Information processing method, device, equipment and storage medium

Abstract

The present disclosure provides an information processing method, device, equipment and storage medium, which relates to the field of information security technology. The method includes: obtaining multiple call information of a target phone number; classifying and processing the target phone number according to the multiple call information to obtain target category information of the target phone number; performing hierarchical analysis according to the multiple call information and the target category information of the target phone number to obtain the abnormal index weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal index weight. The method realizes the effective identification of phone numbers that perform abnormal behavior.

Description

Technical Field

[0001] The present disclosure relates to the field of information security technology, and in particular to an information processing method, apparatus, device and readable storage medium. Background Art

[0002] With the development of information technology, there are more and more cases of abnormal behaviors using communications, such as fraud through phone calls, the Internet, and text messages. How to effectively identify phone numbers that perform abnormal behaviors has become an urgent problem to be solved.

[0003] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0004] The purpose of the present disclosure is to provide an information processing method, apparatus, device and readable storage medium, which can effectively identify telephone numbers that perform abnormal behaviors.

[0005] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0006] According to one aspect of the present disclosure, there is provided an information processing method, comprising: obtaining a plurality of call information of a target phone number; classifying the target phone number according to the plurality of call information to obtain target category information of the target phone number; performing hierarchical analysis according to the plurality of call information and the target category information of the target phone number to obtain an abnormality indicator weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormality indicator weight.

[0007] According to an embodiment of the present disclosure, the abnormal indicator weight includes multiple weights of multiple abnormal indicator levels relative to the target telephone number; hierarchical analysis is performed according to the multiple call information and the target category of the target telephone number to obtain the abnormal indicator weight of the target telephone number, including: obtaining a comparison matrix of the multiple call information according to the target category information of the target telephone number, the comparison matrix of the multiple call information is used to indicate the importance of each two call information in the multiple call information relative to the target telephone number; obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information, the comparison matrix of the multiple abnormal indicator levels is used to indicate the importance of each two abnormal indicator levels in the multiple abnormal indicator levels relative to various call information; obtaining multiple weights of the multiple abnormal indicator levels relative to the target telephone number based on the comparison matrix of the multiple call information and the comparison matrix of the multiple abnormal indicator levels.

[0008] According to an embodiment of the present disclosure, the multiple call information includes outgoing number information, call duration information, call time information and traffic usage information; the target category information of the target phone number includes an outgoing number abnormal category weight, a call duration abnormal category weight, a call time abnormal category weight and a traffic usage abnormal category weight; the target phone number is classified and processed according to the multiple call information to obtain the target category information of the target phone number, including: obtaining a call feature vector of the target phone number according to the outgoing number information, the call duration information, the call time information and the traffic usage information; inputting the call feature vector of the target phone number into a convolutional neural network to obtain the outgoing number abnormal category weight, the call duration abnormal category weight, the call time abnormal category weight and the traffic usage abnormal category weight of the target phone number; obtaining a comparison matrix of the multiple call information according to the target category information of the target phone number, including: comparing the abnormal category weights of each two call information in the multiple call information according to the outgoing number abnormal category weight, the call duration abnormal category weight, the call time abnormal category weight and the traffic usage abnormal category weight of the target phone number to obtain the comparison matrix of the multiple call information.

[0009] According to an embodiment of the present disclosure, the method further includes: performing a consistency check on the comparison matrix of the multiple call information according to the target category information of the target telephone number; if the comparison matrix of the multiple call information passes the consistency check, obtaining a normalized feature vector of the comparison matrix of the multiple call information; performing a consistency check on the comparison matrix of the multiple abnormal indicator levels according to the multiple call information; if the comparison matrix of the multiple abnormal indicator levels passes the consistency check, obtaining a normalized feature vector of the comparison matrix of the multiple abnormal indicator levels; based on the comparison matrix of the multiple call information and the comparison matrix of the multiple abnormal indicator levels, obtaining multiple weights of the multiple abnormal indicator levels respectively relative to the target telephone number, including: obtaining multiple weights of the multiple abnormal indicator levels respectively relative to the target telephone number according to the normalized feature vector of the comparison matrix of the multiple call information and the normalized feature vector of the comparison matrix of the multiple abnormal indicator levels.

