CN114330263A

CN114330263A - Message identification method, device, equipment and storage medium

Info

Publication number: CN114330263A
Application number: CN202011047114.3A
Authority: CN
Inventors: 陈志强
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-09-29
Filing date: 2020-09-29
Publication date: 2022-04-12

Abstract

The application provides a message identification method, a message identification device, message identification equipment and a message identification storage medium, and belongs to the technical field of the Internet. The method comprises the following steps: for a message generated based on a template, a fingerprint sample of such template, i.e. a plurality of reference character strings capable of representing edited text content in the message, may be obtained to match with corresponding character strings in the message to determine whether the message is of the target type. By adopting the matching between the character strings representing the edited text content, the calculation amount of data processing in the message identification process is reduced, a large amount of model calculation is not needed, the maintenance cost of the fingerprint sample is low, in addition, the method makes full use of the characteristic that the message generated based on the message template has a similar or same grammatical structure, the matching accuracy is high, and the accuracy and the efficiency of the message identification are improved.

Description

Message identification method, device, equipment and storage medium

Technical Field

The present application relates to the field of internet technologies, and in particular, to a method, an apparatus, a device, and a storage medium for message identification.

Background

With the development of computer technology and the diversification of terminal functions, the social application provides a social platform which is convenient for communication among users, and the purpose of communication can be achieved through message interaction on the social platform. For example, the social platform provides a message interaction mode based on a template, and a user can call a fixed template provided by the social platform, and edit the fixed template based on the typesetting, the information type and the like specified by the fixed template so as to generate a message to be pushed. Some malicious users may use such messages to perform malicious activities, such as advertising, fraud, etc., which may cause unnecessary annoyances to the user and may also pose a threat to the user's property security.

In the related art, in order to avoid harassment or threat to property security to a user, a server of a social platform may identify message content to be delivered, and the specific identification method may include: and directly inputting a large number of acquired messages into a text recognition model as texts, and recognizing the texts by using the recognition model so as to determine whether the messages are malicious messages.

In the above process, the order of magnitude of the message to be recognized is large, which results in a large amount of computation in the recognition process of the server, and a large amount of data processing resources are consumed, and since the complete text of the message is recognized, many redundant information, such as format information, is included therein, which results in low recognition accuracy and efficiency.

Disclosure of Invention

The embodiment of the application provides a message identification method, a message identification device, message identification equipment and a message identification storage medium, and the accuracy and the efficiency of message identification are improved. The technical scheme is as follows:

in one aspect, a message identification method is provided, and the method includes:

acquiring a first message, wherein the first message is generated based on a template;

determining a fingerprint sample, wherein the fingerprint sample comprises a plurality of reference character strings obtained based on a sample message of a target type, and the plurality of reference character strings are used for representing edited text content of the sample message;

acquiring a plurality of character strings included in the first message, wherein the character strings are used for representing the edited text content of the first message;

and matching the plurality of reference character strings with the plurality of character strings to obtain a matching result, and determining that the first message belongs to the target type in response to the matching result meeting a target condition.

In another aspect, an apparatus for identifying a message is provided, the apparatus including:

the message acquisition module is used for acquiring a first message, and the first message is generated based on a template;

a sample determination module for determining a fingerprint sample, wherein the fingerprint sample comprises a plurality of reference character strings obtained based on a sample message of a target type, and the plurality of reference character strings are used for representing edited text content of the sample message;

a character string obtaining module, configured to obtain a plurality of character strings included in the first message, where the character strings are used to represent edited text content of the first message;

the matching module is used for matching the plurality of reference character strings with the plurality of character strings to obtain a matching result;

and the type determining module is used for determining that the first message belongs to the target type in response to the matching result meeting the target condition.

In an optional implementation manner, the matching result meeting the target condition means:

the number of the successfully matched reference character strings is greater than or equal to the target number;

the ratio of the number of successfully matched reference character strings to the number of the plurality of reference character strings is greater than or equal to a target ratio.

In an optional implementation, the apparatus further includes:

a comparison module for periodically comparing a first set of fingerprints comprising fingerprint samples of the target type with a second set of fingerprints comprising fingerprint samples of a non-target type;

a deleting module, configured to delete a coincident fingerprint sample from the first fingerprint set if the coincident fingerprint sample exists in the first fingerprint set and the second fingerprint set.

In an optional implementation, the apparatus further includes:

the sample message acquisition module is used for acquiring sample messages of the template, wherein the sample messages belong to the target type;

a deleting module, configured to delete a meaningless character in the sample message, and fill a target symbol in a deleted position to obtain a plurality of first sample character strings, where every two first sample character strings are separated by the target symbol;

the deleting module is further configured to delete a target sample character string in the plurality of first sample character strings, where the target sample character string is a character string between two target symbols and the number of words between the two target symbols does not satisfy a word number threshold, so as to obtain a plurality of second sample character strings;

a reference character string obtaining module, configured to obtain the plurality of reference character strings from the plurality of second sample character strings.

In another aspect, a computer device is provided, which includes a processor and a memory, where the memory is used to store at least one program code, and the at least one program code is loaded and executed by the processor to implement the operations performed in the message identification method in the embodiments of the present application.

