Method and device for intention recognition, storage medium and electronic device

Abstract

The present disclosure provides a method, device, electronic device and storage medium for intent recognition, and relates to the field of computer technology. The method includes: receiving a user information set to be identified sent by a user terminal; parsing the user information set to be identified, obtaining user service data, user terminal data and user behavior time data collected by the user terminal, and obtaining the receiving time point of receiving the user information to be identified; performing intent recognition for the target intent according to the user service data, user terminal data, user behavior time data and receiving time point, and obtaining the recognition result value of the target intent; determining the degree to which the user information set to be identified represents the target intent according to the recognition result value. The method can identify the degree to which the user information set to be identified represents the target intent from multiple dimensions based on the user information set to be identified obtained from the user terminal.

Description

Background Art

[0002] In telecommunication operations, channel agents often use card maintenance to forge user card usage records in order to profit from operators. That is, they use some technical means to forge the illusion that users are using phone cards. For example, through some programs or card maintenance equipment, multiple phone cards can be attached to the same terminal to generate records of text messages, calls, and traffic usage, but in fact these phone cards are not used by normal users. A large number of forged user card usage records will cause an increase in operator customers, resulting in increased operating and activity costs but a decrease in operator business revenue, thus causing losses to the operator. Therefore, it is necessary to identify the intention of card maintenance.

[0003] In the related technology, whether there is an intention to maintain the card is usually determined by analyzing the user's business data, which has the problem of inaccurate identification.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the present disclosure is to provide a method, device, electronic device and storage medium for intention recognition to solve the problem of inaccurate recognition of card maintenance intention.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a method for intent identification is provided, including: receiving a user information set to be identified sent by a user terminal; parsing the user information set to be identified, obtaining user business data, user terminal data and user behavior time data collected by the user terminal, and obtaining a receiving time point of receiving the user information to be identified; performing intent identification for a target intent based on the user business data, user terminal data, user behavior time data and receiving time point, and obtaining an identification result value for the target intent; determining the extent to which the user information set to be identified represents the target intent based on the identification result value.

[0008] In one embodiment of the present disclosure, the target intention is the card maintenance intention; the step of performing intent recognition for the target intention according to the user business data, the user terminal data, the user behavior time data and the receiving time point to obtain the recognition result value of the target intention includes: performing business feature recognition for the target intention according to the user business data to obtain the business recognition result value; performing terminal environment recognition for the target intention according to the user terminal data to obtain the terminal recognition result value; performing data authenticity recognition according to the user behavior time data and the receiving time point to obtain the data recognition result value; respectively obtaining the weight values ​​corresponding to the business recognition value, the terminal recognition value and the data recognition value; determining the recognition result value according to the business recognition value, the terminal recognition value, the data recognition value and the corresponding weight values.

[0009] In one embodiment of the present disclosure, user service data includes at least one of the following: user card number registration place, minimum consumption service type, package service type, monthly service type, average income value per user, calling duration, called duration, consumption value in the current month of network access, consumption value in the next month of network access, and user belonging channel type; business feature identification for target intention is performed according to user service data, and the step of obtaining the service identification result value includes: obtaining a card maintenance identification model, inputting the user service data into the card maintenance identification model; obtaining the card maintenance intention probability output by the card maintenance identification model; and using the card maintenance intention probability as the service identification result value; wherein, the card maintenance identification model is trained based on the historical service data of card maintenance users and the historical service data of non-card maintenance users.

[0010] In one embodiment of the present disclosure, the step of performing terminal environment identification for target intent according to user terminal data and obtaining a terminal identification result value includes: determining the number of user terminal data that meet the card maintenance environment conditions; determining the terminal identification result value according to the number; wherein the card maintenance environment conditions include: no international mobile equipment identity code on the terminal, no local number on the terminal, no user identification card number on the terminal, no brand type information on the terminal, no CPU information on the terminal, no Mac address information on the terminal, no battery level change information on the terminal, no baseband information on the terminal, the control group information cannot be read on the terminal, the process group information cannot be read on the terminal, there is no wlan driver information on the terminal, the wlan driver information on the terminal is abnormal, the number of sensors on the terminal is less than the sensor number threshold, and the number of pre-installed applications on the terminal is less than the application number threshold.

[0011] In one embodiment of the present disclosure, before the step of receiving the user information set to be identified collected through the user terminal, the method also includes: configuring a collection instruction; configuring a collection trigger condition in the collection instruction; sending the collection instruction to the user terminal, so that when the user terminal triggers the collection trigger condition, the user service data, user terminal data and user behavior time data are collected according to the collection instruction, thereby generating the user information set to be identified.

[0012] In one embodiment of the present disclosure, the collection instruction also configures the target business data type to be collected, the target terminal data type to be collected, and the user behavior time acquisition strategy; and, the user business data is obtained by collecting the user terminal according to the target business data type; the user terminal data is obtained by collecting the user terminal according to the target terminal data type; and the user behavior time data is obtained by the user terminal according to the user behavior time acquisition strategy.