[0010] According to an embodiment of the present disclosure, the multiple call information includes call voice information; the method also includes: obtaining preset sensitive words; obtaining a comparison matrix of the multiple call information according to the target category information of the target phone number, including: obtaining a comparison matrix of the multiple call information according to the target category information of the target phone number and the preset sensitive words; obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information, including: obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information and the preset sensitive words.

[0011] According to one embodiment of the present disclosure, the call voice information includes a call text obtained by performing natural language processing on the call voice; a comparison matrix of the multiple call information is obtained according to the target category information of the target phone number and the preset sensitive words, including: matching the call text according to the preset sensitive words to obtain sensitivity information of the call text; obtaining a comparison matrix of the multiple call information according to the target category information of the target phone number and the sensitivity information of the call text; obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information and the preset sensitive words, including: obtaining a comparison matrix of the multiple abnormal indicator levels according to the sensitivity information of the multiple call information and the call text.

[0012] According to an embodiment of the present disclosure, the multiple abnormal indicator levels include a blacklist level, a graylist level and a whitelist level. The abnormality degree of the blacklist level is higher than that of the graylist level, and the whitelist level is a non-abnormal level.

[0013] According to another aspect of the present disclosure, there is provided an information processing device, comprising: an acquisition module for acquiring multiple call information of a target phone number; a processing module for classifying the target phone number according to the multiple call information to obtain target category information of the target phone number; and an analysis module for performing hierarchical analysis according to the multiple call information and the target category information of the target phone number to obtain an abnormality indicator weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormality indicator weight.

[0014] According to an embodiment of the present disclosure, the abnormal indicator weight includes multiple weights of multiple abnormal indicator levels relative to the target telephone number; the analysis module is also used to: obtain a comparison matrix of the multiple call information according to the target category information of the target telephone number, the comparison matrix of the multiple call information is used to indicate the importance of each two call information in the multiple call information relative to the target telephone number; obtain a comparison matrix of the multiple abnormal indicator levels according to the multiple call information, the comparison matrix of the multiple abnormal indicator levels is used to indicate the importance of each two abnormal indicator levels in the multiple abnormal indicator levels relative to various call information; obtain multiple weights of the multiple abnormal indicator levels relative to the target telephone number based on the comparison matrix of the multiple call information and the comparison matrix of the multiple abnormal indicator levels.

[0015] According to an embodiment of the present disclosure, the multiple call information includes outgoing number information, call duration information, call time information and traffic usage information; the target category information of the target phone number includes an outgoing number abnormal category weight, a call duration abnormal category weight, a call time abnormal category weight and a traffic usage abnormal category weight; the processing module is further used to: obtain a call feature vector of the target phone number according to the outgoing number information, the call duration information, the call time information and the traffic usage information; input the call feature vector of the target phone number into a convolutional neural network to obtain the outgoing number abnormal category weight, the call duration abnormal category weight, the call time abnormal category weight and the traffic usage abnormal category weight of the target phone number; the analysis module is further used to compare the abnormal category weights of each two call information in the multiple call information according to the outgoing number abnormal category weight, the call duration abnormal category weight, the call time abnormal category weight and the traffic usage abnormal category weight of the target phone number to obtain a comparison matrix of the multiple call information.

[0016] According to one embodiment of the present disclosure, the analysis module is further used to: perform a consistency check on the comparison matrices of the multiple call information according to the target category information of the target telephone number; if the comparison matrices of the multiple call information pass the consistency check, obtain the normalized feature vectors of the comparison matrices of the multiple call information; perform a consistency check on the comparison matrices of the multiple abnormal indicator levels according to the multiple call information; if the comparison matrices of the multiple abnormal indicator levels pass the consistency check, obtain the normalized feature vectors of the comparison matrices of the multiple abnormal indicator levels; obtain multiple weights of the multiple abnormal indicator levels relative to the target telephone number according to the normalized feature vectors of the comparison matrices of the multiple call information and the normalized feature vectors of the comparison matrices of the multiple abnormal indicator levels.

[0017] According to one embodiment of the present disclosure, the multiple call information includes call voice information; the acquisition module is also used to obtain preset sensitive words; the analysis module is also used to: obtain a comparison matrix of the multiple call information based on the target category information of the target phone number and the preset sensitive words; obtain a comparison matrix of the multiple abnormal indicator levels based on the multiple call information and the preset sensitive words.