In another aspect, a computer-readable storage medium is provided, in which at least one program code is stored, and the at least one program code is loaded and executed by the processor to implement the operations performed in the message identification method in the embodiments of the present application.

In another aspect, a computer program product is provided that includes computer program code stored in a computer readable storage medium. The processor of the computer device reads the computer program code from the computer-readable storage medium, and the processor executes the computer program code, causing the computer device to perform the message identification method provided in the various alternative implementations described above.

The technical scheme provided by the embodiment of the application has the following beneficial effects:

in the embodiment of the present application, a message identification method is provided, where for a message generated based on a template, a fingerprint sample of the template, that is, a plurality of reference character strings capable of representing edited text content in the message, may be obtained to match with corresponding character strings in the message, so as to determine whether the message belongs to a target type. By adopting the matching between the character strings representing the edited text content, the calculation amount of data processing in the message identification process is reduced, a large amount of model calculation is not needed, the maintenance cost of the fingerprint sample is low, in addition, the method makes full use of the characteristic that the message generated based on the message template has a similar or same grammatical structure, the matching accuracy is high, and the accuracy and the efficiency of the message identification are improved.

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 message identification method according to an embodiment of the present application;

fig. 2 is a flowchart of a message identification method according to an embodiment of the present application;

FIG. 3 is a flow chart of another message identification method provided in accordance with an embodiment of the present application;

FIG. 4 is a flowchart of a fingerprint sample acquisition method provided in an embodiment of the present application;

fig. 5 is a block diagram of a message identification apparatus according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of a server provided 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.

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

Next, an implementation environment of the message identification method provided in the embodiment of the present application is described, and fig. 1 is a schematic diagram of an implementation environment of the message identification method provided in the embodiment of the present application. The implementation environment includes: a terminal 101 and a server 102.

The terminal 101 and the server 102 can be directly or indirectly connected through wired or wireless communication, and the application is not limited herein. Optionally, the terminal 101 is a smart phone, a tablet computer, a notebook computer, a desktop computer, or the like, but is not limited thereto. The terminal 101 can be installed and run with an application. Optionally, the application is a social application that can provide social related functionality such as session functionality. Illustratively, the terminal 101 is a terminal used by a user, and a user account of the user is registered in an application running in the terminal 101.

The server 102 is a server of a social application platform, which refers to a network architecture that connects people to each other through social relationships and/or common interests (or common interests), and users can perform daily communications and process daily transactions through clients provided by the social application platform. Each user may have a network identity that is recognized for other users on the social application platform.

On the social application platform, different users can establish a social relationship in a mutually confirmed manner, for example, mutually friend or mutually pay attention to each other. When two users establish a social relationship, they become social network contacts of each other. A group of users can form a social group by forming a social relationship with each other in a self-selection mode. Each member in the group is a social network contact of all other members in the group.

A user or organization may establish a public social network identity on a social application platform and allow the public (e.g., any user on the social application platform) to communicate with the public social network identity on the social application platform, which may be based on a one-way confirmation without mutual confirmation between users. For example, a user may choose to subscribe to a public social network identification (e.g., "follow" the public social network identity) message or publish information to become a social network contact of the public social network identification through a one-way confirmation such as subscription. The owner of the public social network identity may also take other users who subscribe to their messages or published information as their social network contacts. The social application platform provides a message template for the public social network identification, so that the public social network identification generates messages based on the message template, thereby performing batch pushing of the messages, and the user identification having a social relationship with the public social network identification can receive the messages pushed in batch.

Each user and each public social network identifier on the social application platform are provided with a social network contact list so that the users or the public social network identifiers in the social network contact list can communicate with the users or the public social network identifiers in the social network contact list in the forms of instant messaging messages and the like. For example, users within a social group may communicate with each other through an interface provided by the social application platform, and users may also communicate with each other through an interface provided by the social application platform.

The server 102 may be an independent physical server, 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, cloud storage, a network service, cloud communication, a middleware service, a domain name service, a security service, a CDN, and a big data and artificial intelligence platform.

Optionally, in the message identification process, the server 102 undertakes primary computation, and the terminal 101 undertakes secondary computation; or, the server 102 undertakes the secondary computing work, and the terminal 101 undertakes the primary computing work; alternatively, the server 102 or the terminal 101 can be respectively capable of separately assuming the calculation work.

Optionally, the terminal 101 generally refers to one of a plurality of terminals, and this embodiment is only illustrated by the terminal 101. Those skilled in the art will appreciate that the number of terminals 101 can be greater. For example, the number of the terminals 101 is dozens or hundreds, or more, and the implementation environment of the message identification method also includes other terminals. The number of terminals and the type of the device are not limited in the embodiments of the present application.

Optionally, the wireless network or wired network described above uses standard communication techniques and/or protocols. The Network is typically the Internet, but can be any Network including, but not limited to, a Local Area Network (LAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a mobile, wireline or wireless Network, a private Network, or any combination of virtual private networks. In some embodiments, data exchanged over a network is represented using techniques and/or formats including Hypertext Mark-up Language (HTML), Extensible Markup Language (XML), and the like. All or some of the links can also be encrypted using conventional encryption techniques such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec). In other embodiments, custom and/or dedicated data communication techniques can also be used in place of or in addition to the data communication techniques described above.

