute similarity between the age of the target object and the age of the first object, the reference attribute similarity between the gender of the target object and the gender of the first object, the reference attribute similarity between the region where the target object is located and the region where the first object is located, and the reference attribute similarity between the work type of the target object and the work type of the first object, respectively, in the process of determining the attribute similarity between the target object and the first object. And the terminal performs weighted summation on the four reference attribute similarities to obtain the attribute similarity between the target object and the first object.

It should be noted that step 304 is described by taking an execution subject as an example, and in other possible embodiments, step 304 may be executed by a server, which is not limited in the embodiment of the present application.

305. The terminal determines a plurality of state similarities of the target object based on the target state information and the state information in the target object information, wherein the state similarities are state similarities between the target object and the first object.

The state information is the state information of a plurality of first objects. In the case of predicting a patient's disease, the target state information is disease information of the target patient, and the state information is disease information of the plurality of first objects. The disease information includes physical disease information and mental disease information. The state similarity can reflect the similarity of the target object and the first object in suffering from the disease.

In a possible implementation manner, the terminal generates a state matrix based on the target state information and the state information, wherein the state matrix is used for representing the target object and the corresponding relation between the plurality of first objects and the plurality of states. And the terminal determines a plurality of state similarities of the target object based on the state matrix.

In this embodiment, the terminal is capable of generating a state matrix describing the correspondence between the target object and the plurality of first objects and the plurality of states based on the target state information and the state information, and determining the correspondence between the target object and the plurality of first objects and the disease in a scenario of predicting a disease of the patient, that is, based on the target state information and the plurality of state information. The similarity in disease between the target object and the first object can subsequently be determined based on the state matrix.

In order to more clearly explain the above embodiment, the above embodiment will be explained in two parts.

The first part, the terminal, generates a state matrix based on the target state information and the state information.

In one possible embodiment, the terminal generates an initial state matrix based on the target state information and the state information, the columns of the initial state matrix being used to represent the state and the rows of the initial state matrix being used to represent the object. And the terminal fills a first numerical value, a second numerical value and a third numerical value in the initial state matrix based on the target state information and the state information to obtain the state matrix, wherein the first numerical value is used for representing that the object and the state have a primary corresponding relationship, the second numerical value is used for representing that the object and the state have a secondary corresponding relationship, and the third numerical value is used for representing that the object and the state have a tertiary corresponding relationship.

The initial state matrix is a blank matrix, and the process of generating the initial state matrix by the terminal is also a process of defining the meaning of the rows and columns of the initial state information. Since the columns of the initial state matrix are used to represent states, the number of columns of the initial state matrix, i.e., the target state information and the total number of states in the state information, corresponds to one state per column of the initial state matrix. In the scenario of predicting a disease of a patient, the number of columns of the initial state matrix is the target state information and the number of diseases in the state information, and each column of the initial state information corresponds to a disease. Since the rows of the initial state matrix are used to represent the objects, the number of rows of the initial state matrix, i.e. the total number of the target object and the plurality of first objects, corresponds to one object per row in the initial state matrix. In the context of predicting patient disease, the number of rows of the initial state matrix, i.e., the number of target patients and first patients, each row in the initial state information corresponds to one patient. The primary corresponding relation represents that the object is or is in a corresponding state; the secondary correspondence represents that the object may be in a corresponding state; having a three level correspondence indicates that it is not in a corresponding state. The first numerical value, the second numerical value and the third numerical value are set by a technician according to an actual situation, and the embodiment of the present application does not limit this. In some embodiments, the first-level correspondence is a direct correspondence and the second-level correspondence is an indirect correspondence. For example, when a patient has a cough, for example, then the patient has a direct correspondence with the cough, and a first value may be assigned; since cough is a respiratory disease, the patient has an indirect correspondence with all diseases that are respiratory diseases and can be assigned a second value; the third-level correspondence relationship is no correspondence relationship, and a third numerical value can be directly given. In some embodiments, the primary correspondence is also referred to as a dominant correspondence, the secondary correspondence is also referred to as a recessive correspondence, and the tertiary correspondence is also referred to as a no correspondence.

In this embodiment, the terminal can distinguish the corresponding relationship between the object and the state by filling different values, and the obtained state matrix can reflect the corresponding relationship between the object and the state as a whole, which is helpful for improving the efficiency and effect of the subsequent processing.

For example, for any position in the initial state matrix, the terminal determines an object and a state corresponding to the position, where the object may be a target object or a first object. And in the case that the state information of the object indicates that the object has a primary corresponding relationship with the state, the terminal fills the first value in the position. And in the case that the state information of the object indicates that the object and the state have a three-level corresponding relationship and the object and the target state have a two-level corresponding relationship, the terminal fills the second numerical value in the position, and the type of the target state is the same as that of the state. And in the case that the state information of the object indicates that the object has a three-level correspondence with the state, and the object also has a three-level correspondence with the target state, the terminal fills the third value in the position.