[0013] In one embodiment of the present disclosure, an offset time period, a hash algorithm and a number of hash calculations are configured in a user behavior time acquisition strategy; and user behavior time data includes: an offset time period, a behavior occurrence time point of a target behavior determined based on the offset time period, and a calculation time period spent on iteratively calculating designated data using a hash algorithm according to the number of hash calculations; wherein the user behavior is determined as follows: a first moment that triggers a collection trigger condition is determined, a moment determined according to an offset time period before the first moment is taken as a second moment, and the most recent user behavior before the second moment is determined as a target behavior; the target behavior includes at least one of a mouse click, a keyboard operation, a gesture behavior and a gyroscope behavior; the designated data includes: at least one of user business data, user terminal data and behavior log data of the target behavior.

[0014] In one embodiment of the present disclosure, the steps of performing data authenticity identification based on user behavior time data and a receiving time point to obtain a data identification result value include: calculating a behavior prediction occurrence time point based on the receiving time point, a calculation time period, and an offset time period; determining a difference between the behavior prediction occurrence time point and the behavior occurrence time point, and using the difference as a behavior time difference; determining a true probability of the user behavior time data based on a relationship between the behavior time difference and a time difference threshold, as a data authenticity identification result; and determining a data identification result value based on the data authenticity identification result.

[0015] According to another aspect of the present disclosure, a device for intent recognition is provided, including: a receiving module, used to receive a user information set to be identified sent by a user terminal; a parsing module, used to parse the user information set to be identified, obtain user business data, user terminal data and user behavior time data collected by the user terminal, and obtain a receiving time point for receiving the user information to be identified; an identification module, used to perform intent recognition for a target intent based on the user business data, user terminal data, user behavior time data and the receiving time point, and obtain a recognition result value for the target intent; and a determination module, used to determine the extent to which the user information set to be identified represents the target intent based on the recognition result value.

[0016] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the method for intent recognition described above is implemented.

[0017] According to another aspect of the present disclosure, there is provided an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the above-mentioned method for intent recognition by executing the executable instructions.

[0018] The method for intent identification provided by the embodiments of the present disclosure can identify the degree to which the user information set to be identified represents the target intent from multiple dimensions such as user business data, user terminal data and user behavior time data based on the user information set to be identified obtained from the user terminal, thereby achieving better target intent identification effect.

[0019] Furthermore, the method for intent identification provided by the embodiments of the present disclosure can also configure the hash algorithm and the number of hash calculations in the collection instructions when collecting the information set of user information to be identified, so that the terminals used by normal users are not affected while the card-maintaining terminals with the intention of maintaining the card have their computing power continuously consumed, thereby achieving the effect of making the card-maintaining behavior incur a cost.

[0020] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0022] Figure 1 A schematic diagram showing an exemplary system architecture to which the method for intent recognition according to an embodiment of the present disclosure can be applied;

[0023] Figure 2 A flowchart of a method for intent recognition according to an embodiment of the present disclosure is shown;

[0024] Figure 3 A flowchart of obtaining a recognition result value in a method for intent recognition according to an embodiment of the present disclosure is shown;

[0025] Figure 4 A schematic diagram showing iterative calculation in a method for intent recognition according to an embodiment of the present disclosure is shown;

[0026] Figure 5 A schematic diagram showing a method for determining a recognition result value in an intention recognition method according to an embodiment of the present disclosure is shown;

[0027] Figure 6 A flowchart of a method for intent recognition according to an embodiment of the present disclosure is shown;

[0028] Figure 7 A block diagram showing an apparatus for intent recognition according to an embodiment of the present disclosure; and

[0029] Figure 8 A structural block diagram of a computer device for intent recognition in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0030] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0031] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0032] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present disclosure, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0033] In view of the technical problems existing in the above-mentioned related technologies, the embodiments of the present disclosure provide a method for intent recognition, so as to solve at least one or all of the above-mentioned technical problems.

[0034] Figure 1 A schematic diagram showing an exemplary system architecture to which the method for intent recognition according to an embodiment of the present disclosure can be applied; Figure 1 As shown:

[0035] The system architecture may include a server 101, a network 102, and a client 103. The network 102 is used to provide a medium for a communication link between the client 103 and the server 101. The network 102 may include various connection types, such as wired, wireless communication links, or optical fiber cables.

[0036] The server 101 may be a server that provides various services, such as a background management server that provides support for identification processing of data obtained from the client 203. The background management server may send a collection instruction to the client 103, receive and parse the user information set to be identified returned by the client 103, and perform identification based on the parsed data.

