CN116738407A

CN116738407A - Method, device and medium for determining abiotic user

Info

Publication number: CN116738407A
Application number: CN202311018887.2A
Authority: CN
Inventors: 李丹; 肖新光
Original assignee: Beijing Antiy Network Technology Co Ltd
Current assignee: Beijing Antiy Network Technology Co Ltd
Priority date: 2023-08-14
Filing date: 2023-08-14
Publication date: 2023-09-12
Anticipated expiration: 2043-08-14
Also published as: CN116738407B

Abstract

The invention relates to a method, a device and a medium for determining a non-biological user, which are applied to first equipment, wherein a target program for information transmission is installed in the first equipment; comprising the following steps: responding to the information transmission between the first equipment and the second equipment through the target program, and acquiring information to be received, which is sent to the second equipment by the first equipment; adding hidden information into the information to be received to obtain target information; the hidden information is information which can trigger the artificial intelligence program to perform interactive response and cannot be recognized by human beings; transmitting the target information to a second device; acquiring information to be identified, which is returned by the second equipment and aims at the target information; and if the information to be identified contains information corresponding to the hidden information, judging that the user of the target program of the second equipment is a non-biological user. The invention can accurately detect the abiotic user, and carry out risk prompt for the user carrying out information interaction with the abiotic user so as to protect the information and the asset safety.

Description

Method, device and medium for determining abiotic user

Technical Field

The present invention relates to the field of information security technologies, and in particular, to a method, an apparatus, and a medium for determining a non-biological user.

Background

With the rapid development of new generation information technology and application, new technology and new application such as artificial intelligence bring convenience to work, life and study of people, and meanwhile, the new technology and new application can be utilized by others to become tools for stealing information, phishing and the like, so that great risks and challenges are brought to network security and social security. Most of the current technical research fields are to research the optimization and application of the artificial intelligence technology, so that the research on behavior detection is less, and the artificial intelligence technology is difficult to effectively detect due to the strong humanoid capability, which tends to cause the increase of related application risks, and is very unfavorable for the information and asset security of individuals and enterprise institutions.

Disclosure of Invention

In view of the above, the present invention provides a method, apparatus and medium for determining an abiotic user, which can effectively detect an abiotic user of artificial intelligence type, help a person or enterprise organization user identify an artificial intelligence application risk, avoid a security trap, and at least partially solve the problems existing in the prior art.

The specific invention comprises the following steps:

the method for determining the non-biological user is applied to a first device, wherein a target program for information transmission is installed in the first device; the method comprises the following steps:

Responding to the information transmission between the first equipment and the second equipment through the target program, and acquiring information to be received, which is sent to the second equipment by the first equipment; and the second equipment is internally provided with the target program so as to realize information transmission with the first equipment.

Adding hidden information into the information to be received to obtain target information; the hidden information is information which can trigger the artificial intelligent program to perform interactive response and cannot be recognized by human beings; and the target information is used for replacing the information to be received and sending the information to the second equipment.

And sending the target information to the second equipment.

And acquiring information to be identified aiming at the target information, which is returned by the second equipment after receiving the target information.

And if the information to be identified contains information corresponding to the hidden information, judging that the user of the target program of the second equipment is a non-biological user.

Further, after the obtaining the information to be received sent by the first device to the second device, the method further includes:

determining the category of the information to be received, and adding hidden information of a corresponding category into the information to be received according to the category of the information to be received; the categories include: text, image, audio.

Further, if the category of the information to be received is text, a process of adding hidden information into the information to be received includes:

and carrying out semantic analysis on the information to be received to obtain semantic features corresponding to the information to be received.

Determining semantic features corresponding to each piece of text information in a preset text library; the preset text library comprises a plurality of pieces of text information, and each piece of text information can trigger an artificial intelligent program to perform interactive response.

And calculating the similarity of the semantic features corresponding to the information to be received and the semantic features corresponding to each piece of text information, and determining the text information corresponding to the similarity calculation result smaller than a preset threshold value as target text information.

And determining a piece of text information from a plurality of pieces of target text information as text information to be added into the information to be received.

And determining an interface format of the information transmission interface of the target program.

And converting the format of the text information to be added into the information to be received according to the interface format to obtain hidden information.

Adding the hidden information to a target position of the information to be received so as to obtain the target information; the target location includes: first character, last character, target character.

and segmenting the information to be received to obtain a word sequence corresponding to the information to be received.

Carrying out grammar analysis on the information to be received based on the word sequence to obtain a grammar information sequence corresponding to the word sequence; the grammar information sequence comprises a plurality of grammar information, the grammar information corresponds to words in the word sequence one by one, and each grammar information comprises: word parts and labels of words which can be linked before and after the word corresponding to the grammar information; the tag is used for marking features of words, and the features comprise: characters, animals, food, tools, books, landscapes.

Determining the part of speech and the label of the addable word corresponding to each target position according to the word sequence and the grammar information sequence to obtain the addition requirement when each target position carries out word addition; the target location includes before the first word, after the last word, between every two words in the sequence of words.

At least one of the target locations is determined as an addition location.

Screening words meeting the addition requirements when adding words at the addition positions from a preset word stock, and adding the screened words to the corresponding addition positions to obtain a plurality of target word sequences; the preset word stock comprises a plurality of words which are used for being added into the information to be received as hidden information.

And connecting words in each target word sequence according to the sequence order to obtain a target text corresponding to each target word sequence.

And carrying out semantic analysis on each target text to obtain semantic features corresponding to each target text.

And calculating the similarity between the semantic features corresponding to the information to be received and the semantic features corresponding to each target text, and determining the target text corresponding to the minimum similarity calculation result as the text to be sent.

Determining information display attributes of the target program; the information display attributes comprise background color, shading color, brightness and transparency of the display window.

And adjusting the display attribute of each added word contained in the text to be sent according to the information display attribute to obtain the target information, so that the added word contained in the target information cannot be recognized by human eyes when the target information is displayed on a display window of the target program.

Further, the determining the part of speech and the tag of the addable word corresponding to each target position according to the word sequence and the grammar information sequence includes:

Determining first information and second information corresponding to each grammar information according to the grammar information sequence; the first information is the part of speech and the label of the word which can be linked before the word, and the second information is the part of speech and the label of the word which can be linked after the word.

And determining the part of speech and the label of the word which can be added before the first word in the word sequence according to the first information corresponding to the first grammar information in the grammar information sequence.

And respectively determining intersections of second information corresponding to each grammar information in the grammar information sequence and first information corresponding to the next grammar information to obtain parts of speech and labels of words which can be added between every two words in the word sequence.

And determining the part of speech and the label of the word which can be added after the last word in the word sequence according to the second information corresponding to the last grammar information in the grammar information sequence.

Further, adding the screened word to the corresponding addition location includes:

and determining a plurality of screened words corresponding to each adding position to obtain a word set corresponding to each adding position.

And selecting a word from each word set to enumerate and combine to obtain a plurality of combined sequences.

The words in each combined sequence are added to the corresponding addition positions so as to obtain the target word sequences.

Further, if the category of the information to be received is an image, a process of adding hidden information to the information to be received includes:

and identifying the image characteristics of the information to be received to obtain the image characteristic vector corresponding to the information to be received.

Determining an image feature vector of each image data in a preset image library; the preset image library comprises a plurality of image data, and each image data can trigger an artificial intelligent program to perform interactive response; the image data is used to be added as hidden information to the information to be received.

And calculating the similarity between the image feature vector corresponding to the information to be received and the image feature vector of each image data, and determining the image data corresponding to the minimum similarity calculation result as a watermark image.

Determining the image attribute of the information to be received; the image attributes include image size and display attributes including: resolution, pixels.

And adjusting the image attribute of the watermark image according to the image attribute of the information to be received so that the image size of the watermark image is not larger than the image size of the information to be received, and the display attribute is the same as the display attribute of the information to be received.