In some embodiments, the terminal determines the type of the state based on a classification table, wherein the classification table describes a correspondence between the state and the type. In some embodiments, there is a hierarchical relationship between types of states, that is, one large class may correspond to multiple small classes whose corresponding states are the same type of state. In the scenario of predicting the disease of the patient, the patient is an International Classification of Diseases (ICD), such as ICD-9, ICD-10 or other versions of the Classification table, which is not limited in this embodiment of the application. In the above example, each position in the state matrix is used to represent a correspondence between an object and a state. That is, the indication that the state information of the object indicates that the object has a primary correspondence with the state in the above description means that the patient corresponding to the position has a disease corresponding to the position at one time or at the present time; the state information of the object indicates that the object and the state have a tertiary corresponding relationship, and the object and the target state have a secondary corresponding relationship, namely that the patient corresponding to the position does not have the disease corresponding to the position, but the patient has the target disease at one time or at present, and the type of the target disease is the same as the type of the disease corresponding to the position; the state information of the object indicates that the object and the state have three-level correspondence, that is, the patient corresponding to the position does not have the disease corresponding to the position, and the patient does not have the target disease. In some embodiments, the state matrix is also referred to as a collaborative filtering matrix. Table 1 shows the form of the state matrix, see table 1.

TABLE 1

The number is the classification number, patient a is the target patient, i.e. the target object, and patient i is the first patient numbered i, i.e. the first object numbered i. In the above table, the first value is 1, indicating that the patient has or is suffering from the disease; a second value of 0.3, indicating that the patient has not suffered from the disease, but has suffered from the same type of disease as the disease; a third value of 0 indicates that the patient has not suffered from the disease, nor has the same type of disease as the disease. For example, if patient a has acute pharyngitis, the location of patient a corresponding to acute pharyngitis is filled with the first value of 1. Patient a did not have acute sinusitis, but acute sinusitis was the same type of disease as acute pharyngitis, and the location of patient a corresponding to acute sinusitis was filled with the second value of 0.3.

And a second part, the terminal determines a plurality of state similarities of the target object based on the state matrix.

In a possible implementation manner, for any one of the plurality of first objects, the terminal determines a first inverse frequency of each state in the state matrix in the state information of the first object, where the first inverse frequency is associated with the number of occurrences of each state in the state matrix and the total number of occurrences of all states in the state matrix. The terminal determines the state similarity between the target object and the first object based on the state matrix, the first inverse frequency and a second inverse frequency, wherein the second inverse frequency is the inverse frequency of each state corresponding to the target object in the state matrix.

The Inverse Frequency (IDF) can reflect the importance degree of an element in an element set, and the more the occurrence number of an element in the element set, the lower the importance degree of the element is represented; the fewer occurrences of an element in a set of elements, the higher the importance of the element. In the above embodiment, the first inverse frequency is used to represent the degree of importance of the state in the state matrix.

In the above embodiment, the description is given taking an example of determining the state similarity between the target object and any one of the plurality of first objects, and the terminal can repeatedly execute the above steps to obtain a plurality of state similarities between the target object and the plurality of first objects.

A. The terminal determines a first inverse frequency of each state in the state matrix in the state information of the first object.

In a possible implementation, for any state in the state information of the first object, the terminal determines a first number of occurrences of the state in the state matrix and a total number of occurrences of all states in the state matrix. And the terminal divides the total occurrence times and the first occurrence times to obtain an initial inverse frequency. And the terminal carries out logarithm operation on the initial inverse frequency to obtain the first inverse frequency.

The first occurrence number of the state in the state matrix is the number of the state corresponding to the first value, and in the case of predicting the disease of the patient, the number of the target object and the first objects which have or currently have the disease is the number of the target object and the first objects. The total occurrence number of all states in the state matrix refers to the number of the first numerical values in the state matrix, and in the context of predicting the disease of the patient, refers to the total number of the diseases that the target object and the plurality of first objects have or currently have. For example, referring to Table 1 above, for disease 401, essential hypertension, the first number of occurrences of the disease in the status matrix is 1, i.e., only patient i has or is currently suffering from the disease. The total number of occurrences of all diseases in the state matrix is 5, i.e., the number of the first numerical value in the state matrix, then the initial inverse frequency of the disease is 5, and the first inverse frequency of the disease is log 5.

In some embodiments, the terminal can determine the first inverse frequency of the state by equation (2) below.

Wherein f is_jIs the first inverse frequency of the state numbered j, n is the total number of occurrences of all states in the state matrix, n_jThe first number of occurrences of the state numbered j.

B. The terminal determines a state similarity between the target object and the first object based on the state matrix, the first inverse frequency and the second inverse frequency.

In a possible implementation manner, the terminal obtains, from the state matrix, a value corresponding to each state of the target object and a value corresponding to each state of the first object. And the terminal fuses the first inverse frequency, the second inverse frequency, the numerical values corresponding to the states of the target object and the numerical values corresponding to the states of the first object to obtain the state similarity between the target object and the first object.

The numerical value corresponding to each state is the first numerical value, the second numerical value or the third numerical value in the above description.

For example, the terminal obtains values corresponding to the states of the target object and values corresponding to the states of the first object from the state matrix. The terminal determines a plurality of first fusion parameters based on the numerical values corresponding to the states of the target object and the second inverse frequencies of the states. The terminal determines a plurality of second fusion parameters based on the numerical values corresponding to the states of the first object and the first inverse frequencies of the states. And the terminal accumulates the plurality of first fusion parameters and the plurality of second fusion parameters to obtain the state similarity between the target object and the first object.

In order to more clearly describe the technical solution provided by the embodiment of the present application, a method for determining a first convergence parameter by a terminal is described below.