In an embodiment of the present invention, a message identification method is provided, in which a message is generated based on a message template. The format content in the message at least accounts for half of the message content, and except the format content, the content which can be customized by a message sender is very limited, so that the customized partial content is only required to be checked, the semantic analysis and word segmentation statistics of the complete message are not required, and the workload is greatly reduced. Moreover, in order to adapt to a multi-account, multi-template, wide-broadcast network and a working mode with fast content change, a sender of some malicious messages inevitably needs to reduce personalized content to enhance reusability of the content, and meanwhile, in order to meet the requirement of rapid change, a new message template cannot be created every time of calling, so that the message generated based on the same template usually has highly consistent language style and grammar structure except format content. The embodiment of the application makes full use of the above features, processes the first message to be issued to obtain a limited character string, and identifies the message type of the first message to be issued by matching with the fingerprint sample, and for a more specific description, refer to the following embodiments.

Fig. 2 is a flowchart of a message identification method according to an embodiment of the present application, and as shown in fig. 2, the message identification method is described in the embodiment of the present application by taking an application to a server as an example. The message identification method comprises the following steps:

201. a first message is obtained, and the first message is generated based on a template.

In the embodiment of the present application, the first message is a message to be published, or a published message stored in a user message database, and for different identification requirements, the message may be from different sources, for example, if the message to be published needs to be identified, the first message is extracted from a message publishing request to perform subsequent steps, and then relevant publishing steps are correspondingly performed, where the message publishing request is used to instruct a server to publish the message; if the published message needs to be identified, the first message is extracted from the user message database for storing the published message to perform subsequent steps, and then the relevant processing steps for the published message are correspondingly performed, which is not limited in the embodiment of the present application.

The template is a message template predefined by the server, that is, a template used for sending a message. In some embodiments, the first message is provided by a public social network identifier on the social application platform, and the server extracts the first message from a message publishing request by receiving the message publishing request of the public social network identifier, or extracts the first message from a published message of the public social network identifier in the user message database.

202. A fingerprint sample is determined, the fingerprint sample including a plurality of reference strings derived based on a sample message of a target type, the plurality of reference strings being used to represent edited text content of the sample message.

In the embodiment of the application, the fingerprint sample is acquired from a sample message belonging to a target type, the sample message is a fingerprint original text of the fingerprint sample, and the target type is generated based on a message template, wherein the target type refers to that the type of the fingerprint sample is an abnormal type. The fingerprint sample comprises a plurality of reference character strings, the reference character strings are used for representing edited text content of the sample message, the edited text content is self-defined content except template format content in a message template, and the edited text content is obtained by editing operations such as input and the like of a user.

203. A plurality of character strings included in the first message are acquired, and the character strings are used for representing edited text content of the first message.

In the embodiment of the application, the server processes the first message to obtain a plurality of character strings only used for representing the edited text content.

204. And matching the plurality of reference character strings with the plurality of character strings to obtain a matching result, and determining that the first message belongs to the target type in response to the matching result meeting the target condition.

In the embodiment of the present application, the matching result refers to the number of successful matching of the plurality of reference character strings in the fingerprint sample. The target condition is used to indicate that the sample message to which the first message corresponds to the fingerprint sample is a similar message.

Fig. 3 is a flowchart of a message identification method according to an embodiment of the present application, and as shown in fig. 3, an interaction between a terminal and a server is taken as an example in the embodiment of the present application for description. The message identification method comprises the following steps:

301. the terminal responds to the editing operation of the user based on the template and generates a first message.

In the embodiment of the application, the social application client of the terminal can provide a message editing function based on the template, and a user selects the template and performs editing operations such as text input based on the template through a message editing interface of the social application client, so that a message to be pushed is generated.

In an alternative implementation manner, the social application client of the terminal provides at least one message template, for example, the message template is a "credit card swiping notification" template, a "commodity purchase success reminding" template, and the like, wherein specific styles of the "credit card swiping notification" template are "[ respected user ] … … [ your account ] … … [ your consumption amount ] … …", "respected user", "your account", and "your consumption amount" are fixed contents preset in the message template, and the rest are contents customizable by the user.

In an optional implementation manner, a social application client of the terminal displays a message editing interface, where the message editing interface includes at least one message template, obtains a message template that needs to be edited and is selected by a user in the message editing interface, displays the template editing interface of the selected message template, and further obtains characters input by the user in the displayed template editing interface to generate a first message. For example, the message template selected by the user is a "credit card swiping notification" template, the social application client of the terminal displays a message editing interface based on the template, obtains text input of the user on the message editing interface, and generates "[ honored user ] if you want wealth to add value, [ your account ] if you want to be overnight riot, [ your consumption amount ] you need to pay attention to the public number of the XXX teacher".

302. And the terminal responds to the message issuing instruction and sends a message issuing request to the server, wherein the message issuing request carries the first message and the template identifier of the template.

In the embodiment of the present application, the message issuing instruction refers to an instruction for a user to issue a message, and the template identifier refers to a unique identifier of each message template. In some embodiments, a user may trigger a corresponding message publishing instruction on a social application client of a terminal by clicking a message publishing button or inputting a voice, and the terminal sends a message publishing request to a server in response to the message publishing instruction, where the message publishing request is used to instruct the server to publish a message.