And adjusting the transparency of the watermark image according to a first preset threshold value so that the watermark image cannot be recognized by human eyes.

And carrying out image superposition on the watermark image and the information to be received so as to obtain the target information.

Further, if the category of the information to be received is audio, a process of adding hidden information into the information to be received includes:

and identifying the information to be received, and determining voiceprint characteristics and semantic information of the information to be received.

And determining semantic features corresponding to the information to be received according to the semantic information of the information to be received.

Determining semantic features corresponding to each piece of audio data in a preset audio library; the audio library comprises a plurality of audio data, and each audio data can trigger an artificial intelligent program to perform interactive response; the audio data is used as hidden information to be added to the information to be received.

And calculating the similarity between the semantic features corresponding to the information to be received and the semantic features corresponding to each piece of audio data, and determining the audio data corresponding to the minimum similarity calculation result as target audio.

And converting the target audio according to the voiceprint characteristics of the information to be received so that the voiceprint characteristics of the target audio are identical to the voiceprint characteristics of the information to be received.

Adjusting the frequency of the target audio to a target frequency band; the target frequency band is a frequency band which cannot be perceived by human ears.

Adding the target audio to a target position of the information to be received so as to obtain the target information; the target position comprises positions without effective signals before and after the information to be received starts.

The position where the information to be received has no effective signal is determined by the following steps:

and carrying out Fourier transform on the information to be received to obtain a frequency domain signal corresponding to the information to be received.

And calculating the energy spectrum density of the frequency domain signal, and determining a frequency interval with energy lower than a second preset threshold value as the position where the information to be received has no effective signal.

Further, before the obtaining the information to be received, which is sent by the first device to the second device, the method further includes:

information sent by a prescribed number of second devices to the first device is acquired.

And carrying out word segmentation statistics on the acquired information, and determining target words with word frequency larger than a third preset threshold value.

And matching the target word with words in a preset keyword library.

And if the matching is successful, acquiring information to be received, which is sent to the second equipment by the first equipment.

Further, the information to be identified contains information corresponding to the hidden information, and the information to be identified is judged by the following steps:

determining keywords of reply information generated by the artificial intelligence program after triggering interactive response of the artificial intelligence program according to the hidden information; the key word of the reply information is information corresponding to the hidden information.

And if the information to be identified contains the key word of the reply information, judging that the information to be identified contains the information corresponding to the hidden information.

A determination apparatus of a non-biological user, applied to a first device in which a target program for information transmission is installed; the device comprises:

the information acquisition module is used for responding to the information transmission between the first equipment and the second equipment through the target program and acquiring information to be received, which is sent to the second equipment by the first equipment; and the second equipment is internally provided with the target program so as to realize information transmission with the first equipment.

The target information generation module is used for adding hidden information into the information to be received to obtain target information; the hidden information is information which can trigger the artificial intelligent program to perform interactive response and cannot be recognized by human beings; and the target information is used for replacing the information to be received and sending the information to the second equipment.

And the information sending module is used for sending the target information to the second equipment.

The user determining module is used for acquiring information to be identified aiming at the target information, which is returned by the second equipment after receiving the target information; and if the information to be identified contains information corresponding to the hidden information, judging that the user of the target program of the second equipment is a non-biological user.

A non-transitory computer readable storage medium having stored therein at least one instruction or at least one program loaded and executed by a processor to implement the foregoing method.

The beneficial effects of the invention are as follows:

the invention has wide application scene, can be used for transmitting the existence information of chatting, meeting, network shopping and the like, can be applied to various network environments such as the Internet, local area network, ad hoc network and the like, and the user of the first equipment can be a person or an enterprise mechanism. The invention can accurately detect the abiotic user, carry out risk prompt for the user carrying out information interaction with the abiotic user, help the individual or enterprise organization user to make correct judgment so as to protect information and asset security, and simultaneously provide effective guarantee for maintaining network security and social security. The invention can realize the addition of the hidden information under the condition that the user has no sense, and minimize the influence on the user experience of the first equipment side.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for determining a non-biological user according to an embodiment of the present invention;

FIG. 2 is a flowchart of another method for determining a non-biological user according to an embodiment of the present invention;

FIG. 3 is a flowchart of a method for adding hidden information to be received in a method for determining a non-biological user according to an embodiment of the present invention;

FIG. 4 is a flowchart of another method for adding hidden information to be received in a method for determining a non-biological user according to an embodiment of the present invention;

FIG. 5 is a flowchart of another method for adding hidden information to be received in a method for determining a non-biological user according to an embodiment of the present invention;

FIG. 6 is a flowchart of another method for adding hidden information to be received in a method for determining a non-biological user according to an embodiment of the present invention;

Fig. 7 is a block diagram of a determination device for a non-biological user according to an embodiment of the present invention.

Detailed Description

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

It should be noted that, without conflict, the following embodiments and features in the embodiments may be combined with each other; and, based on the embodiments in this disclosure, all other embodiments that may be made by one of ordinary skill in the art without inventive effort are within the scope of the present disclosure.

It is noted that various aspects of the embodiments are described below within the scope of the following claims. It should be apparent that the aspects described herein may be embodied in a wide variety of forms and that any specific structure and/or function described herein is merely illustrative. Based on the present disclosure, one skilled in the art will appreciate that one aspect described herein may be implemented independently of any other aspect, and that two or more of these aspects may be combined in various ways. For example, an apparatus may be implemented and/or a method practiced using any number of the aspects set forth herein. In addition, such apparatus may be implemented and/or such methods practiced using other structure and/or functionality in addition to one or more of the aspects set forth herein.

The invention provides a non-biological user determining method embodiment, which is applied to first equipment, wherein a target program for information transmission is installed in the first equipment; the first equipment comprises equipment such as a mobile phone and a computer which can install and run the target program; an embodiment of the method is shown in fig. 1, and includes:

s11: responding to the information transmission between the first equipment and the second equipment through the target program, and acquiring information to be received, which is sent to the second equipment by the first equipment; the second equipment is internally provided with the target program so as to realize information transmission with the first equipment; the method for acquiring the information to be received comprises the following steps: and the information is acquired through a display window and an information transmission interface.

S12: adding hidden information into the information to be received to obtain target information; the hidden information is information which can trigger the artificial intelligent program to perform interactive response and cannot be recognized by human beings; and the target information is used for replacing the information to be received and sending the information to the second equipment.

S13: and sending the target information to the second equipment.

S14: and acquiring information to be identified aiming at the target information, which is returned by the second equipment after receiving the target information.

S15: and if the information to be identified contains information corresponding to the hidden information, judging that the user of the target program of the second equipment is a non-biological user.

The embodiment shown in fig. 1 obtains the information to be received after the first device confirms that the information is sent to the second device, and adds the hidden information to the information to be received, so that the hidden information can be added under the condition that the user has no sense, and the influence on the user experience of the first device is minimized. The target information obtained after the hidden information is added is used for replacing the information to be received and sending the information to the second device, and because the hidden information is the information which can trigger the interaction response of the artificial intelligent program and cannot be identified by human beings, if the user of the target program of the second device is the human beings, the hidden information in the target information cannot be judged and the corresponding reply is made, and the corresponding reply can be made only for the information to be received in the target information; if the user of the target program of the second device is a non-biological user of the artificial intelligence type, namely an artificial intelligence application, an artificial intelligence program and the like, the hidden information can be accurately identified and corresponding replies can be made; therefore, the information to be identified for the target information returned by the second device after receiving the target information is obtained, and if the information to be identified contains information corresponding to the hidden information, the user of the target program of the second device is judged to be a non-biological user. Since the non-biological user is a human-controlled application or program, which itself has no intelligent judgment capability, information interaction with the non-biological user is at a security risk. After step S15 of the embodiment shown in fig. 1, an early warning prompt process may be added according to an actual application scenario, for example, after it is determined that the user of the target program of the second device is a non-biological user, safety prompt information is sent to the user of the target program of the first device, for example, "monitor that the other party is a non-biological user, please alert about the risk of information transmission, do not trust the information content sent by the other party", etc.