In a possible implementation manner, the terminal obtains values corresponding to a plurality of states of the target object in the state matrix. For any state in the plurality of states, the product of the value corresponding to the state and the first inverse frequency of the state is used as a first reference parameter. And the terminal takes the product square sum of the numerical value corresponding to each state and the first inverse frequency of each state as a second reference parameter. And the terminal accumulates the plurality of second reference parameters to obtain a third reference parameter. And the terminal determines the ratio of the first reference parameter to the third reference parameter as the first fusion parameter.

It should be noted that the method for determining the first fusion parameter and the second fusion parameter in other states by the terminal is the same as the above embodiment, and the implementation process is not described again.

A method for the terminal to acquire the state similarity between the target object and the first object is described below by formula (3).

Wherein, w_disease(a, i) is the state similarity between the target object a and the first object i, f_jIs the first inverse frequency, v, of state j_a·jIs the value of the target object and the state j in the state matrix, v_i·jIs the value of the first object and the state j in the state matrix, f_kIs in state kIs of a first inverse frequency, v_a·kIs the value of the target object and the state k in the state matrix, upsilon_i·kIs the value of the first object and the state k in the state matrix, J_aSet of states in the target state information for the target object, J_iIs a set of states in the state information for the first object.

In the scene of predicting the disease of the patient, because the difference between the diseases is large, when the similarity of the diseases among the patients is determined, the similarity of the patients taking the rare diseases into consideration is large, and the rare diseases are reflected by the inverse frequency of the diseases.

It should be noted that step 305 is described by taking an execution subject as an example, and in other possible embodiments, step 305 may be executed by a server, which is not limited in the embodiments of the present application.

306. The terminal determines a plurality of first similarities of the target object based on the plurality of attribute similarities and the plurality of state similarities.

Wherein the first similarity reflects the similarity of the target object and the first object in both attributes and states.

In a possible implementation manner, for any one of the plurality of first objects, the terminal uses a first weight and a second weight to fuse the attribute similarity and the state similarity between the target object and the first object, so as to obtain a first similarity between the target object and the first object, where the first weight and the second weight are weights whose cross loss meets a target condition.

Wherein, the first weight and the second weight cross loss meeting the target condition refers to the first weight and the second weight corresponding to the minimum loss in cross validation.

For example, the terminal obtains the first similarity between the target object and the first object by the following formula (4).

w(a，i)＝α₁w_info(a，i)+a₂w_disease(O，i) (4)

Wherein, w_info(a，i)Is the similarity of the attributes between the target object a and the first object i, w_disease(a, i) is the state similarity between the target object a and the first object i, which is α₁Is a first weight, α₂Is a second weight, α₁+α₂＝1；≤α₁，α₂≤1。

It should be noted that step 306 is described by taking an execution subject as an example, and in other possible embodiments, step 306 may be executed by a server, which is not limited in the embodiment of the present application.

307. The terminal determines a plurality of first probabilities of the target object based on the frequency of occurrence of a plurality of candidate states in the object information of the plurality of first objects and the plurality of first similarities, wherein the first probabilities are probabilities that the target object is to be in the candidate states.

In a possible implementation manner, for any candidate state in the plurality of candidate states, the terminal obtains a plurality of second similarities from the plurality of first similarities, where the plurality of second similarities are similarities between the target object and a plurality of second objects, and the second object is a first object in or once in the candidate state among the plurality of first objects. The terminal determines a first probability that the target object will be in the candidate state based on the frequency of occurrence of the candidate state in the plurality of first objects, the plurality of first similarities and the plurality of second similarities.

The first part and the terminal acquire a plurality of second similarities from the plurality of first similarities.

In one possible embodiment, the terminal determines the plurality of second objects from the plurality of first objects based on the state information of the plurality of first objects and the candidate state. The terminal obtains a plurality of second similarities from the plurality of first similarities based on the plurality of second objects.

And a second part, determining a first probability that the target object is to be in the candidate state by the terminal based on the occurrence frequency of the candidate state in the plurality of first objects, the plurality of first similarities and the plurality of second similarities.

In a possible implementation manner, the terminal divides a first fusion similarity by a second fusion similarity to obtain a third fusion similarity, where the first fusion similarity is a sum of the plurality of second similarities, and the second fusion similarity is a sum of the plurality of first similarities. And the terminal fuses the third fusion similarity and the occurrence frequency of the candidate state in the plurality of first objects to obtain a first probability that the target object is to be in the candidate state.

For example, the terminal determines the first probability that the target object will be in the candidate state by the following formula (5).

Wherein p is_c(a, j) refers to a first probability that the target object a will be in the candidate state j,

the probability of the candidate state j in the first objects is also referred to as the probability of the candidate state j,

is the sum of a plurality of second similarity degrees,

is the sum of a plurality of first similarities.

It should be noted that step 307 is described by taking an execution subject as an example, and in other possible embodiments, step 307 may be executed by a server, which is not limited in the embodiments of the present application.

308. The terminal determines a target candidate state from the candidate states based on the first probabilities of the target object, wherein the target candidate state is a candidate state with the first probability meeting the target probability condition.

In one possible implementation, the terminal determines the candidate state with the highest first probability of the previous target bits from the plurality of candidate states as the target candidate state.

The target number of bits is set by a skilled person according to actual situations, for example, set to 5, 10, or other values, which is not limited in this embodiment of the application.