[0037] In some optional embodiments, the server 101 can receive a set of user information to be identified sent by a user terminal; the server 101 can parse the set of user information to be identified, obtain user business data, user terminal data and user behavior time data collected by the user terminal, and obtain the receiving time point of receiving the user information to be identified; the server 101 can perform intent recognition for the target intention based on the user business data, user terminal data, user behavior time data and the receiving time point, and obtain an recognition result value for the target intention; the server 101 can determine the extent to which the user information set to be identified represents the target intention based on the recognition result value.

[0038] The client 103 may be a mobile terminal such as a mobile phone, a game console, a tablet computer, an e-book reader, smart glasses, smart home devices, AR (Augmented Reality) devices, VR (Virtual Reality) devices, etc., or the client 103 may also be a personal computer, such as a laptop computer and a desktop computer.

[0039] The client 103 can receive the collection instructions sent by the server 101, and collect data on the terminal based on the collection instructions. It can also perform hash calculations on the terminal based on the collection instructions, so that when a terminal is used as a card maintenance terminal by a large number of phone cards, computing power is consumed on the terminal, generating card maintenance costs.

[0040] It should be understood that Figure 1 The number of clients, networks and servers is merely illustrative, and the server 101 may be a physical server, a server cluster consisting of multiple servers, or a cloud server. It may have any number of clients, networks and servers according to actual needs.

[0041] Below, each step of the method for intent recognition in the exemplary embodiment of the present disclosure will be described in more detail with reference to the accompanying drawings and embodiments.

[0042] Figure 2 A flow chart of a method for intention recognition according to an embodiment of the present disclosure is shown. The method provided by the embodiment of the present disclosure can be performed as follows: Figure 1 The process is executed in the server or client shown in the figure, but the present disclosure is not limited to this.

[0043] In the following examples, the server cluster 101 is used as the execution subject for example description.

[0044] like Figure 2 As shown, the method for intent recognition provided by the embodiment of the present disclosure may include the following steps:

[0045] Step S201, receiving a user information set to be identified sent by a user terminal;

[0046] Step S203, parsing the user information set to be identified, obtaining user service data, user terminal data and user behavior time data collected by the user terminal, and obtaining the receiving time point of the user information to be identified;

[0047] Step S205, performing intent recognition for the target intent according to the user service data, the user terminal data, the user behavior time data and the receiving time point, and obtaining a recognition result value for the target intent;

[0048] Step S207, determining the degree to which the to-be-identified user information set represents the target intention according to the recognition result value.

[0049] The method for intent identification provided by the present disclosure can obtain a user information set to be identified from a user terminal, and parse the user information set to be identified to obtain user business data, user terminal data and user behavior time data, and then use the user business data, user terminal data and user behavior time data in multiple dimensions to identify the degree to which the user information set to be identified represents the target intent, thereby achieving a better target intent identification effect.

[0050] Next, we will combine Figure 2 The steps S201 to S207 of the method for intention recognition in this exemplary embodiment are described in more detail.

[0051] Step S201: receiving a user information set to be identified sent by a user terminal.

[0052] The user terminal may first collect data for generating the user information set to be identified locally, so as to generate the user information set to be identified according to the collected data, and then the user terminal may send the user information set to be identified to the server.

[0053] In some embodiments, before step S201, the following steps may also be included: configuring collection instructions, wherein collection trigger conditions may be pre-configured in the collection instructions; sending collection instructions to the user terminal so that when the collection trigger conditions are triggered, the user terminal collects user business data, user terminal data and user behavior time data according to the collection instructions, thereby generating a user information set to be identified.

[0054] Specifically, the above collection trigger condition can be a specified trigger behavior, such as "dialing behavior", or a specified time, such as "3:00 p.m. every day", so that when the terminal detects that the collection trigger condition is met, the above collection instruction is executed.

[0055] In some embodiments, the collection instruction also configures the target business data type to be collected, the target terminal data type to be collected, and the user behavior time acquisition strategy; and, the user business data is obtained by collecting the target business data type through the user terminal; the user terminal data is obtained by collecting the target terminal data type through the user terminal; and the user behavior time data is obtained by the user terminal according to the user behavior time acquisition strategy.

[0056] Specifically, user service data may include service information such as the user card number registration location, minimum consumption service type, package service type, etc., and user terminal data may include information on power changes on the terminal, application quantity information, terminal category information, etc. Both user service data and user terminal data can be searched and obtained from the local storage space of the terminal.

[0057] Step S203, parse the user information set to be identified, obtain the user service data, user terminal data and user behavior time data collected by the user terminal, and obtain the receiving time point of the user information to be identified. After receiving the user information set to be identified sent by the terminal, the server can directly parse to obtain the data collected by the terminal, and can record the receiving time point of the user information to be identified for subsequent identification steps.

[0058] Step S205, perform intention recognition for the target intention according to the user service data, user terminal data, user behavior time data and receiving time point, and obtain the recognition result value of the target intention. In the embodiment of the present disclosure, the target intention can be the card maintenance intention.