[0018] According to one embodiment of the present disclosure, the call voice information includes call text obtained by performing natural language processing on the call voice; the analysis module is also used to: match the call text according to the preset sensitive words to obtain sensitivity information of the call text; obtain a comparison matrix of the multiple call information according to the target category information of the target phone number and the sensitivity information of the call text; obtain a comparison matrix of the multiple abnormal indicator levels according to the multiple call information and the sensitivity information of the call text.

[0019] According to an embodiment of the present disclosure, the multiple abnormal indicator levels include a blacklist level, a graylist level and a whitelist level. The abnormality degree of the blacklist level is higher than that of the graylist level, and the whitelist level is a non-abnormal level.

[0020] According to another aspect of the present disclosure, a device is provided, comprising: a memory, a processor, and executable instructions stored in the memory and executable in the processor, wherein the processor implements any of the above methods when executing the executable instructions.

[0021] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, and when the executable instructions are executed by a processor, any of the above methods is implemented.

[0022] The information processing method provided by the embodiment of the present disclosure obtains multiple call information of the target phone number, classifies the target phone number according to the multiple call information, obtains the target category information of the target phone number, and then performs hierarchical analysis based on the multiple call information and the target category information of the target phone number to obtain the abnormal indicator weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal indicator weight. By performing hierarchical analysis on the multiple call information and the target category information of the target phone number, the abnormal indicator weight that characterizes the abnormal risk degree of the target phone number is obtained, thereby effectively identifying the phone number that performs abnormal behavior.

[0023] It is to be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and other objects, features and advantages of the present disclosure will become more apparent by describing in detail example embodiments thereof with reference to the attached drawings.

[0025] Figure 1 A schematic diagram showing a system structure in an embodiment of the present disclosure.

[0026] Figure 2 A flow chart of an information processing method in an embodiment of the present disclosure is shown.

[0027] Figure 3 Shows Figure 2 FIG. 4 is a schematic diagram of the processing process of step S204 and step S206 in one embodiment.

[0028] Figure 4 A flowchart of another method in an embodiment of the present disclosure is shown.

[0029] Figure 5 Shows Figure 2 FIG. 5 is a schematic diagram of the processing process of step S310 in one embodiment.

[0030] Figure 6 Shows Figure 2 FIG. 5 is a schematic diagram of the processing process of step S206 in another embodiment.

[0031] Figure 7 is based on Figures 2 to 6 A schematic diagram of an abnormal telephone number identification process is shown.

[0032] Figure 8 A block diagram of an information processing device in an embodiment of the present disclosure is shown.

[0033] Fig. 9 A schematic structural diagram of an electronic device in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0034] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The accompanying drawings are only schematic illustrations of the disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0035] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, devices, steps, etc. may be adopted. In other cases, well-known structures, methods, devices, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0036] In addition, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present disclosure, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. The symbol " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0037] In this disclosure, unless otherwise clearly specified and limited, terms such as "connection" should be understood in a broad sense, for example, it can be an electrical connection or can communicate with each other; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in this disclosure can be understood according to specific circumstances.

[0038] With the development of information technology, there are more and more telecom frauds, such as fabricating false information and setting up scams through telephone, Internet and text messages, usually by impersonating others and counterfeiting various legal cloaks and forms, such as impersonating public security, judicial organs, merchants, companies, manufacturers, state agency staff, bank staff and other institutions, counterfeiting and impersonating recruitment, marriage, loans, winning, mobile phone positioning and other forms, to carry out remote, contactless fraud on victims, inducing victims to make payments or transfers, and achieving the purpose of deception. Preventing telecom fraud is very necessary to protect people's lives and property. Before telecom fraud occurs, conducting fraud risk analysis on phone numbers is an effective way to prevent telecom fraud.

[0039] Therefore, the present disclosure provides an information processing method, which, after obtaining multiple call information of a target phone number, classifies the target phone number according to the multiple call information to obtain target category information of the target phone number, and then performs hierarchical analysis based on the multiple call information and the target category information of the target phone number to obtain the abnormal indicator weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal indicator weight. By performing hierarchical analysis on the multiple call information and the target category information of the target phone number, the abnormal indicator weight that characterizes the abnormal risk degree of the target phone number is obtained, thereby effectively identifying the phone number that performs abnormal behavior.