In this embodiment, the terminal can determine, as the target candidate state, the candidate state of the previous target bit number with the highest first probability in the plurality of candidate states, which is also the candidate state with the higher probability that the target object may be in, so that the user can process the target candidate state in time.

In one possible implementation, the terminal determines a candidate state of the plurality of candidate states whose first probability is greater than or equal to a probability threshold as the target candidate state.

The probability threshold is set by a technician according to an actual situation, for example, set to 0.7, 0.8, or other values, which is not limited in this embodiment of the application.

In this embodiment, the terminal can determine the target candidate state from the candidate states with the first probability greater than or equal to the probability threshold, and thus can output all the candidate states with higher probability, thereby facilitating the user to process the target candidate state in time.

It should be noted that, the above is described by taking the example of determining the target candidate state based on a plurality of first objects, and the plurality of first objects may be regarded as a first object group, and in other possible embodiments, the terminal may further determine the target candidate state based on the plurality of object groups, so as to improve the accuracy of the target candidate state, and the method is as follows:

in a possible implementation manner, for any candidate state, the terminal determines a target probability that the target object will be in the candidate state based on a first probability and a plurality of second probabilities corresponding to the candidate state, where the plurality of second probabilities are probabilities that the target object will be in the candidate state determined based on a second object group, and the second object group and the first object group correspond to different states in the target state information. The terminal determines the target candidate state from the candidate states based on a plurality of target probabilities of the target object.

The first probability is determined based on the target object and the plurality of first objects, and the plurality of second probabilities are determined based on other objects in the candidate objects and the target object, so that the implementation process belongs to the same inventive concept, and is not repeated herein. Other objects in the candidate object form a second object group, and the number of the second object group is one or more, which is not limited in the embodiment of the present application.

The first part and the terminal determine a target probability that the target object is to be in the candidate state based on a first probability and a plurality of second probabilities corresponding to the candidate state.

In one possible embodiment, the terminal determines the highest probability of the first probability and the plurality of second probabilities as the target probability.

In some embodiments, the terminal determines the target probability by equation (6) below.

Wherein p (a, j) is the target probability,

to be based on object groups_Ck is the probability between the target object a and the state j, k is the number of the object group, and the probability is the first probability when the object group is the first object group; in the case where the object group is a second object group, the probability is a second probability. D_aThe state of the target state information indication for the target objectAnd (5) state collection.

And a second step of determining the target candidate state from the candidate states by the terminal based on a plurality of target probabilities of the target object.

In a possible implementation manner, the terminal determines the candidate state of the front target bit number with the highest target probability in the plurality of candidate states as the target candidate state.

In one possible implementation, the terminal determines a candidate state with a target probability greater than or equal to a probability threshold among the plurality of candidate states as the target candidate state.

It should be noted that step 308 is described by taking an execution subject as an example, and in other possible embodiments, step 308 may be executed by a server, which is not limited in the embodiment of the present application.

In some embodiments, after step 308, the terminal is further capable of performing any of the following:

in one possible embodiment, in the case that there is a target state in the target candidate states, the terminal transmits alarm information indicating that the target object is to be in the target state.

Wherein, the target state is the state set by the technician according to the actual situation. Under the scene of predicting the disease of the patient, the target state is a preset disease, and under the condition that the determined target candidate state comprises the preset disease, the terminal sends alarm information. This embodiment can be used in the case of infectious disease prevention, and accordingly the predetermined disease is also an infectious disease. For example, referring to table 2, after the subject information of the target patient is input, the terminal can output a disease list of the target patient. Under the condition that the preset infectious diseases exist in the disease list, the terminal sends alarm information to medical care personnel or disease control personnel, so that the medical care personnel or the disease control personnel can conveniently process the information.

TABLE 2

Object information	List of diseases
		Male, 17 years old, Shenzhen, student	Influenza virus
Dyspepsia	Bronchiectasis disease
		Acute bronchitis	Asthma (asthma)
Pulmonary emphysema	Common cold
		Dental caries	Typhoid fever

Wherein, because the preset infectious disease flu exists in the disease list, the terminal sends alarm information which is used for indicating that the patient is about to suffer from the flu.

In one possible implementation, the terminal displays the target candidate status.

In the implementation mode, the terminal can display the determined target candidate state, and under the scene of predicting the disease of the patient, the user can determine the disease possibly suffered by the target object by checking the target candidate state, so that prevention can be performed in time; in the scene of disease prevention, a user can know possible diseases by looking at the target candidate state, so that physical examination is performed in a targeted manner, and the target candidate state is protected in advance.

For example, referring to fig. 5, a user inputs target object information 501 including target attribute information and target state information on a terminal, wherein the target attribute information includes gender, age, location area, and the like. The target status information includes historical diseases, current diseases, and the like. The terminal inputs the target attribute information into the disease prediction model 502, and the disease prediction model 502 is used for executing the steps 301-308. The terminal displays the target candidate state output by the disease prediction model in the disease list 503. The user can perform targeted protection and examination according to the disease list 503.

In a possible implementation, the above step 301-308 is performed by a state prediction model, referring to fig. 6, the state prediction model 600 includes a clustering module 601, a collaborative filtering module 602, and an aggregation module 603. The clustering module 601 is configured to determine a plurality of object groups according to the target state information of the target object and the state information of the plurality of other objects. The collaborative filtering module 602 is configured to determine a state matrix of each object group according to the target attribute information and the attribute information of the plurality of other objects, where the state matrix is also referred to as a collaborative filtering town. The collaborative filtering module 602 is further configured to determine a similarity between the target object and the other objects based on the collaborative filtering matrix. The collaborative filtering module 602 is further configured to determine a first probability based on the state information and the similarity. The aggregation module 603 is configured to determine a target probability and a target candidate state according to information obtained by the clustering module 601 and the collaborative filtering module 602.