303. After receiving the message publishing request, the server determines a fingerprint sample corresponding to the template according to the template identifier of the template, wherein the fingerprint sample comprises a plurality of reference character strings, and the plurality of reference character strings are used for representing the edited text content of the sample message.

In the embodiment of the application, the fingerprint sample comprises a plurality of reference character strings, the reference character strings are used for representing edited text content of the sample message, the edited text content refers to self-defined content except for template format content in the message template, and is obtained by editing operations such as input and the like of a user. For example, there are 3 reference character strings in one fingerprint sample for indicating edit text contents of "wealth added value", "overnight wealth", and "public number of XXX-focused teacher", respectively.

In an alternative implementation, the server is associated with a first fingerprint set, where the first fingerprint set includes fingerprint samples of a target type, and the target type refers to the type of the fingerprint sample being an abnormal type. In some embodiments, the first fingerprint set may be a black fingerprint set, and the content property of the sample message corresponding to the fingerprint sample in the first fingerprint set points to abnormal content. After receiving the message publishing request, the server takes the template identification of the template as an index, and queries the fingerprint sample corresponding to the template in the first fingerprint set.

In an alternative implementation manner, the server is further associated with a second fingerprint set on the basis of the first fingerprint set, and the second fingerprint set comprises fingerprint samples of a non-target type, wherein the non-target type means that the type of the fingerprint sample is a normal type. In some embodiments, the second set of fingerprints may be a white set of fingerprints, and the content property of the sample message corresponding to the fingerprint sample in the second set of fingerprints points to normal content. And, this second fingerprint set can be as the supplement of first fingerprint set, avoid discerning normal content as unusual content.

In an alternative implementation, the implementation manner of supplementing the second set of fingerprints is: the server periodically compares the first fingerprint set with the second fingerprint set, determines the coincident fingerprints existing in the two sets through comparison, and then deletes the coincident fingerprints from the first fingerprint set. By periodically comparing the two fingerprint sets, the accuracy of the fingerprint sample can be ensured, and the condition that the normal content is identified as the abnormal content subsequently, the error processing and the like can not be caused.

For example, in a first fingerprint sample set, that is, a black fingerprint set, fingerprint samples whose content is "continuously maintained and maintained" are included, and when the server compares the two fingerprint sets, the server detects a second fingerprint sample set, that is, a white fingerprint set, and also includes fingerprint samples "continuously maintained and maintained", and at this time, the server deletes the fingerprint sample from the black fingerprint set and retains the fingerprint sample in the white fingerprint set. That is, the server needs to ensure that there are no fingerprint samples in the white fingerprint set in the black fingerprint set, so that when a template message issued by a government agency or other good-intentioned user carries some fingerprint samples in the black fingerprint set, the server does not confirm the information as abnormal content because there are fingerprint samples in the white fingerprint set in the information.

It should be noted that, the foregoing steps 302 to 303 are an implementation manner of determining a fingerprint sample corresponding to a template according to a template identifier of the template provided in an embodiment of the present application. In another alternative implementation manner, after receiving the message publishing request, the server determines any fingerprint sample, where the fingerprint sample is based on any message template, and then the server continues to perform subsequent steps 304 to 311, which is not limited in this embodiment of the present application.

In addition, it should be noted that, in the embodiment of the present application, the process of acquiring the fingerprint sample will be described in the subsequent steps 401 to 404, and therefore, the description thereof is omitted here.

304. The server deletes meaningless characters in the first message, fills target symbols at the deleted positions to obtain a plurality of first character strings, and separates every two first character strings by using the target symbols.

In the embodiment of the present application, the nonsense character includes a symbol and a character not belonging to the target language, wherein the target language is the language to which the first message belongs. The target symbol is a special symbol that does not exist in the text corresponding to the first message, for example, the target symbol is "@", "#", "&", and the like. The server deletes information without practical significance, such as symbols, in the first message, deletes characters which do not belong to the target language in the first message, and fills the deleted blank with the target symbols, so that the data volume of subsequent processing can be reduced, and advance preparation can be made for splitting the first message.

The following explains a specific implementation manner of this step, taking the target language as the chinese language as an example:

in an alternative implementation manner, the server sequentially identifies each character in the first message in a left-to-right order, and deletes the character when the character is identified to be represented as a symbol, such as a punctuation mark or other common symbols; when the character is recognized to represent non-Chinese content, such as letters, numbers and the like, the character is deleted; after deleting all symbols and non-Chinese content in the first message, the server fills all blank spaces in the first message, which appear due to character deletion, with target symbols.

In an alternative implementation manner, the server sequentially identifies each character in the first message from left to right, deletes the character corresponding to the symbol or the non-chinese content when the character is identified as the symbol or the non-chinese content, and fills a target symbol in a blank space appearing at the character deletion position, that is, fills the target symbol while deleting the character corresponding to the symbol and the non-chinese content. Optionally, the server performs the above-mentioned identification process from right to left, which is not limited in this embodiment.

It should be noted that, in the above, taking the target language of the first message as the chinese language as an example, in another optional implementation manner, when the target language of the first message is the english language, the server deletes the symbols and the non-english content in the first message, and fills the blank positions after deletion with the target symbols to obtain a plurality of first character strings. The embodiment of the present application does not limit the target language of the first message.