[0040] Figure 1 An exemplary system architecture 10 to which the information processing method or information processing apparatus of the present disclosure can be applied is shown.

[0041] like Figure 1 As shown, the system architecture 10 may include a terminal device 102, a network 104, and a server 106. The terminal device 102 may be a variety of electronic devices having a display screen and supporting input and output, including but not limited to a smart phone, a tablet computer, a laptop portable computer, a desktop computer, a wearable device, a virtual reality device, a smart home, etc. The network 104 is used to provide a medium for a communication link between the terminal device 102 and the server 106. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc. The server 106 may be a server or server cluster that provides various services, such as a background processing server, a database server, etc.

[0042] The user can use the terminal device 102 to interact with the server 106 through the network 104 to receive or send data, etc. For example, the user downloads various call information of the target phone number from the server 106 to the terminal device 102 through the network 104, and then classifies and processes it through the information processing software on the terminal device 102. For another example, the user operates on the terminal device 102 and calls the server 106 through the network 104 to process the various call information of the target phone number. For another example, the server 106 can perform hierarchical analysis based on the target category information of the various call information and the target phone number, and then send the obtained abnormal index weight of the target phone number to the terminal device 102 through the network 104, and the user can view the abnormal index weight of the target phone number on the terminal device 102.

[0043] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0044] Figure 2FIG. 1 is a flowchart of an information processing method according to an exemplary embodiment. Figure 2 The method shown can be applied to the server side of the above system, for example, and can also be applied to the terminal device of the above system.

[0045] refer to Figure 2 , the method 20 provided in the embodiment of the present disclosure may include the following steps.

[0046] In step S202, various call information of the target phone number is obtained.

[0047] In some embodiments, for example, the target phone number may be the phone numbers of all online users available to the operator, that is, all phone numbers that have generated calls, regardless of whether they have passed user real-name authentication, and their related call information can be recorded by the network equipment of the communication network.

[0048] In some embodiments, for example, various call information may include outgoing number information, call duration information, call time information, and traffic usage information. Among them, the outgoing number information may include an outgoing number list, whether the outgoing number is a registered telecommunications fraud number, the number of outgoing numbers, the region to which the outgoing number belongs, etc.; the call duration information may include the call duration, whether it is less than a preset duration threshold (for example, 15 seconds, 30 seconds, 1 minute, etc.); the call time information may include the call time point, the outgoing call frequency, etc., for example, whether the call time point is a time point in a fixed period of time every day, whether the frequency of calling the registered telecommunications fraud number is higher than the preset frequency threshold (for example, 1 time / week, 2 times / week, 3 times / week, etc.); the traffic usage information may include the time of continuous use of the traffic network.

[0049] In some embodiments, for example, the various call information may include call voice information, which may include a recording of an answering machine call, and may also include a call text obtained by performing natural language processing (Natural Language Processing, NLP) on the call voice.

[0050] In some embodiments, for example, various call information may include a travel track obtained based on the call connection base station, as well as SMS sending frequency and time, etc.

[0051] In some embodiments, for example, after obtaining the initial information of all user telephone numbers collected by the network device, preliminary classification can be performed according to outgoing number information, call duration information, call time information, traffic usage information, and call voice information, etc. For example, regular matching and other methods can be used to obtain corresponding types of information.

[0052] In step S204, the target telephone number is classified according to the various call information to obtain target category information of the target telephone number.

[0053] In some embodiments, the target category information of the target phone number may include an abnormal category weight for an outgoing number, an abnormal category weight for a call duration, an abnormal category weight for a call time, and an abnormal category weight for traffic usage. According to the specific contents of the outgoing number information, call duration information, call time information, and traffic usage information in the various call information, a convolutional neural network may be used for classification to obtain the weights of each abnormal category, and then the abnormal category weight for the outgoing number, call duration, call time, and traffic usage of the target phone number may be applied to the hierarchical analysis of step S206. For a specific implementation method, please refer to Figure 3 .

[0054] In some embodiments, the target category information may include a sensitive word exception category. For example, if multiple call information includes call voice information, after obtaining the call voice text through NLP processing, the call voice text can be matched according to pre-set sensitive words, and then the target phone number can be judged according to the set sensitive word quantity threshold whether it belongs to the sensitive word exception category.