In the experimental process, the effect of the state prediction model 600 was evaluated by using several evaluation indexes, and the data set used was a MIMICIV data set, which contains 60000 patient data and 898 diseases in total. That is, the experimental process is performed based on the scenario of disease prediction. The evaluation index and the evaluation result will be described below.

The evaluation indexes comprise prediction diversity, prediction coverage rate, prediction accuracy rate, average rank and half-life accuracy rate.

1. Diversity is used to describe the diversity of predicted disease categories, and the formula is as follows:

diversity ═ N_{disease_predict}；

Wherein N is_{disease_predict}The number of disease categories output for the model.

2. Coverage is used to describe the comprehensiveness of predicting accurate disease categories, and the formula is as follows:

predicted Coverage ratio Coverage N_{disease_true}/N_disease；

Wherein N is_{disease_true}To predict the exact number of disease types, N_diseaseThe number of disease types to be predicted.

3. The accuracy is used for describing the credibility of the prediction result, and the formula is as follows:

accuracy N_true/N_total；

Wherein N is_trueTo predict the exact number of diseases, N_totalIs the number of diseases to be predicted.

4. The average rank is used to describe the ranking of actual diseases in the predicted diseases, and the higher the ranking of actual diseases, the better the effect. The formula is as follows:

average Rank (Σ Rank)_true)/N_true；

Wherein, Rank_trueFor ranking of actual diseases in prediction, N_trueTo predict the exact number of diseases.

The evaluation results are shown in Table 3.

TABLE 3

In the experimental process, a traditional collaborative filtering matrix and an ICARE (aid decision) model are selected for comparison, wherein the ICARE is an aggregation and collaborative filtering model. Experimental results show that the effect of the state prediction model provided by the application is first in multiple indexes, and the diversity and coverage rate are improved remarkably.

All the above optional technical solutions may be combined arbitrarily to form optional embodiments of the present application, and are not described herein again.

In the context of predicting diseases, the embodiments of the present application provide future possible disease predictions (target candidate states) including partial infectious diseases, chronic diseases, etc. using the disease information (state information) of patients and the social demographic information (attribute information). The prediction can assist doctors in identifying new diseases, help patients to prevent diseases, or help patients to select examination items in a targeted manner, so that the comprehensiveness and pertinence of physical examination are improved. The technical scheme provided by the embodiment of the application predicts the possible future diseases of the patient through the existing data, and can provide more references for disease prevention and control.

Fig. 7 is a schematic structural diagram of a device for determining a target state according to an embodiment of the present application, and referring to fig. 7, the device includes: an object information acquisition module 701, a first similarity determination module 702, a first probability determination module 703, and a target candidate state determination module 704.

An object information obtaining module 701, configured to obtain object information of a plurality of first objects, where the object information includes attribute information and status information.

A first similarity determination module 702, configured to determine a plurality of first similarities of a target object based on target object information of the target object and object information of the plurality of first objects, where the first similarities are used to represent similarities between the target object and the first objects.

A first probability determination module 703, configured to determine, based on the frequency of occurrence of multiple candidate states in the object information of the multiple first objects and the multiple first similarities, multiple first probabilities of the target object, where the first probabilities are probabilities that the target object will be in the candidate states.

A target candidate state determining module 704, configured to determine a target candidate state from the plurality of candidate states based on the plurality of first probabilities of the target object, where the target candidate state is a candidate state with a first probability that meets a target probability condition.

In a possible implementation manner, the first similarity determining module 702 is configured to determine a plurality of attribute similarities of the target object based on the target attribute information in the target object information and the attribute information, where the attribute similarities are attribute similarities between the target object and the first object. And determining a plurality of state similarities of the target object based on the target state information and the state information in the target object information, wherein the state similarities are the state similarities between the target object and the first object. And determining a plurality of first similarity of the target object based on the attribute similarities and the state similarities.

In a possible implementation manner, the first similarity determining module 702 is configured to determine, for any first object in the plurality of first objects, a plurality of reference attribute similarities between the target object and the first object, where the reference attribute similarities are similarities between a plurality of attributes in the target attribute information and corresponding attributes in the attribute information of the first object. And fusing the multiple reference attribute similarities to obtain multiple attribute similarities of the target object.

In a possible implementation, the first similarity determining module 702 is configured to generate a state matrix based on the target state information and the state information, where the state matrix is used to represent the target object and the corresponding relationship between the plurality of first objects and the plurality of states. Based on the state matrix, a plurality of state similarities of the target object are determined.

In a possible implementation, the first similarity determination module 702 is configured to generate an initial state matrix based on the target state information and the state information, columns of the initial state matrix are used for representing states, and rows of the initial state matrix are used for representing objects. And filling a first numerical value, a second numerical value and a third numerical value in the initial state matrix based on the target state information and the state information to obtain the state matrix, wherein the first numerical value is used for representing that the object and the state have a primary corresponding relationship, the second numerical value is used for representing that the object and the state have a secondary corresponding relationship, and the third numerical value is used for representing that the object and the state have a tertiary corresponding relationship.