[0059] Figure 3 A flowchart of obtaining a recognition result value in a method for intent recognition according to an embodiment of the present disclosure is shown. Figure 3 As shown, in some embodiments Figure 2 Step S205 in the embodiment may further include the following steps S301 to S309:

[0060] Step S301, identify the service characteristics of the target intention according to the user service data to obtain a service identification result value. In this embodiment, the identification can be performed from the service characteristic dimension.

[0061] In some embodiments, user service data includes at least one of the following: user card number registration place, minimum consumption service type, package service type, monthly service type, average income value per user, calling duration, called duration, consumption value in the current month of network access, consumption value in the next month of network access, and user belonging channel type; step S301 may include: obtaining a card maintenance identification model, inputting user service data into the card maintenance identification model; obtaining the card maintenance intention probability output by the card maintenance identification model; using the card maintenance intention probability as the service identification result value; wherein the card maintenance identification model is trained based on the historical service data of card maintenance users and the historical service data of non-card maintenance users.

[0062] The above-mentioned user service data can characterize the service usage status or situation of the phone card user, and each information in the user service data can be used to determine whether the user is using the phone card normally; in some practical applications, the user service data of a large number of users can be obtained as training samples, and the neural network model can be trained according to the training samples, and then the trained neural network model can be used for service feature recognition.

[0063] Step S303: Perform terminal environment recognition for the target intention according to the user terminal data to obtain a terminal recognition result value. In this embodiment, the recognition can be performed from the terminal environment dimension.

[0064] In some embodiments, step S303 may include: determining the number of user terminal data that meet the card maintenance environment conditions; determining the terminal identification result value based on the number; wherein the card maintenance environment conditions include: no international mobile equipment identity code on the terminal, no local number on the terminal, no user identification card number on the terminal, no brand type information on the terminal, no CPU information on the terminal, no Mac address information on the terminal, no battery level change information on the terminal, no baseband information on the terminal, the control group information cannot be read on the terminal, the process group information cannot be read on the terminal, there is no wlan driver information on the terminal, the wlan driver information on the terminal is abnormal, the number of sensors on the terminal is less than the sensor number threshold, and the number of pre-installed applications on the terminal is less than the application number threshold.

[0065] The above user terminal data can be used to characterize the terminal environment. When a normal user uses a phone card, their terminal usually does not meet the above card maintenance environment conditions. For example, the terminal used by a normal user usually has an International Mobile Equipment Identity code and has power changes. Therefore, the terminal identification result value can be determined by judging whether the user terminal data meets the card maintenance environment conditions and the number of card maintenance environment conditions met if they are met.

[0066] For example, the terminal recognition result value can be set to 0 and 1, where 0 can indicate that there is no intention to maintain the card, and 1 can indicate that there is an intention to maintain the card; it can be set that if the number of card maintenance environment conditions is 0, the corresponding terminal recognition result value is 0, and if the number of card maintenance environment conditions is greater than or equal to 1, the corresponding terminal recognition result value is 1. For another example, the terminal recognition result value can be set to 0, 0.2, 0.4, 0.6, 0.8 and 1, and it can be set that if the number of card maintenance environment conditions is 0, the corresponding terminal recognition result value is 0, if the number of card maintenance environment conditions is greater than or equal to 1 and less than or equal to 2, the corresponding terminal recognition result value is 0.2, if the number of card maintenance environment conditions is greater than or equal to 3 and less than or equal to 4, the corresponding terminal recognition result value is 0.4, and so on.

[0067] Step S305, performing data authenticity identification based on the user behavior time data and the receiving time point to obtain a data identification result value.

[0068] In some embodiments, for the aforementioned collection instruction, an offset time period, a hash algorithm, and a number of hash calculations may also be configured in the user behavior time acquisition strategy of the collection instruction, so that the terminal can obtain user behavior time data according to the user behavior time acquisition strategy configured with the offset time period, the hash algorithm, and the number of hash calculations. The user behavior time data may include: an offset time period, a behavior occurrence point at which a target behavior determined based on the offset time period occurs, and a calculation time period spent on iteratively calculating the specified data using the hash algorithm according to the number of hash calculations.

[0069] The user behavior can be determined as follows: first determine the first moment that triggers the collection trigger condition, then use the moment determined according to the offset time period before the first moment as the second moment, and then determine the most recent user behavior before the second moment as the target behavior. In addition, the target behavior may include at least one of the behavior of clicking the mouse, the behavior of operating the keyboard, the gesture behavior, and the gyroscope behavior; the specified data may include: at least one of the user service data, the user terminal data, and the behavior log data of the target behavior.

[0070] For example, if the first moment can be 8:00, and the offset time period is set to two hours, it can be calculated that the moment two hours before 8:00 is 6:00, then the second moment can be determined to be 6:00, and then a query is performed in the terminal local storage space to find the target behavior that was last recorded before 6:00, and then the occurrence time of the target behavior is obtained as 5:00, and then the behavior occurrence time point can be determined to be 5:00. The log data of the target behavior can also be used as the specified data, and the hash algorithm is used to iteratively calculate the specified data according to the number of hash calculations, and then the time spent on the iterative calculation is recorded, for example, 2 seconds, and then 2 seconds can be used as the calculation time period. Therefore, "the offset time period is two hours", "the behavior occurrence time point is 5:00", and "the calculation time period is 2 seconds" can be used as user behavior time data.