[0055] In step S206, hierarchical analysis is performed based on the target category information of the target phone number and the multiple call information to obtain the abnormal index weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal index weight.

[0056] In some embodiments, a hierarchical analysis method can be used to evaluate the collected call information of the target phone number and classify the target phone number into different levels. A phone number hierarchical analysis model can be created based on the characteristics of harassment and fraudulent calls. In the phone number analysis model, the risk level of the phone number is determined by hierarchical analysis based on different weights for information such as the outgoing number, call duration, scheduled calls, and traffic usage.

[0057] The analytic hierarchy process is to decompose the problem into different components according to the nature of the problem and the overall goal to be achieved, and to cluster and combine the factors at different levels according to the mutual correlation and influence between the factors and the subordinate relationship, so as to form a multi-level analysis structure model, so as to ultimately reduce the problem to the determination of the relative importance weights of the lowest level (schemes and measures for decision-making, etc.) relative to the highest level (overall goal) or the ranking of the relative advantages and disadvantages. For example, the target phone number can be the highest level, the various call information of the target phone number can be the various factors of the middle level, and the abnormal index level of the target phone number can be the lowest level. The target category information of the target phone number can be used to obtain the relative importance weights of the middle level relative to the highest level, and the relative importance weights of the lowest level relative to the middle level. Finally, the relative importance weights of the lowest level relative to the highest level are obtained, which are the abnormal index weights including the multiple weights of multiple abnormal index levels relative to the target phone number. The specific implementation method of the telephone number analysis model using the analytic hierarchy process can be referred to. Figures 3 to 6 .

[0058] In some embodiments, the multiple abnormality index levels may include a blacklist level, a graylist level, and a whitelist level. The abnormality of the blacklist level is higher than that of the graylist level, and the whitelist level is a non-abnormal level. For example, if in the hierarchical analysis result of the target phone number, the abnormality index weight of the blacklist level is higher than the abnormality index weight of the graylist level and the abnormality index weight of the whitelist level, it is determined that the target phone number is classified into the blacklist, and corresponding reminder processing is performed. The specific implementation method of processing phone numbers of different abnormal levels can be referred to Figure 7 .

[0059] According to the information processing method provided by the embodiment of the present disclosure, after obtaining multiple call information of the target phone number, the target phone number is classified and processed according to the multiple call information to obtain the target category information of the target phone number, and then a hierarchical analysis is performed based on the multiple call information and the target category information of the target phone number to obtain the abnormal indicator weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal indicator weight. By performing a hierarchical analysis on the multiple call information and the target category information of the target phone number, the abnormal indicator weight that characterizes the abnormal risk degree of the target phone number is obtained, thereby effectively identifying the phone number that performs abnormal behavior.

[0060] According to the information processing method provided by the embodiment of the present disclosure, all mobile phone numbers can be comprehensively analyzed based on big data, and their call records can be subjected to convolutional neural classification and hierarchical analysis. According to the analysis results, these users can be divided into black, white and gray lists, and different identifiers can be added to the black, white and gray lists to remind users who receive the calls, so as to prevent the occurrence of telecommunications fraud and protect the safety of life and property.

[0061] Figure 3 Shows Figure 2 Schematic diagram of the processing process of step S204 and step S206 in one embodiment. Figure 3 As shown, in the embodiment of the present disclosure, the above-mentioned step S204 and step S206 may further include the following steps.

[0062] Step S302, obtaining a call feature vector of the target phone number according to the outgoing number information, call duration information, call time information and traffic usage information.

[0063] In some embodiments, feature vectors of outgoing number information, call duration information, call time information and traffic usage information may be extracted according to set rules. For example, the feature vector dimensions of outgoing number information may include the number of outgoing numbers that are registered telecommunications fraud numbers, the total number of outgoing numbers, the number of outgoing numbers per day, etc. The feature vector dimensions of call duration information may include a call duration less than a preset duration threshold being recorded as 1, a call duration not less than a preset duration threshold being recorded as 0, etc. The feature vector dimensions of call time information may include a call time point identifier, a frequency of calling a registered telecommunications fraud number higher than a multiple of a preset frequency threshold, etc. The feature vector dimensions of traffic usage information may include a continuous traffic usage duration identifier, etc.