In a possible implementation manner, the first similarity determining module 702 is configured to determine, for any position in the initial state matrix, an object and a state corresponding to the position. And filling the position with the first numerical value under the condition that the state information of the object indicates that the object has a primary corresponding relation with the state. And filling the second numerical value in the position under the condition that the state information of the object indicates that the object and the state have a three-level corresponding relationship and the object and a target state have a two-level corresponding relationship, wherein the type of the target state is the same as that of the state. And filling the third numerical value in the position in the case that the state information of the object indicates that the object has three-level correspondence with the state and the object also has three-level correspondence with the target state.

In a possible implementation manner, the first similarity determining module 702 is configured to determine, for any first object in the plurality of first objects, a first inverse frequency of each state in the state information of the first object in the state matrix, where the first inverse frequency is associated with the number of occurrences of each state in the state matrix and a total number of occurrences of all states in the state matrix. And determining the state similarity between the target object and the first object based on the state matrix, the first inverse frequency and a second inverse frequency, wherein the second inverse frequency is the inverse frequency of each state corresponding to the target object in the state matrix.

In a possible implementation manner, the first similarity determining module 702 is configured to determine, for any state in the state information of the first object, a first occurrence number of the state in the state matrix and a total occurrence number of all states in the state matrix. And dividing the total occurrence number by the first occurrence number to obtain an initial inverse frequency. And carrying out logarithm operation on the initial inverse frequency to obtain the first inverse frequency.

In a possible implementation manner, the first similarity determining module 702 is configured to obtain, from the state matrix, values corresponding to the states of the target object and values corresponding to the states of the first object. And fusing the first inverse frequency, the second inverse frequency, the numerical values corresponding to the states of the target object and the numerical values corresponding to the states of the first object to obtain the state similarity between the target object and the first object.

In a possible implementation manner, the first similarity determining module 702 is configured to, for any first object in the plurality of first objects, fuse the attribute similarity and the state similarity between the target object and the first object by using a first weight and a second weight to obtain the first similarity between the target object and the first object, where the first weight and the second weight are weights for which a cross loss meets a target condition.

In a possible implementation manner, the first probability determining module 703 is configured to, for any candidate state in the plurality of candidate states, obtain a plurality of second similarities from the plurality of first similarities, where the plurality of second similarities are similarities between the target object and a plurality of second objects, and the second object is a first object in or once in the candidate state in the plurality of first objects.

Determining a first probability that the target object will be in the candidate state based on the frequency of occurrence of the candidate state in the plurality of first objects, the plurality of first similarities and the plurality of second similarities.

In a possible implementation manner, the first probability determining module 703 is configured to divide a first fused similarity and a second fused similarity to obtain a third fused similarity, where the first fused similarity is a sum of the plurality of second similarities, and the second fused similarity is a sum of the plurality of first similarities. And fusing the third fusion similarity and the occurrence frequency of the candidate state in the plurality of first objects to obtain a first probability that the target object is to be in the candidate state.

In a possible implementation, the target candidate state determination module 704 is configured to perform any one of the following:

and determining the candidate state with the highest first probability target digit in the plurality of candidate states as the target candidate state.

In a possible implementation manner, the apparatus further includes an alarm information sending module, configured to send alarm information, in a case where a target state exists in the target candidate states, where the alarm information is used to indicate that the target object is to be in the target state.

In a possible implementation, the apparatus further comprises a first object determination module for comparing the target state information with state information of a plurality of candidate objects. And acquiring the candidate object as the first object when the target state information is matched with the state information of any candidate object.

In a possible implementation, the plurality of first objects belong to a first object group, the state information and the target state information in the target object information both have a first state, and the target candidate state determination module 704 is configured to determine, for any candidate state, a target probability that the target object will be in the candidate state based on a first probability corresponding to the candidate state and a plurality of second probabilities determined based on a second object group corresponding to different states in the target state information. The target candidate state is determined from the plurality of candidate states based on a plurality of target probabilities of the target object.

In one possible implementation, the target candidate state determination module 704 determines the highest probability of the first probability and the plurality of second probabilities as the target probability.

It should be noted that: the determining apparatus for determining a target state provided in the above embodiment is only illustrated by the division of the above functional modules when determining the target state, and in practical applications, the above function distribution may be completed by different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the above described functions. In addition, the target state determining apparatus provided in the foregoing embodiments and the target state determining method embodiment belong to the same concept, and specific implementation processes thereof are described in the method embodiment and are not described herein again.

An embodiment of the present application provides a computer device, configured to perform the foregoing method, where the computer device may be implemented as a terminal or a server, and a structure of the terminal is introduced below:

fig. 8 is a schematic structural diagram of a terminal according to an embodiment of the present application. The terminal 800 may be: a smartphone, a tablet, a laptop, or a desktop computer. The terminal 800 may also be referred to by other names such as user equipment, portable terminal, laptop terminal, desktop terminal, etc.

In general, the terminal 800 includes: one or more processors 801 and one or more memories 802.

The processor 801 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and so forth. The processor 801 may be implemented in at least one hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 801 may also include a main processor and a coprocessor, where the main processor is a processor for Processing data in an awake state, and is also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 801 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the display screen. In some embodiments, the processor 801 may further include an AI (Artificial Intelligence) processor for processing computing operations related to machine learning.