The foregoing steps are an implementation manner of deleting the meaningless character in the first message in the embodiment of the present application, and the server may also have other manners of deleting the meaningless character, for example, deleting an exclamation word in the first message, deleting an quantifier in the first message, and the like. This is not particularly limited by the present application.

305. The server deletes a target character string of the plurality of first character strings, wherein the target character string is a character string between two target symbols, and the number of words of the character string does not meet a word number threshold value, so that a plurality of second character strings are obtained, and the plurality of second character strings are used for representing edited text content of the first message.

The above step 305 includes the following steps: the server checks the number of characters of each first character string; if the number of the characters of the first character string is smaller than the threshold value of the number of the characters, the first character string is a target character string, and the target character string is deleted; if the number of the characters of the first character string is larger than or equal to the word number threshold, the first character string is reserved; and using the checked plurality of first character strings as a plurality of second character strings.

The server deletes the target character strings of which the number of characters does not meet the word number threshold value, and keeps the first character strings meeting the word number threshold value to obtain a plurality of second character strings of the first message, so that the text content without practical significance in the message is prevented from being identified, and the subsequent data processing amount is further reduced.

For example, the server may set the word number threshold to be 4 chinese characters in length represented by the number of characters of the first character string, that is, when the server detects that the length of the chinese character represented by the first character string is less than 4 chinese characters, the first character string is deleted; when the server detects that the length of Chinese represented by the first character string is greater than or equal to 4 characters, the first character string is reserved.

It should be noted that, in this embodiment, the server first obtains fingerprint samples according to the foregoing steps 303 to 305 in a sequence from front to back, and then processes the first message to obtain a plurality of first character strings, and then obtains a plurality of second character strings. In another alternative implementation manner, the server may first process the first message according to

steps

304 and 305, and then perform step 303 after obtaining a plurality of second character strings, to obtain a fingerprint sample corresponding to the first message. This is not particularly limited by the embodiments of the present application.

306. And the server matches the plurality of reference character strings in the fingerprint sample of the template with the plurality of second character strings to obtain a matching result.

In the embodiment of the application, a plurality of reference character strings exist in the fingerprint sample corresponding to the template, and a plurality of second character strings are obtained by processing the first message. And the server matches the plurality of reference character strings in the fingerprint sample with the plurality of second character strings, and when two character strings are the same, the two character strings are considered to be successfully matched.

It should be noted that the first message and the fingerprint sample are generated based on the template, and have high similarity with the original fingerprint text of the fingerprint sample of the template in terms of grammatical structure, typesetting characteristics and language style, and the plurality of second character strings obtained by processing based on the first message and the plurality of reference character strings in the fingerprint sample all represent edited text content, and are obtained by editing operations such as user input. For example, the content represented by the first one of the plurality of second strings is "hello", and the content represented by the first reference string of the fingerprint sample is "hello"; the content represented by the second character string of the plurality of second character strings is "fast attention", the content represented by the second reference character string of the fingerprint sample is "attention required", and the like, all represent the custom content of the template, i.e., the edited text content.

In an alternative implementation, the server determines, for each of a plurality of reference character strings of the fingerprint sample, a character string having the same character string number as that of the plurality of reference character strings from a plurality of second character strings according to the character string number of each of the reference character strings, and then matches the reference character string with the determined character string. For example, the server obtains the number of each reference character string in the fingerprint sample, determines that the numbers of the reference character strings are odd numbers, then the server takes the character strings with the odd numbers in the second character strings as the character strings to be matched, then the server performs one-to-one corresponding matching on the character strings to be matched in the second character strings and the reference character strings in the fingerprint sample, and finally obtains the matching result. By the above method for matching the character strings according to the serial numbers of the character strings, the matching accuracy is ensured.

307. And the server determines whether the first message and the sample message corresponding to the fingerprint sample meet the target condition according to the matching result, if so, the steps 308 to 311 are executed, and if not, the processing flow is ended.

In the embodiment of the application, the greater the number of successful matching between the plurality of reference character strings and the plurality of second character strings in the fingerprint sample of the template, the greater the number of text contents which are the same in the text contents represented by the plurality of reference character strings and the plurality of second character strings, and therefore, whether the sample message corresponding to the first message and the fingerprint sample meets the target condition is determined through the number of successful matching between the character strings, and the accuracy is high.

In an optional implementation manner, the server obtains the number of successfully matched reference character strings and the number of second character strings in the fingerprint sample of the template, and determines that the sample message corresponding to the first message and the fingerprint sample is a similar message if the number of successfully matched reference character strings is greater than or equal to a target number or the ratio of the number of successfully matched reference character strings to the number of the plurality of reference character strings is greater than or equal to a target ratio.

In an alternative implementation manner, the server may preset a number threshold for representing the target condition, and when the number of reference character strings for which matching is successful is greater than or equal to the number threshold, it is determined that the first message and the sample message corresponding to the fingerprint sample are similar messages, for example, the number threshold may be set to 10. In another alternative implementation manner, the server may preset a ratio threshold for representing a target condition, and when a ratio of the number of successfully matched reference strings to the number of the plurality of reference strings is greater than or equal to the ratio threshold, determine that the first message and the sample message corresponding to the fingerprint sample are similar messages. For example, the ratio threshold may be set to 80%, and when more than 80% of the plurality of second character strings are successfully matched with the reference character string in the fingerprint sample, the first message and the sample message corresponding to the fingerprint sample are considered to be similar messages. The embodiment of the present application does not specifically limit the specific numerical value and the representation type of the preset threshold.