[0064] Step S304, inputting the call feature vector of the target phone number into the convolutional neural network to obtain the abnormal category weight of the outgoing number, abnormal category weight of the call duration, abnormal category weight of the call time and abnormal category weight of the traffic usage of the target phone number.

[0065] In some embodiments, for example, a convolutional neural network (CNN) may be trained using manually labeled suspected fraud phone numbers, wherein the output of the CNN may include weights of multiple abnormal categories, such as weights of abnormal categories for outgoing numbers, weights of abnormal categories for call duration, weights of abnormal categories for call time, weights of abnormal categories for traffic usage, etc. Then, the call feature vector of the target phone number to be predicted is input into the trained CNN to obtain the abnormal category weights of the outgoing number, abnormal category weights for call duration, abnormal category weights for call time, and abnormal category weights for traffic usage of the target phone number.

[0066] Step S306, based on the outgoing number abnormality category weight, call duration abnormality category weight, call time abnormality category weight and traffic usage abnormality category weight of the target phone number, compare the abnormality category weights of each two call information in the multiple call information to obtain a comparison matrix of the multiple call information.

[0067] In some embodiments, a comparison matrix of multiple call information that represents the importance of each two types of call information relative to the target phone number can be obtained based on the outgoing number abnormality category weight, call duration abnormality category weight, call time abnormality category weight, and traffic usage abnormality category weight of the target phone number. The factors in the middle layer may include outgoing number abnormality factors, call duration abnormality factors, call time abnormality factors, and traffic usage abnormality factors. The comparison matrix can be obtained using Santy's 1-9 scaling method. For example, if the outgoing number abnormality category weight of the target phone number is 0.8 and the call duration abnormality category weight is 0.1, then the outgoing number abnormality factor of the target phone number is more important than the call duration abnormality factor. Then, element A in the comparison matrix of multiple call information in the middle layer is ij =7,A ji =1 / 7, i represents the abnormal factor of the outgoing number, and j represents the abnormal factor of the call duration.

[0068] Step S308, obtaining a comparison matrix of multiple abnormal indicator levels according to the multiple call information.

[0069] In some embodiments, a comparison matrix of multiple abnormal indicator levels for representing the importance of each two abnormal indicator levels in multiple abnormal indicator levels relative to various call information can be obtained based on the content of multiple call information of the target phone number. For example, for the middle-level outgoing number abnormality factor, if the number of outgoing numbers that are registered telecommunications fraud numbers in the outgoing number information of the target phone number is 500 and the total number of outgoing numbers is 550, then for the outgoing number abnormality factor of the target phone number, the blacklist level and the graylist level of the lowest level are more important than the outgoing number abnormality factor, that is, based on the number of registered telecommunications fraud numbers and the total number of outgoing numbers, the target phone number is very likely to be at the blacklist level, then the element B in the comparison matrix of multiple call information at the lowest level is pq =8,B qp =1 / 8, p represents the abnormal factor of the outgoing number, and q represents the abnormal factor of the call duration.

[0070] Step S310: obtaining a plurality of weights of the plurality of abnormal index levels relative to the target telephone number based on a comparison matrix of the plurality of call information and a comparison matrix of the plurality of abnormal index levels.

[0071] In some embodiments, after obtaining the comparison matrix of the middle layer and the lowest layer, the normalized feature vectors of the comparison matrix of the middle layer and the lowest layer can be obtained respectively, and then the normalized feature vector is used as the weight vector of the layer, and then according to each abnormal indicator level of the lowest layer (such as blacklist level, graylist level and whitelist level), the weights of each layer are multiplied to obtain multiple weights of multiple abnormal indicator levels relative to the target phone number. The specific implementation method can be referred to Figure 5 .

[0072] Figure 4 Shows Figure 2 The processing process of step S206 in one embodiment is shown in FIG. Figure 4 As shown, in the embodiment of the present disclosure, the above step S206 may further include the following steps.

[0073] Step S402: obtaining a comparison matrix of multiple call information according to the target category information of the target phone number, wherein the comparison matrix of multiple call information is used to indicate the importance of each two call information in the multiple call information relative to the target phone number.

[0074] In some embodiments, the specific implementation of step S402 may refer to step S306 and step S604.