[0071] In some practical applications, the hash algorithm and the number of hash calculations can be adjusted and set according to actual conditions; the hash algorithm can be regarded as an algorithm that can consume terminal computing power, for example, the hash algorithm can be the SHA256 algorithm, the SHA512 algorithm, etc., and the number of hash calculations can be, for example, 5 times, 1000 times, 5000 times, etc. The calculation time period can also be determined by recording the time of a hash calculation and obtaining the calculation time period according to the product of the time of a hash calculation and the number of hash calculations.

[0072] Figure 4 A schematic diagram showing iterative calculation in a method for intent recognition according to an embodiment of the present disclosure is shown. Figure 4 As shown, a schematic diagram of five hash iteration calculations is shown, including:

[0073] You can first obtain the specified data;

[0074] First calculation: Get the current collection time, process the specified data and the current collection time through the SHA256 hash algorithm, get the hash value Hash1 of the first hash calculation, and record the time spent on this calculation, that is, time slice t;

[0075] Second calculation: Get the current collection time, process the specified data, current collection time and Hash1 through the SHA256 hash algorithm, get the hash value Hash2 of the second hash calculation, and record the time spent on this calculation, that is, time slice t;

[0076] The third calculation: Get the current collection time, process the specified data, the current collection time and Hash2 through the SHA256 hash algorithm, get the hash value Hash3 of the third hash calculation, and record the time spent on this calculation, that is, the time slice t;

[0077] Fourth calculation: Get the current collection time, process the specified data, current collection time and Hash3 through the SHA256 hash algorithm, get the hash value Hash4 of the fourth hash calculation, and record the time spent on this calculation, that is, time slice t;

[0078] Fifth calculation: Get the current collection time, process the specified data, current collection time and Hash4 through the SHA256 hash algorithm, get the hash value Hash5 of the fifth hash calculation, and record the time spent on this calculation, that is, time slice t;

[0079] At this point, the iterative calculation is completed.

[0080] Obviously, by implementing Figure 4 In the method shown, the previous hash calculation result is used as the input parameter of the next SHA-256 algorithm for calculation. Then, the calculation result can only be generated after the behavior occurs and a certain hash calculation is performed. Therefore, this method increases the cost of counterfeiting.

[0081] Since a large number of phone cards are usually used in a card maintenance terminal, when each phone card causes the card maintenance terminal to perform a large number of hash calculations, the computing power of the card maintenance terminal can be consumed. It can be seen that when collecting information of the user information set to be identified, the hash algorithm and the number of hash calculations can be configured in the collection instruction, so that the terminals used by normal users are not affected, while the card maintenance terminals with the intention of maintaining cards continue to consume computing power, thereby achieving the effect of making the card maintenance behavior costly.

[0082] In the embodiment of step S305, the identification can be performed based on whether the acquired data is forged. The steps in this embodiment can be used to determine whether the data used for identification in step S301 and step S303 is forged by the terminal to avoid inaccurate identification results caused by the terminal using forged data.

[0083] In some embodiments, step S305 may include: calculating the predicted behavior occurrence time point according to the receiving time point, the calculation time period, and the offset time period; determining the difference between the predicted behavior occurrence time point and the behavior occurrence time point, and using the difference as the behavior time difference; determining the true probability of the user behavior time data according to the relationship between the behavior time difference and the time difference threshold, as the data authenticity identification result; determining the data identification result value according to the data authenticity identification result. Among them, the time difference threshold can be adjusted and set according to the actual situation, for example, the minimum threshold can be set to 0 and the maximum threshold can be set to 2 hours.

[0084] For example, if it has been determined that the offset time period T1 is two hours, the behavior occurrence time point t1 is 5:00, the calculation time period T2 is 1 minute, and the receiving time point t2 is 8:05, then the behavior prediction occurrence time point t3 can be calculated based on the receiving time point, calculation time period, and offset time period. For example, the formula (t3=t2-T2-T1) can be used to calculate the behavior prediction occurrence time point t3 as 6:04; then determine the difference between the behavior prediction occurrence time point and the behavior occurrence time point, that is, the value of (t3-t1), and obtain the behavior time difference of 1 hour and 4 minutes, and then obtain the time difference threshold, and determine the true probability of the user behavior time data based on the relationship between the behavior time difference and the time difference threshold; specifically, a probability mapping table corresponding to the time difference threshold can be obtained, and the probability mapping table can include the corresponding relationship between the behavior time difference and the true probability, so that the true probability corresponding to the behavior time difference can be found.