The embodiment of fig. 1 has a very wide application scenario, and may be used in chat, conference, online shopping, etc. in the scenario of presence information transmission, and may be applied in various network environments such as the internet, a local area network, an ad hoc network, etc., where the user of the first device may be a person or an enterprise organization. The embodiment shown in fig. 1 can accurately detect non-biological users, prompt risks for users who interact with the non-biological users, help individuals or enterprise organization users to make correct judgment, protect information and asset security, and provide effective guarantee for maintaining network security and social security.

Preferably, as shown in fig. 2, after the obtaining the information to be received sent by the first device to the second device in step S11, the method further includes:

s111: determining the category of the information to be received, and adding hidden information of a corresponding category into the information to be received according to the category; the categories include: text, image, audio. The method and the device can ensure that the hidden information and the information to be received are fused better, the information to be received is taken as a carrier to transmit the hidden information better, the difficulty of the second equipment in reverse analysis of the hidden information is increased to a certain extent, and the accuracy of whether the user of the target program of the second equipment is a non-biological user judgment result is ensured.

Preferably, in combination with the foregoing preferred solution, if the category of the information to be received is text, an embodiment of a method for adding hidden information to the information to be received in a method for determining a non-biological user is provided, as shown in fig. 3, including:

s31: and carrying out semantic analysis on the information to be received to obtain semantic features corresponding to the information to be received. The semantic analysis of the information to be received can be completed through a semantic analysis algorithm or model.

S32: determining semantic features corresponding to each piece of text information in a preset text library; the preset text library comprises a plurality of pieces of text information, and each piece of text information can trigger an artificial intelligent program to perform interactive response; the text information is exemplified as follows: how weather is today, what breakfast is, how open the day is to learn, and what color is to be covered.

The semantic features in step S31 and step S32 are expressed in the form of spatial vectors, each spatial vector is determined according to the word vector of each word included in the corresponding information to be received or the text information, and the determining manner includes: the mean value of each word vector is calculated, or each word vector is calculated by using a model. The semantic features corresponding to each piece of text information in the preset text library can be calculated in advance and then stored in the appointed position such as the database, and the process of the step S32 can be realized by acquiring the semantic features from the appointed position, so that the calculation power consumption is reduced, and the calculation efficiency is improved.

S33: and calculating the similarity of the semantic features corresponding to the information to be received and the semantic features corresponding to each piece of text information, and determining the text information corresponding to the similarity calculation result smaller than a preset threshold value as target text information. The preset threshold value may be set according to actual requirements, for example, 0.1, 0.2, etc., and since the higher the similarity between the semantic feature corresponding to the information to be received and the semantic feature corresponding to each piece of text information is, the higher the obtained value of the similarity calculation result is, so that the smaller the value set by the preset threshold value is, the greater the assistance to the accuracy of the final non-biological user determination result is.

S34: and randomly determining a piece of text information from a plurality of pieces of target text information as the text information to be added into the information to be received.

The text information to be added to the information to be received and the information to be received determined in the steps S33 and S34 are greatly distinguished from each other in terms of semantics, so that the reply information can contain information which does not correspond to the information to be received completely after the target information is identified by the artificial intelligence program, and the situation that the artificial intelligence program merges and replies the content of the information to be received and the content of the hidden information, so that the information corresponding to the hidden information cannot be distinguished from the information to be identified is avoided. The text information to be added to the information to be received is determined through steps S33 and S34, which is beneficial to ensuring the accuracy of the non-biological user determination result. In the actual application, the steps S33 and S34 may be adjusted according to the requirement, for example, the text information corresponding to the minimum similarity calculation result is directly determined as the text information to be added into the information to be received.

S35: and determining an interface format of the information transmission interface of the target program.

S36: and converting the format of the text information to be added into the information to be received according to the interface format to obtain hidden information. Step S36 converts the format of the text information to be added to the information to be received into the interface format, so that the text information to be added to the information to be received is not displayed on the display interface of the target program, the purpose of transmitting is completed only through the interface, interference to the user of the first device is avoided, and the influence on the user experience of the first device side is minimized.

S37: adding the hidden information to a target position of the information to be received so as to obtain the target information; the target location includes: first character, last character, target character. The target character comprises a line feed character, a period character, a special character or the like.

The invention is exemplified in connection with the embodiment described in fig. 3 as follows:

the information to be received is "which play is going to today? The "after step S31, step S32 and step S33, the determined target text information includes" what color is covered "," how is the read "," how is you will be the haircut "," how is the read "is randomly determined as the text information to be added to the information to be received", then the data format of "how is the read" is converted according to the interface format of the target program, and finally "how is the read" after the format conversion is added to "what is the play going today? After the last character of "to get the target information" which play today? "how read" is hidden information, where " how read" is unrecognizable by human beings, if the user of the target program of the second device is a person or an enterprise user, the information returned to the first device after receiving the target information does not include information corresponding to " how read", if the user of the target program of the second device is an artificial intelligence type non-biological user, the user of the target program of the second device may recognize the hidden information, and answer " how read" to the information returned to the first device after receiving the target information, that is, the returned information includes information corresponding to " how read", and then it may be determined that the user of the target program of the second device is a non-biological user, and the information transmission between the user corresponding to the first device and the second device has a security risk.

The embodiment of fig. 3 shows an implementation manner of adding hidden information in to-be-received information in a non-biological user determining method when the type of to-be-received information is text, first determining a plurality of target text information with larger semantic difference from the to-be-received information in a preset text library, then randomly determining text information to be added into the to-be-received information in a plurality of target text information, then converting the format of the text information to be added into the to-be-received information into an interface format corresponding to the target program, obtaining the hidden information, and finally adding the hidden information into a target position of the to-be-received information, so as to obtain the target information. The embodiment shown in fig. 3 can achieve the purpose that the hidden information is not displayed on the display interface of the target program, and is only transmitted through the interface, so that the user of the first device is not interfered, and the influence on the user experience of the first device side is reduced to the minimum. Meanwhile, after the target information is identified, the information which is not corresponding to the information to be received can be contained in the reply information by the artificial intelligence program, so that the artificial intelligence program is prevented from merging and replying the content of the information to be received and the content of the hidden information, the information corresponding to the hidden information is not distinguished from the information to be identified, and the accuracy of the non-biological user judgment result is ensured.

Preferably, in combination with the foregoing preferred solution, if the category of the information to be received is text, another embodiment of a method for adding hidden information to the information to be received in a method for determining a non-biological user is provided, as shown in fig. 4, including:

s41: and segmenting the information to be received to obtain a word sequence corresponding to the information to be received. The word segmentation of the information to be received is completed through a word segmentation algorithm.

S42: carrying out grammar analysis on the information to be received based on the word sequence to obtain a grammar information sequence corresponding to the word sequence; the grammar information sequence comprises a plurality of grammar information, the grammar information corresponds to words in the word sequence one by one, and each grammar information comprises: word parts and labels of words which can be linked before and after the word corresponding to the grammar information; the tag is used for marking the characteristics of the word, the characteristics of the word correspond to the meaning of the word, and the characteristics comprise: characters, animals, foods, tools, books, landscapes, etc., such as: the label corresponding to the term "wheat wave" is "landscape" and the label corresponding to the term "Tiananmen" is "place". The parsing of the information to be received may be accomplished by a parsing algorithm or model. The parts of speech include: nouns, verbs, adjectives, adverbs, real words, imaginary words, personification, and exclamation.

S43: determining the part of speech and the label of the addable word corresponding to each target position according to the word sequence and the grammar information sequence to obtain the addition requirement when each target position carries out word addition; the target location includes before the first word, after the last word, between every two words in the sequence of words.