[0075] Step S404: obtaining a comparison matrix of multiple abnormality indicator levels according to the multiple call information. The comparison matrix of the multiple abnormality indicator levels is used to indicate the importance of each two abnormality indicator levels in the multiple abnormality indicator levels relative to the various call information.

[0076] In some embodiments, the specific implementation of step S404 may refer to step S308 and step S606.

[0077] Step S406: obtaining a plurality of weights of the plurality of abnormal index levels relative to the target telephone number based on a comparison matrix of the plurality of call information and a comparison matrix of the plurality of abnormal index levels.

[0078] In some embodiments, the specific implementation of step S406 can refer to Figure 5 .

[0079] Figure 5 Shows Figure 2 The processing diagram of step S310 in one embodiment is shown in FIG. Figure 5 As shown, in the embodiment of the present disclosure, the above step S310 may further include the following steps.

[0080] Step S502: performing consistency check on comparison matrices of multiple call information according to target category information of the target phone number.

[0081] In some embodiments, the size and eigenvalue of the comparison matrix can be used as parameters to define its consistency index. When the consistency index calculated according to the size and eigenvalue of the comparison matrix meets the consistency requirements, it can be considered to have passed the consistency test, otherwise it has failed the consistency test.

[0082] Step S504: if the comparison matrix of the multiple call information passes the consistency check, a normalized feature vector of the comparison matrix of the multiple call information is obtained.

[0083] In some embodiments, if the comparison matrix of the various call information does not pass the consistency check, the values ​​in the comparison matrix may be adjusted until it passes the consistency check.

[0084] Step S506: performing consistency check on comparison matrices of multiple abnormal indicator levels according to multiple call information.

[0085] Step S508: if the comparison matrices of the multiple abnormality index levels pass the consistency check, the normalized eigenvectors of the comparison matrices of the multiple abnormality index levels are obtained.

[0086] Step S510: obtaining a plurality of weights of the plurality of abnormal index levels relative to the target telephone number according to the normalized feature vectors of the comparison matrix of the plurality of call information and the normalized feature vectors of the comparison matrix of the plurality of abnormal index levels.

[0087] In some embodiments, the normalized feature vector of each layer is used as the weight vector of the layer, and then the weights of each layer are multiplied according to each abnormal indicator level of the lowest layer (such as blacklist level, gray list level and white list level) to obtain multiple weights of multiple abnormal indicator levels relative to the target phone number.

[0088] According to the information processing method provided by the embodiment of the present disclosure, data is collected from multiple aspects of the user, and the user data is classified using a convolutional neural network based on a deep learning algorithm based on big data, and risk grading is performed in combination with a hierarchical analysis decision method. The data sources are rich, real-time, and have small errors, which improves the accuracy of the data. The process is simple and the effect is significant.

[0089] Figure 6 Shows Figure 2 FIG. 1 is a schematic diagram of the processing process of step S206 in another embodiment. Figure 6 As shown, in the embodiment of the present disclosure, the above step S206 may further include the following steps.

[0090] Step S602: Obtain preset sensitive words.

[0091] In some embodiments, a list of keywords suspected of telecommunications fraud terms can be defined, and the list includes preset sensitive words, such as identity card, bank card, card number, password, verification code, etc.

[0092] Step S604: obtaining a comparison matrix of multiple call information according to the target category information of the target phone number and preset sensitive words.

[0093] Step S6042, matching the call text according to preset sensitive words to obtain sensitivity information of the call text.

[0094] In some embodiments, for example, the number of sensitive words that successfully match the call text can be obtained, such as 5, 10, 20, etc. The number of successfully matched sensitive words can indicate the sensitivity of the call text. The larger the number, the greater the sensitivity of the call text.

[0095] Step S6044, obtaining a comparison matrix of multiple call information according to the target category information of the target phone number and the sensitivity information of the call text.

[0096] In some embodiments, referring to step S306, the middle-layer factors may also include call content sensitive factors, and the importance of the call content sensitive factors relative to the highest layer (target phone number) can be obtained based on the number of sensitive words that successfully match the call text, thereby obtaining the corresponding values ​​in the comparison matrix of multiple call information.

[0097] Step S606, obtaining a comparison matrix of multiple abnormal indicator levels according to the multiple call information and preset sensitive words.