[0085] Step S307, respectively obtain the weight values ​​corresponding to the service identification value, the terminal identification value and the data identification value. The weight values ​​corresponding to the service identification value, the terminal identification value and the data identification value can be preset, and the weight values ​​can be adjusted according to actual conditions.

[0086] It should be noted that the present disclosure does not limit the specific implementation order of steps S301 to 307, and steps S301 to 307 only need to be executed before step S309; ​​specifically, the steps in steps S301 to 307 can be executed simultaneously or in order, and can be adjusted according to actual conditions.

[0087] Step S309, determining the identification result value according to the service identification value, the terminal identification value, the data identification value and the corresponding weight values.

[0088] Figure 5 A schematic diagram of determining a recognition result value in a method for intent recognition according to an embodiment of the present disclosure is shown. Figure 5 As shown, the process of steps S301 to S309 may be:

[0089] First, the corresponding quantized value (i.e., business identification value) P1 is obtained through card maintenance feature identification Y1 (i.e., business feature identification), the corresponding quantized value (i.e., terminal identification value) P2 is obtained through simulator Y2 (i.e., terminal environment identification), and the corresponding quantized value (data identification value) P3 is obtained through anti-counterfeiting identification Y3 (i.e., data authenticity identification), and then the weights W1, W2, and W3 corresponding to each quantized value are obtained respectively, and then weighted calculation (P1*W1+P2*W2+P3*W3) is performed to obtain the identification result value. The higher the identification result value, the greater the card maintenance intention represented by the user information set to be identified.

[0090] Step S207, determining the degree to which the to-be-identified user information set represents the target intention according to the recognition result value.

[0091] A pre-set level mapping table can be obtained, and according to the correspondence between the recognition result value and the card maintenance intention level in the level mapping table, it can be determined whether the user information set to be identified represents the target intention and the degree to which it represents the target intention.

[0092] Figure 6 A flowchart of a method for intent recognition according to an embodiment of the present disclosure is shown. Figure 6 As shown, including:

[0093] Step (1), the user terminal triggers the collection of user behavior data and terminal environment data; records the time of each behavior point and each environment parameter; and calculates the time offset of each behavior data.

[0094] Among them, user behavior may include any one of the following behaviors, such as clicking the mouse, operating the keyboard, gesture behavior, and gyroscope behavior, and any combination thereof; environmental data refers to the software and hardware information of the user terminal, which may include IMEI number (International Mobile Equipment Identity), local phone number, SIM card (subscriber identification card number) serial number, mobile phone brand, manufacturer, model, CPU information, Mac address, battery information, baseband information, number of sensors, number of pre-installed apps by users, etc.

[0095] Step (2): After that, the user terminal performs n (5, for example) SHA-256 hash calculations, and each SHA-256 hash calculation takes a time slice t.

[0096] The results obtained in step (3), step (1) and step (2) form a result set and are sent to the server.

[0097] Step (4), the server receives the result set and performs: 1. Card maintenance feature recognition: using clustering big data modeling; 2. Simulator recognition: using environmental parameter rules to make judgments; 3. Anti-counterfeiting recognition: calculating the difference between the user behavior calculation time and the user behavior collection time, and whether it exceeds a preset time difference threshold.

[0098] Among them, for the identification of card maintenance features (i.e. business feature identification), big data modeling can be used to perform data mining based on the features; specifically, a big data clustering algorithm can be used (based on features such as location, minimum consumption, package type, whether there is a monthly rent, monthly ARPU value, calling duration, consumption in the month of network access, consumption in the next month of network access, developed channels, activated IMEI, etc.) to determine whether to maintain the card.

[0099] For simulator identification (i.e., terminal environment identification), whether it is a simulator can be determined based on the collected user terminal software and hardware parameters; specifically, if one of the following conditions is met, it is determined that the user behavior is running on a simulator: (1) Obtain the mobile phone IMEI number, obtain the local phone number, and obtain the SIM card serial number, if it is empty; (2) Read the mobile phone brand information, including brand, manufacturer, and model, if it is empty; (3) Check the CPU information, if it is empty; (4) Check the Mac address, if it is empty; (5) Check whether the battery information has changed, if it has not changed; (6) Check whether the baseband information is empty, if it is empty; (7) Determine whether the control group information can be read, if not; (8) Determine whether the process group information can be read, if not; (9) Determine whether the wlan driver is unknown or abnormal, if yes; (10) Determine whether the number of sensors is 1, if yes; (12) Determine whether the user's pre-installed apps are less than 5, if yes.

[0100] For anti-counterfeiting identification (i.e. data authenticity identification), it can prevent users from forging action data such as mouse, keyboard, gesture and gyroscope, prevent forging device data such as IP, UA, device fingerprint, prevent machines from posting, brushing orders, posting advertisements, and prevent machines from killing instantly; specifically, a behavior prediction occurrence time point can be calculated by obtaining the receiving time point, calculation time period, and offset time period, and then the behavior prediction occurrence time point is compared with the behavior occurrence time point. On the one hand, the behavior prediction occurrence time point should be later than the behavior occurrence time point, and on the other hand, the time period that the behavior occurrence time point is earlier than the behavior prediction occurrence time point cannot exceed the preset time difference threshold, so that the identification result can be determined according to the comparison result.