308. The server determines that the first message belongs to a target type based on the target type of the fingerprint sample.

In the embodiment of the application, after the server determines that the first message and the sample message corresponding to the fingerprint sample are similar messages, it determines that the target type of the first message is consistent with the target type of the fingerprint sample.

In an optional implementation manner, the target type includes an advertisement type or a type carrying malicious and fraudulent content, for example, the server determines that a message type corresponding to the fingerprint sample is the advertisement type, and if the content of the message sample of the fingerprint sample belongs to the advertisement property, it may be determined that the content property of the first message also belongs to the advertisement property. The embodiment of the present application does not specifically limit the determination of the target type.

The server identifies the first message to be issued through the steps 301 and 308, and since the target type of the first message is determined through the fingerprint sample, a model does not need to be trained, and the fingerprint sample has low maintenance cost and small data processing amount, the accuracy and efficiency of message identification can be improved.

Optionally, after the server identifies the first message as the target type, any one of the following processing steps 309 to 311 is further performed on the first message.

309. The server does not issue the first message.

In the embodiment of the application, after the server determines that the first message belongs to the target type, the server rejects the message publishing request sent by the terminal and does not publish the first message.

Step 309 and step 301-.

310. And the server adds first prompt information in the first message to obtain a second message, and issues the second message, wherein the first prompt information is used for prompting that the issued message content has risks.

In the embodiment of the present application, after the server determines that the first message belongs to the target type, the server adds the first prompting message to the first message, for example, the first prompting message is "this message content contains an advertisement, please browse cautiously". The content of the first prompt information is not limited in the present invention. Based on this, the server generates a second message and issues the second message based on the message publication request.

Step 310 and step 301-.

311. The server issues a first message and second prompt information, and the second prompt information is used for prompting that the first message has risks.

In the embodiment of the application, after determining that the first message belongs to the target type, the server issues the first message based on the message issuing request, and at the same time, issues the second prompt message, for example, the second prompt message is "the first message is at risk, please browse carefully". The content of the second prompt message is not limited in the present invention.

Step 310 and step 301-.

It should be noted that in the message identification method formed in steps 301 to 311, the message to be published by the user is identified to determine whether to publish or how to publish, and based on the identification of the message type, the malicious message can be filtered in the message publishing process. In another possible implementation manner, the server extracts the message published by the user from the user message database to identify, so as to determine whether to delete, prompt and the like the published message, thereby filtering malicious messages in the published message. Correspondingly, after the server identifies the first message as the target type, the server also performs processing such as deleting and prompting on the first message, for example, deleting the first message from the user message database, so that other users cannot receive or access the first message any more, for example, adding the first prompting information in the published message, for example, linking the second prompting information in the published message, and if the user opens the published message, synchronously displaying the second prompting information to achieve the warning effect.

Of course, the different embodiments described above may be combined with each other to form different embodiments, which are not specifically limited by the examples of the present invention.

The method for acquiring the fingerprint sample in the above steps is explained below. Fig. 4 is a flowchart of a fingerprint sample acquiring method according to an embodiment of the present application, and as shown in fig. 4, the fingerprint sample acquiring method is described in the embodiment of the present application by taking an application to a server as an example. The method comprises the following steps:

401. the server obtains a sample message of the template, the sample message belonging to the target type.

In the embodiment of the present application, the sample message is a fingerprint original text of the fingerprint sample, and is a template message generated based on a template. That is, the sample message and the first message in the steps 301 and 308 are generated based on the same template. For example, a sample message is generated based on a "credit card swipe template," the contents of which are "[ respected user ] you want value added, [ your account ] you want evening primrose, [ your amount of consumption ] to pay attention to the public number of the XXX teacher.

402. The server deletes meaningless characters in the sample message, fills target symbols in the deleted positions to obtain a plurality of first sample character strings, and separates every two first sample character strings by using the target symbols.

In the embodiment of the application, the server deletes the characters without practical significance in the sample message, including the symbols and the characters which do not belong to the target language, and fills the deleted blank with the target symbols, so that the data volume of subsequent processing can be reduced, and advance preparation can be made for splitting the sample message. The specific implementation manner is consistent with the step of obtaining the plurality of first character strings in step 304, and therefore, the detailed description thereof is omitted here.

403. The server deletes a target character string in the first sample character strings, wherein the target character string is a character string between two target symbols, and the number of words of the character string does not meet a word number threshold value, so that a plurality of second sample character strings are obtained.

In the embodiment of the present application, the process of obtaining the plurality of second sample character strings is the same as the process of obtaining the plurality of second character strings in step 305, and therefore, the description thereof is omitted here.

404. The server obtains a plurality of reference character strings from the plurality of second sample character strings as fingerprint samples of the template.

In the embodiment of the application, after obtaining the plurality of second sample character strings, the server numbers the character strings in the plurality of second sample character strings, and randomly selects the character strings to obtain the plurality of reference character strings as the fingerprint samples of the template.