S44: at least one of the target locations is determined as an addition location. When adding words, each target position can be added, one or more positions can be determined to be added, and the adding positions can be determined randomly.

S45: screening words meeting the addition requirements when adding words at the addition positions from a preset word stock, and adding the screened words to the corresponding addition positions to obtain a plurality of target word sequences; the preset word stock comprises a plurality of words which are used for being added into the information to be received as hidden information, and each word is marked with a corresponding part of speech and a label.

S46: and connecting words in each target word sequence according to the sequence order to obtain a target text corresponding to each target word sequence.

The part of speech and the label of the addable words corresponding to each target position can be determined through the steps S41 to S45, so that the adding requirement when the words are added at each target position is obtained, namely the adding requirement comprises the finally determined part of speech and label, further, words meeting the adding requirement can be rapidly screened out from a preset word stock, after screening is finished, the accurate words obtained from the preset word stock are added to the adding position, the meaning of the target text obtained after the words are added at each adding position can be ensured to be smooth and correct, no grammar error exists, the condition that the violations and the words are added at the adding position is avoided, the interactive correspondence of an artificial intelligent program can be triggered more accurately, and the accuracy of the non-biological user judging result is ensured.

S47: and carrying out semantic analysis on the information to be received to obtain semantic features corresponding to the information to be received.

S48: and carrying out semantic analysis on each target text to obtain semantic features corresponding to each target text.

S49: and calculating the similarity between the semantic features corresponding to the information to be received and the semantic features corresponding to each target text, and determining the target text corresponding to the minimum similarity calculation result as the text to be sent. Step S49 determines the target text with the largest semantic difference from the information to be received as the text to be transmitted, so that it can be ensured that the artificial intelligence program can contain the information which does not correspond to the information to be received in the reply information after identifying the target information, and the artificial intelligence program is prevented from merging and replying the contents of the information to be received and the hidden information, so that the information corresponding to the hidden information cannot be distinguished from the information to be identified. The determination of the information to be transmitted in step S49 is advantageous in ensuring the accuracy of the non-biological user determination result.

S410: determining information display attributes of the target program; the information display attributes comprise background color, shading color, brightness and transparency of the display window.

S411: and adjusting the display attribute of each added word contained in the text to be sent according to the information display attribute to obtain the target information, so that when the target information is displayed on a display window of the target program, the added word contained in the target information cannot be recognized by human eyes, interference to a user of the first equipment is avoided, and influence on user experience of the first equipment side is reduced to the minimum.

The invention is exemplified in connection with the embodiment described in fig. 4 as follows:

the information to be received is "which play today", the word sequence obtained after word segmentation is "today" →which play ". After grammar analysis, a grammar information sequence containing grammar information corresponding to each word in the word sequence is obtained, and the grammar information corresponding to the word sequence is" noun ((pre-noun, adjective), (character, animal)), (post (noun), (character, animal))) ", wherein the first" noun "represents the part of speech of the word of" today "," (pre-noun, adjective), (character, animal)) "represents the part of speech and labels of words which can be added before the word of" today "," (post-noun), (character, animal)) "represents the part of speech and labels of words which can be added after the word of" today "," which play "," constitutes the grammar information sequence corresponding to the word sequence. Before randomly determining that the adding position is today, acquiring words corresponding to the adding position from a preset word stock, namely 'Xiaoming', 'rabbit', 'teacher', correspondingly acquiring three target texts, namely 'Xiaoming today' going to which play ',' rabbit today 'going to which play' and 'teacher today' going to which play ', determining' rabbit today 'going to which play' as a text to be sent after semantic analysis and semantic similarity calculation, and then adjusting the display attribute of the 'rabbit' word in the text to be sent according to the information display attribute of the target program, so that the 'rabbit' word becomes hidden information which cannot be recognized by human eyes. If the information returned by the second device after receiving the target information contains information corresponding to the term "rabbit", if "I don't know what play the rabbit plays today," then it can be determined that the user of the target program of the second device is a non-biological user, and the security risk exists in the information transmission between the user corresponding to the first device and the second device.

The embodiment shown in fig. 4 shows an implementation manner of adding hidden information in to-be-received information in another method for determining a non-biological user when the type of to-be-received information is text, firstly, word segmentation and grammar analysis are performed on the to-be-received information, part of speech and labels of the addable words corresponding to each target position are determined, then the adding position is determined, words corresponding to the adding position are obtained from a preset word stock and added into the adding position, a plurality of target texts are obtained, the text to be sent is determined as the text to be sent with the largest semantic difference with the to-be-received information in the target texts, and finally, the display attribute of the added words in the text to be sent is adjusted according to the information display attribute of the target program, so that the target information is obtained. The embodiment shown in fig. 4 can ensure that the semantics of the target text obtained after adding the words at each addition position are smooth and correct, have no grammar error, avoid the condition of adding the offence words at the addition position, more accurately trigger the interaction correspondence of the artificial intelligence program, and ensure the accuracy of the non-biological user judgment result. The embodiment shown in fig. 4 can ensure that the artificial intelligence program can contain the information which does not correspond to the information to be received in the reply after identifying the target information, so that the artificial intelligence program is prevented from merging and replying the content of the information to be received and the content of the hidden information, and the information corresponding to the hidden information is not distinguished in the information to be identified, and the accuracy of the non-biological user judgment result is further ensured.

Preferably, the determining the part of speech and the tag of the addable word corresponding to each target position according to the word sequence and the grammar information sequence includes:

In combination with the foregoing examples, the preferred embodiments described above are exemplified as follows:

the information to be received is "which play today," the word sequence obtained after word segmentation is "today- & gtwhich play- & gtthe word sequence is" and the grammar information sequence containing the grammar information corresponding to each word in the word sequence is obtained after grammar analysis, wherein the grammar information corresponding to the word of "today" is "noun ((pro (noun, adjective), (character, animal))", "post (noun), (character, animal)))", the first information corresponding to the grammar information is ((noun, adjective), (character, animal)), and the second information corresponding to the grammar information is ((noun), (character, animal)). Since the grammar information corresponding to the word "today" is the first grammar information in the current grammar information sequence, the part of speech of the word which can be added before the first word in the current word sequence is determined to be "noun" or "adjective" and the label is "person" or "animal". Assuming that the first information corresponding to "which" word is going is ((noun), (person)), the intersection of the second information corresponding to "today" word and the first information corresponding to "which" word is going is ((noun), (person)), it can be determined that the part of speech of the word which can be added between "today" and "which" word is going is "noun", and the label is "person". Assuming that the second information corresponding to the "play" word is ((noun), (toy)), since the grammar information corresponding to the "play" word is the last grammar information in the current grammar information sequence, the part of speech of the word which can be added after the last word in the current word sequence is determined as "noun", and the tag is "toy".

Preferably, adding the screened word to the corresponding addition location includes:

Preferably, in combination with the foregoing preferred solution, if the category of the information to be received is an image, another embodiment of a method for adding hidden information to the information to be received in a method for determining a non-biological user is provided, as shown in fig. 5, including:

s51: and identifying the image characteristics of the information to be received to obtain the image characteristic vector corresponding to the information to be received. The identification of the image features of the information to be received can be accomplished by an image feature identification algorithm or model.

S52: determining an image feature vector of each image data in a preset image library; the preset image library comprises a plurality of image data, each image data can trigger an artificial intelligent program to perform interactive response, and the image data comprises text information which can trigger the artificial intelligent interactive response after being identified by a computer program, for example: how weather is today, what breakfast is, how open the day is to learn, and what color is to be covered; the image data is used to be added as hidden information to the information to be received.