[0101] Step (5), according to the weighted scoring calculation model, a score (i.e., recognition result value) is obtained.

[0102] Among them, the scoring formula can be (P1*W1+P2*W2+P3*W3), and in this step Figure 5 The same contents in the embodiment shown can be found in Figure 5 The textual description of the illustrated embodiments will not be repeated in this disclosure.

[0103] It should be noted that the above figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0104] Figure 7 FIG. 7 is a block diagram of a device 700 for intent recognition in a fifth embodiment of the present disclosure; Figure 7 As shown, including:

[0105] The receiving module 701 is used to receive a user information set to be identified sent by a user terminal;

[0106] The parsing module 702 is used to parse the user information set to be identified, obtain the user service data, user terminal data and user behavior time data collected by the user terminal, and obtain the receiving time point of the user information to be identified;

[0107] The identification module 703 is used to perform intent identification for the target intent according to the user service data, the user terminal data, the user behavior time data and the receiving time point, and obtain the identification result value of the target intent;

[0108] The determination module 704 is used to determine the degree to which the to-be-identified user information set represents the target intention according to the recognition result value.

[0109] Figure 7 For other contents of the embodiment, reference can be made to the other embodiments mentioned above.

[0110] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as a system, method or program product. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as a "circuit", "module" or "system".

[0111] Figure 8 The block diagram of a computer device for intent recognition in an embodiment of the present disclosure is shown. It should be noted that the electronic device shown in the figure is only an example and should not bring any limitation to the function and scope of use of the embodiment of the present invention.

[0112] Refer to the following Figure 8 An electronic device 800 according to this embodiment of the present invention will be described. Figure 8 The electronic device 800 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0113] like Figure 8 As shown, the electronic device 800 is in the form of a general computing device. The components of the electronic device 800 may include but are not limited to: at least one processing unit 810, at least one storage unit 820, and a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810).

[0114] The storage unit stores program codes, which can be executed by the processing unit 810, so that the processing unit 810 performs the steps according to various exemplary embodiments of the present invention described in the above “Exemplary Method” section of this specification. For example, the processing unit 810 can perform the following steps: Figure 2 The method shown in .

[0115] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 8201 and / or a cache memory unit 8202 , and may further include a read-only memory unit (ROM) 8203 .

[0116] The storage unit 820 may also include a program / utility 8204 having a set (at least one) of program modules 8205, such program modules 8205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0117] Bus 830 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0118] The electronic device 800 may also communicate with one or more external devices 900 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 850. Furthermore, the electronic device 800 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 860. As shown, the network adapter 860 communicates with other modules of the electronic device 800 via a bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0119] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present specification.

[0120] The program product for implementing the above method according to the embodiment of the present invention may adopt a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto, and in this document, a readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus, or device.

[0121] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0122] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0123] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0124] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0125] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0126] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0127] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0128] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A method for intent recognition, characterized in that: include: Receiving a user information set to be identified sent by a user terminal; Parsing the user information set to be identified, obtaining user service data, user terminal data and user behavior time data collected by the user terminal, and obtaining a receiving time point of receiving the user information to be identified; Performing intent recognition for a target intent according to the user service data, the user terminal data, the user behavior time data and the receiving time point, and obtaining a recognition result value for the target intent; Determining, according to the recognition result value, the degree to which the to-be-recognized user information set represents the target intention; The target intention is the card maintenance intention; before receiving the user information set to be identified sent by the user terminal, the method further includes: configuring a collection instruction, wherein the collection instruction configures a user behavior time acquisition strategy; The user behavior time acquisition strategy is configured with an offset time period, a hash algorithm, and a number of hash calculations; and, The user behavior time data includes: the offset time period, the behavior occurrence time point of the target behavior determined based on the offset time period, and the calculation time period spent on iteratively calculating the specified data according to the number of hash calculations using the hash algorithm; wherein the hash algorithm is used as an algorithm for consuming terminal computing power to consume the computing power of the user terminal; Among them, the user behavior is determined in the following manner: determine the first moment that triggers the collection trigger condition, take the moment determined according to the offset time period before the first moment as the second moment, and determine the most recent user behavior before the second moment as the target behavior; the target behavior includes at least one of the behavior of clicking the mouse, the behavior of operating the keyboard, the gesture behavior and the gyroscope behavior; the specified data includes: at least one of the user business data, the user terminal data and the behavior log data of the target behavior.