In an alternative implementation manner, the server may select, as the reference string, a string numbered with an odd number in the plurality of second sample strings, for example, the plurality of second sample strings may be "@ S1@ S2@ S3@ S4@ S5@ S6 @", and the selection of the reference string may be: reference character string 1: "S1", reference character string 2: "S3", reference character string 3: "S5", that is, 3 odd numbered character strings are selected from the 6 character strings as reference character strings to obtain fingerprint samples of the template. The embodiment of the application does not limit the selection mode of the character string.

In an alternative implementation, the server stores the acquired fingerprint samples in the first fingerprint set.

It should be noted that, the foregoing steps 401 to 404 are an implementation manner of obtaining a fingerprint sample from a sample message generated based on a template according to an embodiment of the present application, where the template and the message to be identified belong to the same template. In another alternative implementation manner, the fingerprint sample may be obtained from a sample message generated based on any template, which is not specifically limited in this embodiment of the present application.

In the embodiment of the application, a method for acquiring a fingerprint sample is provided, by deleting characters without actual significance in a sample message, further, text content which does not meet a word number threshold in the sample message is deleted, so that the data volume of subsequent processing is greatly saved, and finally, a plurality of reference character strings for representing edited text content of the sample message are selected as the fingerprint sample for confirming the message type of the first message, so that the accuracy is high, and the maintenance cost of the fingerprint sample is low.

Fig. 5 is a block diagram of a message identification apparatus according to an embodiment of the present application. The apparatus is used for executing the steps when the message identification method is executed, referring to fig. 5, and the apparatus includes: a message acquisition module 501, a sample determination module 502, a string acquisition module 503, a matching module 504, and a type determination module 505.

A message obtaining module 501, configured to obtain a first message, where the first message is generated based on a template;

a sample determination module 502 for determining a fingerprint sample, the fingerprint sample comprising a plurality of reference strings obtained based on a sample message of a target type, the plurality of reference strings being used for representing edited text content of the sample message;

a character string obtaining module 503, configured to obtain a plurality of character strings included in a first message, where the character strings are used to represent edited text content of the first message;

a matching module 504, configured to match the multiple reference character strings with the multiple character strings to obtain a matching result;

and a type determining module 505, configured to determine that the first message belongs to the target type in response to the matching result meeting the target condition.

In an alternative implementation manner, the character string obtaining module 503 includes:

a first deleting unit configured to delete a meaningless character in the first message;

the filling unit is used for filling the target symbol at the deletion position to obtain a plurality of first character strings, and every two first character strings are separated by the target symbol;

and a second deleting unit configured to delete a target character string of the plurality of first character strings, the target character string being a character string in which the number of words between two target symbols does not satisfy a word number threshold, to obtain a plurality of second character strings.

In an alternative implementation manner, the first deleting unit is configured to at least one of:

deleting symbols in the first message;

and deleting characters which do not belong to the target language in the first message, wherein the target language is the language to which the first message belongs.

In an alternative implementation, the matching module 504 is configured to:

for each of the plurality of reference character strings, a character string having the same character string number as the plurality of reference character strings is determined from the plurality of character strings according to the character string number of each of the reference character strings, and the reference character string and the determined character string are matched.

In an optional implementation manner, the message identification apparatus further includes:

a message publishing module to any one of:

not issuing the first message;

adding first prompt information in the first message to obtain a second message, and issuing the second message, wherein the first prompt information is used for prompting that the issued message content has risks;

and issuing a first message and second prompt information, wherein the second prompt information is used for prompting that the first message has risks.

In an alternative implementation manner, the matching result meeting the target condition means:

the ratio of the number of successfully matched reference character strings to the number of the plurality of reference character strings is greater than or equal to the target ratio.

the comparison module is used for periodically comparing a first fingerprint set and a second fingerprint set, wherein the first fingerprint set comprises fingerprint samples of a target type, and the second fingerprint set comprises fingerprint samples of a non-target type;

and the deleting module is used for deleting the coincident fingerprint samples from the first fingerprint set if the coincident fingerprint samples exist in the first fingerprint set and the second fingerprint set.

the system comprises a sample message acquisition module, a template analysis module and a template analysis module, wherein the sample message acquisition module is used for acquiring a sample message of a template, and the sample message belongs to a target type;

the deleting module is used for deleting meaningless characters in the sample messages, filling target symbols in the deleted positions to obtain a plurality of first sample character strings, and separating every two first sample character strings by using the target symbols;

the deleting module is further configured to delete a target sample character string of the plurality of first sample character strings, where the target sample character string is a character string between two target symbols and the number of words does not satisfy a word number threshold, and obtain a plurality of second sample character strings;

and the reference character string obtaining module is used for obtaining a plurality of reference character strings from the plurality of second sample character strings.

In the embodiment of the present application, a message identification apparatus is provided, for a message generated based on a template, a fingerprint sample of the template, that is, a plurality of reference character strings capable of representing edited text content in the message, may be obtained to match with corresponding character strings in the message, so as to determine whether the message belongs to a target type. By adopting the matching between the character strings representing the edited text content, the calculation amount of data processing in the message identification process is reduced, a large amount of model calculation is not needed, the maintenance cost of the fingerprint sample is low, in addition, the method makes full use of the characteristic that the message generated based on the message template has a similar or same grammatical structure, the matching accuracy is high, and the accuracy and the efficiency of the message identification are improved.