S53: and calculating the similarity between the image feature vector corresponding to the information to be received and the image feature vector of each image data, and determining the image data corresponding to the minimum similarity calculation result as a watermark image. The image itself contains information such as characters, animals, scenery, text and the like, and the step S53 determines the image data corresponding to the minimum similarity calculation result as a watermark image, so that it can be ensured that the artificial intelligence program can contain information which does not correspond to the information to be received in the reply information after identifying the target information, and the situation that the information corresponding to the hidden information is not distinguished from the information to be identified is avoided, thereby being beneficial to ensuring the accuracy of the non-biological user judgment result.

S54: determining the image attribute of the information to be received; the image attributes include image size and display attributes including: resolution, pixels.

S55: and adjusting the image attribute of the watermark image according to the image attribute of the information to be received so that the image size of the watermark image is not larger than the image size of the information to be received, and the display attribute is the same as the display attribute of the information to be received.

S56: and adjusting the transparency of the watermark image according to a first preset threshold value so that the watermark image cannot be recognized by human eyes. The first preset threshold is determined according to practical application requirements, for example, 5%, 3% and the like, and the lower the numerical value is, the lower the visual recognition rate of the watermark image is, generally, the lower the transparency is, the visual recognition is impossible.

S57: and carrying out image superposition on the watermark image and the information to be received so as to obtain the target information.

The invention is exemplified in connection with the embodiment described in fig. 5 as follows:

and after the information to be received is a human image, determining that an image data is a watermark image from a preset image library through image feature vector similarity calculation, wherein the image data is text image data, and after the computer program identifies that the text information is 'fig mountain in which province', adjusting the image attribute of the watermark image according to the image attribute of the information to be received so that the image size of the watermark image is one eighth of the image size of the information to be received, wherein the display attribute is the same as the display attribute of the information to be received, and finally, carrying out image superposition on the watermark image and the information to be received to obtain target information. If the information returned by the second device after receiving the target information contains the information corresponding to the 'flower and fruit mountain in which province', the user of the target program of the second device can be judged to be a non-biological user, and the safety risk exists in the information transmission between the user corresponding to the first device and the second device.

The embodiment shown in fig. 5 shows an implementation manner of adding hidden information in to-be-received information in a non-biological user determining method when the type of to-be-received information is an image, firstly, performing image feature recognition on the to-be-received information to obtain corresponding image feature vectors, then determining image data with minimum similarity with the image feature vectors of the to-be-received information from a preset image library as a watermark image, and finally, adjusting display attribute and transparency of the watermark image and performing image superposition with the to-be-received information to obtain the target information. The embodiment shown in fig. 5 can ensure that after identifying the target information, the artificial intelligence program can include information that does not correspond to the information to be received in the reply information, so that the artificial intelligence program is prevented from merging and replying the content of the information to be received and the content of the hidden information, so that the information corresponding to the hidden information cannot be distinguished from the information to be identified, and the accuracy of the non-biological user judgment result is further ensured.

Preferably, in combination with the foregoing preferred solution, if the category of the information to be received is audio, another embodiment of a method for adding hidden information to the information to be received in a method for determining a non-biological user is provided, as shown in fig. 6, including:

s61: and identifying the information to be received, and determining voiceprint characteristics and semantic information of the information to be received.

S62: and determining semantic features corresponding to the information to be received according to the semantic information of the information to be received.

S63: determining semantic features corresponding to each piece of audio data in a preset audio library; the audio library comprises a plurality of audio data, and each audio data can trigger an artificial intelligence program to perform interactive response, such as audio containing contents of how the weather is today, what is eaten in breakfast, how to learn in open days, what color is sleeved outside, and the like; the audio data is used as hidden information to be added to the information to be received.

S64: and calculating the similarity between the semantic features corresponding to the information to be received and the semantic features corresponding to each piece of audio data, and determining the audio data corresponding to the minimum similarity calculation result as target audio. The step S64 determines the audio data corresponding to the minimum similarity calculation result as the target audio, so that it can be ensured that the artificial intelligence program can include the information which does not correspond to the information to be received in the reply information after identifying the target information, avoiding that the information corresponding to the hidden information is not distinguished from the information to be identified, and being beneficial to ensuring the accuracy of the non-biological user determination result.

S65: and converting the target audio according to the voiceprint characteristics of the information to be received so that the voiceprint characteristics of the target audio are identical to the voiceprint characteristics of the information to be received. Step S65 can ensure that the hidden information and the information to be received are better fused, and the hidden information is better transmitted by taking the information to be received as a carrier, so that the difficulty of the second device in reversely analyzing the hidden information is increased to a certain extent, and the accuracy of the judgment result of whether the user of the target program of the second device is a non-biological user is ensured.

S66: adjusting the frequency of the target audio to a target frequency band; the target frequency band is a frequency band which cannot be perceived by human ears. The frequency of the sound wave vibration can be perceived by the human ear in the range of 20-20000 hz, and therefore, the target frequency band includes a frequency band having a frequency lower than 20 hz and a frequency band having a frequency exceeding 20000 hz.

S67: adding the target audio to a target position of the information to be received so as to obtain the target information; the target position comprises positions without effective signals before and after the information to be received starts.

And calculating the energy spectrum density of the frequency domain signal, and determining a frequency interval with energy lower than a second preset threshold value as the position where the information to be received has no effective signal. The second preset threshold is set according to the actual application requirement, and because noise data are unavoidable in the audio, the second preset threshold is a non-zero value.

Preferably, before the obtaining the information to be received sent by the first device to the second device in step S11, the method further includes:

information transmitted by a prescribed number of second devices to the first device is acquired, such as 100 pieces of information transmitted by the second devices to the first device.

Performing word segmentation statistics on the acquired information, and determining target words with word frequency greater than a third preset threshold value; the third preset threshold is set according to practical application requirements, for example, 30 times, 50 times and the like.

And matching the target word with words in a preset keyword library. The preset keyword library stores anti-fraud keywords including insurance, stocks, funds, investment, purchase and the like.

The above preferred solution is that a trigger condition is added before the information to be received sent by the first device to the second device is obtained in step S11, the information transmission behavior of the second device is judged, whether the information transmission behavior is malicious is determined, and if the information transmission behavior is malicious, whether the user of the target program in the second device is a non-biological user is further judged. The method is beneficial to more accurately executing the trigger condition, effectively reduces the computing resource, is beneficial to guaranteeing the computing performance and the system operation performance of the first equipment, and further guarantees the user experience of the user at the first equipment side.

Preferably, the information to be identified includes information corresponding to the hidden information, and the information to be identified is determined by the following steps:

The invention also provides an embodiment of a determining device of the non-biological user, which is applied to first equipment, wherein a target program for information transmission is installed in the first equipment; the device is shown in fig. 7, and comprises:

An information obtaining module 71, configured to obtain information to be received sent by the first device to the second device in response to the first device performing information transmission with the second device through the target program; and the second equipment is internally provided with the target program so as to realize information transmission with the first equipment.

A target information generating module 72, configured to add hidden information to the information to be received, so as to obtain target information; the hidden information is information which can trigger the artificial intelligent program to perform interactive response and cannot be recognized by human beings; and the target information is used for replacing the information to be received and sending the information to the second equipment.

An information sending module 73, configured to send the target information to the second device.

A user determining module 74, configured to obtain information to be identified for the target information returned by the second device after receiving the target information; and if the information to be identified contains information corresponding to the hidden information, judging that the user of the target program of the second equipment is a non-biological user.