Memory 802 may include one or more computer-readable storage media, which may be non-transitory. Memory 802 may also include high speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in the memory 802 is used to store at least one computer program for execution by the processor 801 to implement the method of determining a target state provided by the method embodiments of the present application.

In some embodiments, the terminal 800 may further include: a peripheral interface 803 and at least one peripheral. The processor 801, memory 802 and peripheral interface 803 may be connected by bus or signal lines. Various peripheral devices may be connected to peripheral interface 803 by a bus, signal line, or circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 804, a display screen 805, a camera assembly 806, an audio circuit 807, a positioning assembly 808, and a power supply 809.

The peripheral interface 803 may be used to connect at least one peripheral related to I/O (Input/Output) to the processor 801 and the memory 802. In some embodiments, the processor 801, memory 802, and peripheral interface 803 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 801, the memory 802, and the peripheral interface 803 may be implemented on separate chips or circuit boards, which are not limited by this embodiment.

The Radio Frequency circuit 804 is used for receiving and transmitting RF (Radio Frequency) signals, also called electromagnetic signals. The radio frequency circuitry 804 communicates with communication networks and other communication devices via electromagnetic signals. The rf circuit 804 converts an electrical signal into an electromagnetic signal to be transmitted, or converts a received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 804 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a subscriber identity module card, and so forth.

The display screen 805 is used to display a UI (User Interface). The UI may include graphics, text, icons, video, and any combination thereof. When the display 805 is a touch display, the display 805 also has the ability to capture touch signals on or above the surface of the display 805. The touch signal may be input to the processor 801 as a control signal for processing. At this point, the display 805 may also be used to provide virtual buttons and/or a virtual keyboard, also referred to as soft buttons and/or a soft keyboard.

The camera assembly 806 is used to capture images or video. Optionally, camera assembly 806 includes a front camera and a rear camera. Generally, a front camera is disposed at a front panel of the terminal, and a rear camera is disposed at a rear surface of the terminal.

The audio circuit 807 may include a microphone and a speaker. The microphone is used for collecting sound waves of a user and the environment, converting the sound waves into electric signals, and inputting the electric signals to the processor 801 for processing or inputting the electric signals to the radio frequency circuit 804 to realize voice communication.

The positioning component 808 is used to locate the current geographic position of the terminal 800 for navigation or LBS (Location Based Service).

Power supply 809 is used to provide power to various components in terminal 800. The power supply 809 can be ac, dc, disposable or rechargeable.

In some embodiments, terminal 800 also includes one or more sensors 810. The one or more sensors 810 include, but are not limited to: acceleration sensor 811, gyro sensor 812, pressure sensor 813, optical sensor 814, and proximity sensor 815.

The acceleration sensor 811 may detect the magnitude of acceleration in three coordinate axes of the coordinate system established with the terminal 800.

The gyro sensor 812 may acquire a 3D motion of the user with respect to the terminal 800 in cooperation with the acceleration sensor 811.

Pressure sensors 813 may be disposed on the side frames of terminal 800 and/or underneath display 805. When the pressure sensor 813 is disposed on the side frame of the terminal 800, the holding signal of the user to the terminal 800 can be detected, and the processor 801 performs left-right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 813. When the pressure sensor 813 is disposed at a lower layer of the display screen 805, the processor 801 controls the operability control on the UI interface according to the pressure operation of the user on the display screen 805.

The optical sensor 814 is used to collect the ambient light intensity. In one embodiment, the processor 801 may control the display brightness of the display 805 based on the ambient light intensity collected by the optical sensor 814.

The proximity sensor 815 is used to collect the distance between the user and the front surface of the terminal 800.

Those skilled in the art will appreciate that the configuration shown in fig. 8 is not intended to be limiting of terminal 800 and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components may be used.

The computer device may also be implemented as a server, and the following describes a structure of the server:

fig. 9 is a schematic structural diagram of a server provided in this embodiment of the present application, where the server 900 may generate a relatively large difference due to a difference in configuration or performance, and may include one or more processors (CPUs) 901 and one or more memories 902, where the one or more memories 902 store at least one computer program, and the at least one computer program is loaded and executed by the one or more processors 901 to implement the methods provided by the foregoing method embodiments. Certainly, the server 900 may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface, so as to perform input and output, and the server 900 may also include other components for implementing device functions, which are not described herein again.

In an exemplary embodiment, a computer-readable storage medium, such as a memory including a computer program, which is executable by a processor to perform the method of determining a target state in the above embodiments, is also provided. For example, the computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a Compact Disc Read-Only Memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

In an exemplary embodiment, a computer program product or a computer program is also provided, which includes program code stored in a computer-readable storage medium, which is read by a processor of a computer apparatus from the computer-readable storage medium, and which is executed by the processor to cause the computer apparatus to execute the above-described method of determining the target state.

In some embodiments, the computer program according to the embodiments of the present application may be deployed to be executed on one computer device or on multiple computer devices located at one site, or may be executed on multiple computer devices distributed at multiple sites and interconnected by a communication network, and the multiple computer devices distributed at the multiple sites and interconnected by the communication network may constitute a block chain system.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only exemplary of the present application and should not be taken as limiting, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for determining a target state, the method comprising:

2. The method of claim 1, wherein determining the plurality of first similarities for the target object based on the target object information for the target object and the object information for the plurality of first objects comprises:

determining a plurality of attribute similarities of the target object based on target attribute information in the target object information and the attribute information, wherein the attribute similarities are attribute similarities between the target object and the first object;

determining a plurality of state similarities of the target object based on target state information and the state information in the target object information, wherein the state similarities are state similarities between the target object and the first object;

determining a plurality of first similarities of the target object based on the plurality of attribute similarities and the plurality of state similarities.