It should be noted that: in the message identification device provided in the above embodiment, only the division of the above functional modules is used for illustration when identifying a message, 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 device is divided into different functional modules to complete all or part of the above described functions. In addition, the message identification apparatus and the message identification method provided by the above embodiments belong to the same concept, and specific implementation processes thereof are described in the method embodiments in detail and are not described herein again.

Fig. 6 is a schematic structural diagram of a server according to an embodiment of the present application, where the server 600 may generate a relatively large difference due to different configurations or performances, and can include one or more processors (CPUs) 601 and one or more memories 602, where the memory 602 stores at least one program code, and the at least one program code is loaded and executed by the processors 601 to implement the message identification method provided by the above-mentioned method embodiments. Certainly, the server can also have components such as a wired or wireless network interface, a keyboard, an input/output interface, and the like so as to perform input and output, and the server can also include other components for realizing the functions of the device, which is not described herein again.

The embodiment of the present application further provides a computer-readable storage medium, which is applied to a computer device, and the computer-readable storage medium stores at least one program code, and the at least one program code is loaded and executed by a processor to implement the operations performed by the computer device in the message identification method of the foregoing embodiment.

Embodiments of the present application also provide a computer program product comprising computer program code stored in a computer readable storage medium. The processor of the computer device reads the computer program code from the computer-readable storage medium, and the processor executes the computer program code, causing the computer device to perform the message identification method provided in the various alternative implementations described above.

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, where 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 message identification, the method comprising:

2. The method of claim 1, wherein the obtaining the plurality of strings included in the first message comprises:

deleting meaningless characters in the first message, and filling target symbols at deleted positions to obtain a plurality of first character strings, wherein every two first character strings are separated by the target symbols;

and deleting target character strings in the plurality of first character strings, wherein the target character strings are character strings between two target symbols, and the number of words does not meet a word number threshold value, so as to obtain a plurality of second character strings.

3. The method of claim 2, wherein the deleting of the nonsense character in the first message comprises at least one of:

deleting symbols in the first message;

and deleting characters which do not belong to a target language in the first message, wherein the target language is the language to which the first message belongs.

4. The method of claim 1, wherein the matching the plurality of reference strings and the plurality of strings comprises:

for each reference character string in the plurality of reference character strings, according to the character string number of each reference character string, determining a character string with the same character string number as the character string numbers of the plurality of reference character strings from the plurality of character strings, and matching the reference character string with the determined character string.

5. The method of claim 1, wherein after determining that the first message belongs to the target type, the method further comprises any one of:

not issuing the first message;

adding first prompt information into the first message to obtain a second message, and issuing the second message, wherein the first prompt information is used for prompting that the issued message content has risks;

and issuing the first message and second prompt information, wherein the second prompt information is used for prompting that the first message has risks.

6. The method according to any one of claims 1 to 5, wherein the matching result meeting the target condition is that:

7. The method of claim 1, further comprising:

periodically comparing a first set of fingerprints comprising fingerprint samples of the target type with a second set of fingerprints comprising fingerprint samples of a non-target type;

and if the first fingerprint set and the second fingerprint set have coincident fingerprint samples, deleting the coincident fingerprint samples from the first fingerprint set.

8. The method of claim 1, further comprising:

obtaining a sample message of the template, wherein the sample message belongs to the target type;

deleting meaningless characters in the sample message, filling target symbols at deleted positions to obtain a plurality of first sample character strings, wherein every two first sample character strings are separated by the target symbols;

deleting a target sample character string in the first sample character strings, wherein the target sample character string is a character string between two target symbols, and the number of words does not meet a word number threshold value, so as to obtain a plurality of second sample character strings;

obtaining the plurality of reference character strings from the plurality of second sample character strings.

9. An apparatus for message recognition, the apparatus comprising:

a sample determination module for determining a fingerprint sample, the fingerprint sample comprising a plurality of reference character strings obtained based on a sample message of a target type, the plurality of reference character strings being used for representing edited text content of the sample message;

10. The apparatus of claim 9, wherein the string obtaining module comprises:

a first deleting unit, configured to delete a meaningless character in the first message;

the device comprises a filling unit, a deleting unit and a judging unit, wherein the filling unit is used for filling a target symbol at a deleting position to obtain a plurality of first character strings, and every two first character strings are separated by the target symbol;

and the second deleting unit is used for deleting a target character string in the plurality of first character strings, wherein the target character string is a character string between two target symbols, and the number of words does not meet a word number threshold value, so that a plurality of second character strings are obtained.

11. The apparatus of claim 10, wherein the first deleting unit is configured to at least one of:

deleting symbols in the first message;

12. The apparatus of claim 9, wherein the matching module is configured to:

13. The apparatus of any one of claims 9 to 12, further comprising:

a message publishing module to any one of:

not issuing the first message;

14. A computer device, characterized in that the computer device comprises a processor and a memory for storing at least one piece of program code, which is loaded by the processor and which performs the message identification method of any of claims 1 to 8.

15. A computer-readable storage medium for storing at least one program code for performing the message identification method of any of claims 1 to 8.