The embodiment of fig. 7 obtains the information to be received when the first device confirms that the information is sent to the second device, and adds the hidden information to the information to be received, which can implement adding the hidden information without the sense of the user, and minimize the influence on the user experience of the first device. The target information obtained after the hidden information is added is used for replacing the information to be received and sending the information to the second device, and because the hidden information is the information which can trigger the interaction response of the artificial intelligent program and cannot be identified by human beings, if the user of the target program of the second device is the human beings, the hidden information in the target information cannot be judged and the corresponding reply is made, and the corresponding reply can be made only for the information to be received in the target information; if the user of the target program of the second device is a non-biological user of the artificial intelligence type, namely an artificial intelligence application, an artificial intelligence program and the like, the hidden information can be accurately identified and corresponding replies can be made; therefore, the information to be identified for the target information returned by the second device after receiving the target information is obtained, and if the information to be identified contains information corresponding to the hidden information, the user of the target program of the second device is judged to be a non-biological user. Since the non-biological user is a human-controlled application or program, which itself has no intelligent judgment capability, information interaction with the non-biological user is at a security risk. After the user determining module 74 in the embodiment shown in fig. 7 is executed, an early warning prompt module may be added and executed according to the actual application scenario, and is configured to send early warning prompt information to the user at the first device side after determining that the user of the target program in the second device is a non-biological user, for example, after determining that the user of the target program in the second device is a non-biological user, send prompt information such as "monitor that the other party is a non-biological user, please alert about the risk of information transmission, and do not trust the information content sent by the other party" to the user of the target program in the first device.

The application scenario of the embodiment illustrated in fig. 7 is very extensive, and may be used in the scenario of transmitting presence information such as chat, conference, online shopping, etc., and may be applied to various network environments such as the internet, a local area network, an ad hoc network, etc., where the user of the first device may be a person or an enterprise organization. The embodiment illustrated in fig. 7 can accurately detect non-biological users, prompt risks for users interacting with the non-biological users, help individuals or enterprise users to make correct decisions to protect information and asset security, and provide effective guarantees for maintaining network security and social security.

Preferably, after the acquiring the information to be received sent by the first device to the second device, the information acquiring module 71 is further configured to:

Preferably, if the category of the information to be received is text, the process of adding hidden information into the information to be received includes:

At least one of the target locations is determined as an addition location.

Preferably, if the category of the information to be received is an image, the process of adding hidden information to the information to be received includes:

Preferably, if the category of the information to be received is audio, the process of adding hidden information to the information to be received includes:

Preferably, before the acquiring the information to be received sent by the first device to the second device, the information acquiring module 71 is further configured to:

And matching the target word with words in a preset keyword library.

The embodiment shown in fig. 7 is an embodiment of an apparatus corresponding to the embodiment of the method shown in fig. 1 to 6, and a part of implementation procedures and technical effects of the embodiment shown in fig. 7 are similar to those of the embodiment shown in fig. 1 to 6, so that the description of the embodiment shown in fig. 7 is simpler, and please refer to the embodiment shown in fig. 1 to 6 for relevant points.

In a preferred solution of the foregoing embodiment, before the information to be received sent by the first device to the second device in step S11 is acquired and in the execution process of the information acquisition module 71, a trigger condition is added, so as to determine the information transmission behavior of the second device, determine whether the information transmission behavior of the second device has a malicious nature, and if so, further determine whether the user of the target program in the second device is a non-biological user. The method is beneficial to more accurately executing the trigger condition, effectively reduces the computing resource, is beneficial to guaranteeing the computing performance and the system operation performance of the first equipment, and further guarantees the user experience of the user at the first equipment side. The additional trigger condition may be implemented by the following scheme in addition to the foregoing related preferred scheme.

Before the information to be received, which is sent by the first device to the second device, is obtained, under a target condition, each piece of information sent by the first device to the second device and the sending time corresponding to each piece of information are obtained, and a first information set A is obtained; acquiring each piece of information sent to the first equipment by the second equipment and the sending time corresponding to each piece of information to obtain a second information set B; the target condition includes at least one of: the quantity of the information sent to the first device by the second device meets a preset quantity threshold, and the first device and the second device transmit the information through respective target programs for meeting a specified duration. The preset quantity threshold is set according to actual application requirements, such as 50 strips, 100 strips and the like; the specified time length is set according to actual application requirements, such as half an hour, two hours and the like.

A=（（A ₁ ，T ₁ ），（A ₂ ，T ₂ ），…，（A _i ，T _i ），…，（A _n ，T _n ) A) is provided; i=1, 2, …, n; where n is the number of information acquired from the first device to the second device, A _i For the obtained i-th information sent by the first device to the second device, T _i Is A _i Is a transmission time of (a); b= ((B) ₁ ，R ₁ ），（B ₂ ，R ₂ ），…，（B _j ，R _j ），…，（B _m ，R _m ) A) is provided; j=1, 2, …, m; where m is the number of acquired information sent by the second device to the first device, B _j For the obtained j-th information sent by the second device to the first device, R _j Is B _j Is transmitted, and the transmission time of the same is set.

Preferably, after obtaining the first information set a and the second information set B, the method further includes:

determining a target time interval set t according to the first information set A and the second information set B; t= (t ₁ ，t ₂ ，…，t _y ，…，t _w ) The method comprises the steps of carrying out a first treatment on the surface of the y=1, 2, …, w; wherein t is _y For the determined y-th target time interval, w is the number of the determined target time intervals;

the determining of the target time interval set t comprises the following steps:

step 1: obtaining a target value x=1;

step 2: if x is less than or equal to n-1, then according to T _x And the second information set B determines an intermediate time interval set ZT; otherwise, ending the current flow; zt= (ZT) ₁ ，ZT ₂ ，…，ZT _j ，…，ZT _m ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein ZT _j ZT for the jth intermediate time interval _j =R _j -T _x ；

Step 3: according to T _x Traversing the intermediate time interval set ZT, if ZT _j > 0, and ZT _j ＜T _x+1 -T _x ZT is then _j Determining the time interval as a target time interval, and entering a step 4; otherwise, directly entering the step 4;

step 4: obtain x=x+1, and go to step 1.

The above preferred solution can determine the response time interval of the second device for each piece of information transmitted by the first device, that is, the interval between the time when each first device transmits information to the second device and the time when the second device returns the corresponding reply information.

The process of determining the target time interval set t in the above preferred embodiment is as follows:

assuming that n=100, the first information set a= ((a) ₁ ，T ₁ ），（A ₂ ，T ₂ ），…，（A _i ，T _i ），…，（A ₁₀₀ ，T ₁₀₀ ) A) is provided; i=1, 2, …,100; second information set b= ((B) ₁ ，R ₁ ），（B ₂ ，R ₂ ），…，（B _j ，R _j ），…，（B _m ，R _m ））。

Obtaining target values x=1, respectively using R ₁ 、R ₂ 、…、R _j 、…、R _m Subtracting T ₁ Obtaining T ₁ Corresponding intermediate time interval set Zt= (ZT) ₁ ，ZT ₂ ，…，ZT _f ，…，ZT _m ) The method comprises the steps of carrying out a first treatment on the surface of the f=1, 2, …, m; wherein ZT _f ZT for the f-th intermediate time interval _f =R _f -T ₁ The method comprises the steps of carrying out a first treatment on the surface of the According to T ₁ Traversing the current set of intermediate time intervals if an intermediate time interval ZT exists _d ，d=1，2，…，m，ZT _d Which is itself greater than 0, i.e. the corresponding second device sends information B to the first device _d Time R of (2) _d Later than T ₁ And ZT _d ＜T ₂ -T ₁ I.e. less than the first device sending information a to the second device ₁ And A ₂ To the ZT _d Is determined as T ₁ Corresponding target time interval, i.e. the first device sends information a to the second device ₁ Time T of (2) ₁ And the second device is directed to A ₁ Sending reply information B to the first device _d Time R of (2) _d Time interval between. It is known from experimental calculation that if T is obtained ₁ Earlier than R ₁ Then T is calculated ₁ The corresponding target time interval is ZT ₁ Otherwise T ₁ The corresponding target time interval is determined according to the actual calculation result.

Taking x=x+1, i.e. x=2, determining T according to the calculation procedure described above ₂ And corresponding target time intervals, and so on until the corresponding calculation of x=99 is completed, determining each target time interval to determine the target time interval set t. Since it cannot be excluded that the second device replies one message after the first device sends multiple messages, or that the second device replies multiple messages after the first device sends one message, the number w of the target time intervals finally determined in this example is less than or equal to 99.