3. The method of claim 2, wherein the determining a plurality of attribute similarities for the target object based on the attribute information and target attribute information in the target object information comprises:

for any one of the plurality of first objects, determining a plurality of reference attribute similarities between the target object and the first object, where the reference attribute similarities are similarities between a plurality of attributes in the target attribute information and corresponding attributes in the attribute information of the first object;

and fusing the multiple reference attribute similarities to obtain multiple attribute similarities of the target object.

4. The method of claim 2, wherein the determining a plurality of state similarities for the target object based on the target state information and the state information in the target object information comprises:

generating a state matrix based on the target state information and the state information, wherein the state matrix is used for representing the target object and the corresponding relation between the plurality of first objects and a plurality of states;

based on the state matrix, a plurality of state similarities of the target object are determined.

5. The method of claim 4, wherein generating a state matrix based on the target state information and the state information comprises:

generating an initial state matrix based on the target state information and the state information, columns of the initial state matrix being used to represent states, rows of the initial state matrix being used to represent objects;

and filling a first numerical value, a second numerical value and a third numerical value in the initial state matrix based on the target state information and the state information to obtain the state matrix, wherein the first numerical value is used for representing that the object and the state have a primary corresponding relationship, the second numerical value is used for representing that the object and the state have a secondary corresponding relationship, and the third numerical value is used for representing that the object and the state have a tertiary corresponding relationship.

6. The method of claim 5, wherein the populating the initial state matrix with a first value, a second value, and a third value based on the target state information and the state information to obtain the state matrix comprises:

for any position in the initial state matrix, determining an object and a state corresponding to the position;

in the case that the state information of the object indicates that the object has a primary correspondence with the state, filling the position with the first numerical value;

filling the second numerical value in the position under the condition that the state information of the object indicates that the object and the state have a three-level corresponding relation and the object and a target state have a two-level corresponding relation, wherein the type of the target state is the same as that of the state;

and filling the third numerical value in the position in the case that the state information of the object indicates that the object has a three-level correspondence with the state and the object also has a three-level correspondence with the target state.

7. The method of claim 4, wherein determining the plurality of state similarities for the target object based on the state matrix comprises:

for any first object in the plurality of first objects, determining a first inverse frequency of each state in the state matrix in the state information of the first object, wherein the first inverse frequency is associated with the occurrence number of each state in the state matrix and the total occurrence number of all states in the state matrix;

and determining the state similarity between the target object and the first object based on the state matrix, the first inverse frequency and a second inverse frequency, wherein the second inverse frequency is the inverse frequency of each state corresponding to the target object in the state matrix.

8. The method of claim 7, wherein determining a first inverse frequency of each state in the state matrix in the state information of the first object comprises:

for any state in the state information of the first object, determining a first occurrence number of the state in the state matrix and a total occurrence number of all states in the state matrix;

dividing the total occurrence number by the first occurrence number to obtain an initial inverse frequency;

and carrying out logarithmic operation on the initial inverse frequency to obtain the first inverse frequency.

9. The method of claim 7, wherein the determining the state similarity between the target object and the first object based on the state matrix, the first inverse frequency, and a second inverse frequency comprises:

acquiring numerical values corresponding to all states of the target object and numerical values corresponding to all states of the first object from the state matrix;

and fusing the first inverse frequency, the second inverse frequency, the numerical values corresponding to the states of the target object and the numerical values corresponding to the states of the first object to obtain the state similarity between the target object and the first object.

10. The method of claim 1, wherein the determining a plurality of first probabilities of the target object based on the frequency of occurrence of the plurality of candidate states in the object information of the plurality of first objects and the plurality of first similarities comprises:

for any candidate state in the candidate states, obtaining a plurality of second similarities from the plurality of first similarities, where the plurality of second similarities are similarities between the target object and a plurality of second objects, and the second objects are first objects in or once in the candidate state among the plurality of first objects;

11. The method of claim 1, wherein the plurality of first objects belong to a first object group, wherein the state information and target state information in the target object information each have a first state, and wherein determining a target candidate state from the plurality of candidate states based on a plurality of first probabilities of the target object comprises:

for any candidate state, determining a target probability that the target object will be in the candidate state based on a first probability and a plurality of second probabilities corresponding to the candidate states, wherein the plurality of second probabilities are determined based on a second object group, and the second object group and the first object group correspond to different states in the target state information;

determining the target candidate state from the plurality of candidate states based on a plurality of target probabilities of the target object.

12. An apparatus for determining a target state, the apparatus comprising:

13. A computer device, characterized in that the computer device comprises one or more processors and one or more memories in which at least one computer program is stored, the computer program being loaded and executed by the one or more processors to implement the method of determining a target state according to any one of claims 1 to 11.

14. A computer-readable storage medium, in which at least one computer program is stored, which is loaded and executed by a processor to implement the method of determining a target state according to any one of claims 1 to 11.

15. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the method of determining a target state of any one of claims 1 to 11.