2. The method according to claim 1, characterized in that The step of performing intent recognition for a target intent according to the user service data, the user terminal data, the user behavior time data and the receiving time point to obtain a recognition result value for the target intent includes: Performing business feature identification for the target intention according to the user business data to obtain a business identification result value; Performing terminal environment recognition for the target intention according to the user terminal data to obtain a terminal recognition result value; Perform data authenticity identification based on the user behavior time data and the receiving time point to obtain a data identification result value; Respectively obtaining weight values ​​corresponding to the service identification value, the terminal identification value, and the data identification value; The identification result value is determined according to the service identification value, the terminal identification value, the data identification value and corresponding weight values.

3. The method according to claim 2, characterized in that The user service data includes at least one of the following: user card number registration place, minimum consumption service type, package service type, monthly rental service type, average income per user, calling duration, called duration, consumption value in the current month of network access, consumption value in the next month of network access, and user belonging channel type; The step of identifying the service characteristics of the target intention according to the user service data and obtaining the service identification result value comprises: Obtain a card maintenance identification model, and input the user service data into the card maintenance identification model; Obtaining the card maintenance intention probability output by the card maintenance recognition model; The card maintenance intention probability is used as the business identification result value; wherein the card maintenance identification model is obtained by training based on the historical business data of card maintenance users and the historical business data of non-card maintenance users.

4. The method according to claim 2, characterized in that: The step of performing terminal environment recognition for the target intention according to the user terminal data to obtain a terminal recognition result value includes: Determine the number of user terminal data that meets the card maintenance environment conditions; Determine the terminal identification result value according to the number; Among them, the card maintenance environment conditions include: there is no international mobile equipment identity code on the terminal, no local number on the terminal, no user identification card number on the terminal, no brand type information on the terminal, no CPU information on the terminal, no Mac address information on the terminal, no battery level change information on the terminal, no baseband information on the terminal, the control group information cannot be read on the terminal, the process group information cannot be read on the terminal, there is no wlan driver information on the terminal, the wlan driver information on the terminal is abnormal, the number of sensors on the terminal is less than the sensor number threshold, and the number of pre-installed applications on the terminal is less than the application number threshold.

5. The method according to claim 2, characterized in that: The acquisition instruction configures an acquisition trigger condition; the method further includes: The collection instruction is sent to the user terminal, so that when the collection trigger condition is triggered, the user terminal collects the user service data, the user terminal data and the user behavior time data according to the collection instruction, thereby generating the user information set to be identified.

6. The method according to claim 5, characterized in that The collection instruction also configures the target service data type to be collected and the target terminal data type to be collected; and, The user service data is collected and obtained by the user terminal according to the target service data type; the user terminal data is collected and obtained by the user terminal according to the target terminal data type; the user behavior time data is obtained by the user terminal according to the user behavior time acquisition strategy.

7. The method according to claim 1, characterized in that The step of performing data authenticity identification according to the user behavior time data and the receiving time point to obtain a data identification result value includes: The behavior prediction occurrence time point is calculated according to the receiving time point, the calculation time period, and the offset time period; Determine the difference between the predicted behavior occurrence time point and the behavior occurrence time point, and use the difference as the behavior time difference; According to the relationship between the behavior time difference and the time difference threshold, determining the true probability of the user behavior time data as a data authenticity identification result; The data identification result value is determined according to the data authenticity identification result.

8. A device for intention recognition, characterized in that: include: A receiving module, used for receiving a user information set to be identified sent by a user terminal; A parsing module, used to parse the user information set to be identified, obtain user service data, user terminal data and user behavior time data collected by the user terminal, and obtain a receiving time point of the user information to be identified; An identification module, used to perform intent identification for a target intent according to the user service data, the user terminal data, the user behavior time data and the receiving time point, and obtain an identification result value for the target intent; A determination module, used to determine the degree to which the to-be-identified user information set represents the target intention according to the recognition result value; Wherein, the target intention is the card maintenance intention; before receiving the user information set to be identified sent by the user terminal, the device is also used to: configure a collection instruction, wherein the collection instruction configures a user behavior time acquisition strategy; The user behavior time acquisition strategy is configured with an offset time period, a hash algorithm, and a number of hash calculations; and, The user behavior time data includes: the offset time period, the behavior occurrence time point of the target behavior determined based on the offset time period, and the calculation time period spent on iteratively calculating the specified data according to the number of hash calculations using the hash algorithm; wherein the hash algorithm is used as an algorithm for consuming terminal computing power to consume the computing power of the user terminal; Among them, the user behavior is determined in the following manner: determine the first moment that triggers the collection trigger condition, take the moment determined according to the offset time period before the first moment as the second moment, and determine the most recent user behavior before the second moment as the target behavior; the target behavior includes at least one of the behavior of clicking the mouse, the behavior of operating the keyboard, the gesture behavior and the gyroscope behavior; the specified data includes: at least one of the user business data, the user terminal data and the behavior log data of the target behavior.

9. A computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the method for intent recognition according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the method for intent recognition as described in any one of claims 1 to 7.

Images