Preferably, the ratio of the number of target time intervals smaller than the preset time interval threshold to the total number w of target time intervals in the target time interval set t is determined. The preset time interval threshold is set according to actual application requirements, and based on the characteristic that interaction response of the artificial intelligence application or program is rapid and each target time interval is short, the preset time interval threshold is set to be a smaller value, such as 3 seconds, 2 seconds and the like.

And if the proportion is larger than a preset proportion threshold value, judging that the triggering condition is met. The preset proportion threshold is set according to actual application requirements, and in order to accurately identify interaction behaviors of the artificial intelligence, the preset proportion threshold is set to be a larger value, such as 90%, 95% and the like.

According to the preferred scheme, based on the characteristic that the interaction response of the artificial intelligence is quick, the target time interval corresponding to each interaction response is short, whether the second equipment side user has the characteristics of the interaction behavior of the artificial intelligence or not is primarily identified, when the information interaction time interval is smaller than the preset time interval threshold and the information quantity occupying ratio is too large (larger than the preset proportion threshold), the condition that the preset triggering condition is met is judged, whether the second equipment side user is a non-biological user or not is further judged through the target information, and the judging accuracy is improved.

Preferably, the fluctuation value P of the target time interval is determined from said set t of target time intervals.

P=((Σ ^w _y=1 (t _y -t’) ² )/w) ^1/2 The method comprises the steps of carrying out a first treatment on the surface of the Wherein t' = (Σ ^w _y=1 t _y )/w。

And if the fluctuation value P is smaller than a preset first fluctuation threshold value, judging that the triggering condition is met. The preset first fluctuation threshold is set according to actual application requirements, and because the artificial intelligence type user has the characteristics of quick interaction response and short target time interval corresponding to each interaction response, if the second equipment side user is an artificial intelligence type non-biological user, the corresponding fluctuation value P should be a value tending to 0, so that the preset first fluctuation threshold should be set to be a smaller value, such as 0.5, 0.3 and the like.

According to the preferred scheme, based on the characteristic that interaction response of artificial intelligence is rapid, the target time interval corresponding to each interaction response is short, whether the second equipment side user has the characteristics of artificial intelligence interaction behavior or not is primarily identified by combining the fluctuation value P of the target time interval, if yes, the preset triggering condition is judged to be met, and whether the second equipment side user is a non-biological user or not is further judged through the target information. And setting a judgment condition (whether the fluctuation value P is smaller than a preset first fluctuation threshold value) according to the characteristics of the artificial intelligence application or program, so that the judgment accuracy is improved.

Preferably, clustering is performed on the target time intervals contained in the target time interval set t, so as to obtain a plurality of target time interval groups.

Determining the number of target time intervals contained in each target time interval group to obtain a time interval number set Q; q= (S ₁ ，S ₂ ，…，S _a ，…S _v ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein a=1, 2, …, v, v is the number of the target time interval groups, S _a Is the number of target time intervals contained in the a-th target time interval group.

And determining a fluctuation value Z of the time interval number according to the time interval number set Q.

Z=((Σ ^v _a=1 (S _a -S’) ² )/v) ^1/2 The method comprises the steps of carrying out a first treatment on the surface of the Wherein S' = (Σ ^v _a=1 S _a )/v。

And if the fluctuation value Z is smaller than a preset second fluctuation threshold value, judging that the triggering condition is met. The preset second fluctuation threshold should be set to a smaller value that tends to 0, such as 0.3, 0.2, etc.

The artificial intelligent non-biological user has the characteristics of quick interaction response and short target time interval corresponding to each interaction response, the characteristic can be actively avoided, and the behavior characteristics of stable target time intervals are changed through a rule of increasing or decreasing the interaction response time length, so that the artificial intelligent non-biological user has the characteristics of humanization on the interaction response time length, and the difficulty of judging whether the user at the second equipment side is the artificial intelligent non-biological user is greatly increased. The above-mentioned preferred scheme can effectively solve this problem, effectively carries out preliminary discernment to whether second equipment side user possesses artificial intelligence interaction behavior characteristic. The conclusion is obtained through experimental analysis, no matter what method is used for avoiding the behavior characteristics of stable target time intervals, the behavior characteristics are regularly and sought, or the interactive response time is regularly increased or decreased, for example, the interactive response time is increased to 20 seconds after being increased to 2 seconds each time, or the interactive response time is randomly increased or decreased, for example, the target time is arbitrarily increased each time, and the target time comprises 2 seconds, 4 seconds, 10 seconds and the like. Under the target condition, a certain amount of information sent by the second device to the first device is accumulated, a certain amount of target time intervals are also formed in the target time interval set t, no matter what avoidance mode is adopted, after the target time intervals contained in the target time interval set t are clustered, the obtained amount of time intervals contained in each target time interval group is basically consistent, for example, the number of target time intervals is about 10, the number of target time intervals is about 11, and the number of target time intervals is about 15 seconds. Therefore, by adopting the preferable scheme, under the condition that the behavior characteristics of the second equipment side, which are stable in target time interval, of the artificial intelligence type user can be avoided, whether the second equipment side user has the artificial intelligence interaction behavior characteristics or not can be effectively identified preliminarily, whether the second equipment side user is a non-biological user or not can be further judged through the target information, and the accuracy of a judging result is effectively improved.

Furthermore, although the steps of the methods in the present disclosure are depicted in a particular order in the drawings, this does not require or imply that the steps must be performed in that particular order or that all illustrated steps be performed in order to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform, etc.

From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (may be a personal computer, a server, a mobile terminal, or a network device, etc.) to perform the method according to the embodiments of the present disclosure.

In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

Those skilled in the art will appreciate that the various aspects of the application may be implemented as a system, method, or program product. Accordingly, aspects of the application may be embodied in the following forms, namely: an entirely hardware embodiment, an entirely software embodiment (including firmware, micro-code, etc.) or an embodiment combining hardware and software aspects may be referred to herein as a "circuit," module "or" system.

An electronic device according to this embodiment of the application. The electronic device is merely an example, and should not impose any limitations on the functionality and scope of use of embodiments of the present application.

The electronic device is in the form of a general purpose computing device. Components of an electronic device may include, but are not limited to: the at least one processor, the at least one memory, and a bus connecting the various system components, including the memory and the processor.

Wherein the memory stores program code that is executable by the processor to cause the processor to perform steps according to various exemplary embodiments of the application described in the "exemplary methods" section of this specification.

The storage may include readable media in the form of volatile storage, such as Random Access Memory (RAM) and/or cache memory, and may further include Read Only Memory (ROM).

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

The bus may be one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.

The electronic device may also communicate with one or more external devices (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device, and/or with any device (e.g., router, modem, etc.) that enables the electronic device to communicate with one or more other computing devices. Such communication may be through an input/output (I/O) interface. And, the electronic device may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet, through a network adapter. The network adapter communicates with other modules of the electronic device via a bus. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with an electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.

From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (may be a personal computer, a server, a terminal device, or a network device, etc.) to perform the method according to the embodiments of the present disclosure.

In an exemplary embodiment of the present disclosure, a computer-readable storage medium having stored thereon a program product capable of implementing the method described above in the present specification is also provided. In some possible embodiments, the various aspects of the application may also be implemented in the form of a program product comprising program code for causing a terminal device to carry out the steps according to the various exemplary embodiments of the application as described in the "exemplary methods" section of this specification, when said program product is run on the terminal device.

The program product may employ 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 can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The computer readable signal medium may include a data signal propagated in baseband or as part of a carrier wave with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind 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., connected via the Internet using an Internet service provider).

Furthermore, the above-described drawings are only schematic illustrations of processes included in the method according to the exemplary embodiment of the present application, and are not intended to be limiting. It will be readily appreciated that the processes shown in the above figures do not indicate or limit the temporal order of these processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, for example, among a plurality of modules.

It should be noted that although in the above detailed description several modules or units of a device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit in accordance with embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.

The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present application should be included in the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.

Claims

1. A method for determining an abiotic user, which is applied to a first device in which a target program for information transmission is installed; the method comprises the following steps:

responding to the information transmission between the first equipment and the second equipment through the target program, and acquiring information to be received, which is sent to the second equipment by the first equipment; the second equipment is internally provided with the target program so as to realize information transmission with the first equipment;

Adding hidden information into the information to be received to obtain target information; the hidden information is information which can trigger the artificial intelligent program to perform interactive response and cannot be recognized by human beings; the target information is used for replacing the information to be received and sending the information to the second equipment;

transmitting the target information to a second device;

acquiring information to be identified aiming at the target information returned by the second equipment after receiving the target information;

2. The method of claim 1, wherein after the obtaining the information to be received sent by the first device to the second device, the method further comprises:

3. The method according to claim 2, wherein if the category of the information to be received is text, the process of adding hidden information to the information to be received includes:

Carrying out semantic analysis on the information to be received to obtain semantic features corresponding to the information to be received;

determining semantic features corresponding to each piece of text information in a preset text library; the preset text library comprises a plurality of pieces of text information, and each piece of text information can trigger an artificial intelligent program to perform interactive response;

calculating the similarity of the semantic features corresponding to the information to be received and the semantic features corresponding to each piece of text information, and determining the text information corresponding to the similarity calculation result smaller than a preset threshold value as target text information;

determining a piece of text information from a plurality of pieces of target text information as text information to be added into the information to be received;

determining an interface format of an information transmission interface of the target program;

converting the format of the text information to be added into the information to be received according to the interface format to obtain hidden information;

4. The method according to claim 2, wherein if the category of the information to be received is text, the process of adding hidden information to the information to be received includes:

word segmentation is carried out on the information to be received, and a word sequence corresponding to the information to be received is obtained;

carrying out grammar analysis on the information to be received based on the word sequence to obtain a grammar information sequence corresponding to the word sequence; the grammar information sequence comprises a plurality of grammar information, the grammar information corresponds to words in the word sequence one by one, and each grammar information comprises: word parts and labels of words which can be linked before and after the word corresponding to the grammar information; the tag is used for marking features of words, and the features comprise: characters, animals, food, tools, books, landscapes;

determining the part of speech and the label of the addable word corresponding to each target position according to the word sequence and the grammar information sequence to obtain the addition requirement when each target position carries out word addition; the target position comprises the position before the first word, the position after the last word and the position between every two words in the word sequence;

determining at least one of the target locations as an addition location;

Screening words meeting the addition requirements when adding words at the addition positions from a preset word stock, and adding the screened words to the corresponding addition positions to obtain a plurality of target word sequences; the preset word stock comprises a plurality of words which are used for being added into the information to be received as hidden information;

the words in each target word sequence are joined according to the sequence order, and target texts corresponding to each target word sequence are obtained;

carrying out semantic analysis on each target text to obtain semantic features corresponding to each target text;

calculating the similarity between the semantic features corresponding to the information to be received and the semantic features corresponding to each target text, and determining the target text corresponding to the minimum similarity calculation result as the text to be sent;

determining information display attributes of the target program; the information display attributes comprise background color, shading color, brightness and transparency of a display window;

5. The method of claim 4, wherein determining the part of speech and the tag of the additivable word corresponding to each target location based on the word sequence and the grammar information sequence comprises:

determining first information and second information corresponding to each grammar information according to the grammar information sequence; the first information is the part of speech and the label of the word which can be linked before the word, and the second information is the part of speech and the label of the word which can be linked after the word;

determining the part of speech and the label of the word which can be added before the first word in the word sequence according to the first information corresponding to the first grammar information in the grammar information sequence;

respectively determining intersection of second information corresponding to each grammar information in the grammar information sequence and first information corresponding to the next grammar information to obtain part of speech and labels of words which can be added between every two words in the word sequence;

6. The method of claim 5, wherein adding the screened word to the corresponding addition location comprises:

Determining a plurality of screened words corresponding to each adding position to obtain a word set corresponding to each adding position;

selecting a word from each word set to enumerate and combine to obtain a plurality of combined sequences;

7. The method according to claim 2, wherein if the category of the information to be received is an image, the process of adding hidden information to the information to be received includes:

identifying the image characteristics of the information to be received to obtain an image characteristic vector corresponding to the information to be received;

determining an image feature vector of each image data in a preset image library; the preset image library comprises a plurality of image data, and each image data can trigger an artificial intelligent program to perform interactive response; the image data is used for being added into the information to be received as hidden information;

calculating the similarity between the image feature vector corresponding to the information to be received and the image feature vector of each image data, and determining the image data corresponding to the minimum similarity calculation result as a watermark image;

Determining the image attribute of the information to be received; the image attributes include image size and display attributes including: resolution, pixels;

adjusting the image attribute of the watermark image according to the image attribute of the information to be received so that the image size of the watermark image is not larger than the image size of the information to be received, and the display attribute is the same as the display attribute of the information to be received;

according to a first preset threshold, the transparency of the watermark image is adjusted so that the watermark image cannot be recognized by human eyes;

8. The method according to claim 2, wherein if the category of the information to be received is audio, the process of adding hidden information to the information to be received includes:

identifying the information to be received, and determining voiceprint characteristics and semantic information of the information to be received;

determining semantic features corresponding to the information to be received according to the semantic information of the information to be received;

determining semantic features corresponding to each piece of audio data in a preset audio library; the audio library comprises a plurality of audio data, and each audio data can trigger an artificial intelligent program to perform interactive response; the audio data is used as hidden information to be added into the information to be received;

Calculating the similarity of the semantic features corresponding to the information to be received and the semantic features corresponding to each piece of audio data, and determining the audio data corresponding to the minimum similarity calculation result as target audio;

converting the target audio according to the voiceprint characteristics of the information to be received so that the voiceprint characteristics of the target audio are identical to those of the information to be received;

adjusting the frequency of the target audio to a target frequency band; the target frequency band is a frequency band which cannot be perceived by human ears;

adding the target audio to a target position of the information to be received so as to obtain the target information; the target position comprises positions without effective signals before and after the information to be received starts;

performing Fourier transform on the information to be received to obtain a frequency domain signal corresponding to the information to be received;

9. The method according to any of claims 1-8, wherein prior to said obtaining information to be received that the first device sends to the second device, the method further comprises:

Acquiring information sent by a specified number of second devices to the first device;

performing word segmentation statistics on the acquired information, and determining target words with word frequency greater than a third preset threshold value;

matching the target word with words in a preset keyword library;

10. The method according to claim 9, wherein the information to be identified includes information corresponding to the hidden information, and the method is determined by:

determining keywords of reply information generated by the artificial intelligence program after triggering interactive response of the artificial intelligence program according to the hidden information; the key words of the reply information are information corresponding to the hidden information;

11. A determination apparatus of a non-biological user, characterized by being applied to a first device in which a target program for information transmission is installed; the device comprises:

the information acquisition module is used for responding to the information transmission between the first equipment and the second equipment through the target program and acquiring information to be received, which is sent to the second equipment by the first equipment; the second equipment is internally provided with the target program so as to realize information transmission with the first equipment;

The target information generation module is used for adding hidden information into the information to be received to obtain target information; the hidden information is information which can trigger the artificial intelligent program to perform interactive response and cannot be recognized by human beings; the target information is used for replacing the information to be received and sending the information to the second equipment;

the information sending module is used for sending the target information to the second equipment;

12. A non-transitory computer readable storage medium having stored therein at least one instruction or at least one program, wherein the at least one instruction or the at least one program is loaded and executed by a processor to implement the method of any one of claims 1-10.