[0098] Step S6062, obtaining a comparison matrix of multiple abnormal indicator levels according to the sensitivity information of multiple call information and call texts.

[0099] In some embodiments, referring to step S308, the factors of the middle layer may also include call content sensitivity factors, and the importance of each abnormal indicator level relative to the call content sensitivity factors may be obtained based on the number of sensitive words successfully matched in the call text, thereby obtaining the corresponding values ​​in the comparison matrix of multiple abnormal indicator levels.

[0100] According to the information processing method provided by the embodiment of the present disclosure, multiple factors are comprehensively considered, such as whether the target phone number has been in contact with a registered telecommunications fraud mobile phone number, the frequency of contact, and whether there are sensitive fields involving telecommunications fraud during the call, to analyze the risk level of the target phone number as a telecommunications fraud number, thereby improving the accuracy of the analysis results.

[0101] Figure 7 is based on Figures 2 to 6A schematic diagram of an abnormal telephone number identification process is shown in FIG. Figure 7 As shown, the abnormal telephone number identification process may include the following steps.

[0102] Step S702. Definition of sensitive data such as fraud and harassment: Define a list of keywords related to suspected telecommunications fraud, harassment, etc.

[0103] Step S704. Data collection and classification: Obtain the user's original data 7002, which may include, for example, call records, call frequency, travel trajectory, text messages, phone recordings, etc. After collecting the original data 7002, it can be preliminarily classified, for example, according to the city to which the number belongs, the time of network access, whether it is real-name, whether the real-name verification is passed, etc., and then analyze them by category according to the results of the preliminary classification.

[0104] Step S706: Call the call information of the target phone number in the data storage 7004.

[0105] Step S708. Use convolutional neural network to classify the call information of the called target phone number. Use big data technology to conduct a comprehensive analysis of the acquired user information. The big data technology used is a deep learning algorithm. Use convolutional neural network for statistical analysis and real-time processing of the collected data. For specific implementation methods, please refer to Figure 3 .

[0106] Step S710. Fraud risk analysis and result evaluation: Introduce hierarchical analysis method, gray statistical theory and other technologies, use these technologies to evaluate the collected data (refer to Figures 4 to 6 ), divided into different levels, the mobile phone numbers determined to be fraudulent are blacklisted 7006, those suspected of fraud are graylisted 7010, and those not suspected of fraud are whitelisted 7012, and this information is fed back to the mobile phone user 7008 in real time with different identifiers (S714), so that telecommunication fraud can be prevented dynamically and flexibly.

[0107] Step S712. Users who are in contact with the blacklist can be fed back to relevant national agencies in real time, effectively preventing the occurrence of telecommunications fraud.

[0108] According to the abnormal phone number identification method provided by the embodiment of the present disclosure, all mobile phone numbers can be comprehensively analyzed based on big data, and their call records can be subjected to convolutional neural classification and hierarchical analysis. According to the analysis results, these users can be divided into black, white, and gray lists, and different labels can be added to remind users. It can interact with the police system in real time and work with the public security department to prevent telecommunications fraud.

[0109] Figure 8 FIG. 1 is a block diagram of an information processing device according to an exemplary embodiment. Figure 8 The device shown can be applied to the server side of the above system, for example, and can also be applied to the terminal device of the above system.

[0110] refer to Figure 8 The device 80 provided in the embodiment of the present disclosure may include an acquisition module 802, a processing module 804 and an analysis module 806.

[0111] The acquisition module 802 may be used to acquire various call information of a target phone number.

[0112] The various call information may include outgoing number information, call duration information, call time information, and traffic usage information.

[0113] The various call information may include call voice information.

[0114] The call voice information may include a call text obtained by performing natural language processing on the call voice.

[0115] The acquisition module 802 may also be used to acquire preset sensitive words.

[0116] The processing module 804 may be used to classify the target phone number according to various call information to obtain target category information of the target phone number.

[0117] The target category information of the target phone number may include an outgoing number abnormal category weight, a call duration abnormal category weight, a call time abnormal category weight, and a traffic usage abnormal category weight.

[0118] The processing module 804 can also be used to: obtain a call feature vector of the target phone number based on the outgoing number information, call duration information, call time information and traffic usage information; input the call feature vector of the target phone number into a convolutional neural network to obtain the outgoing number abnormal category weight, call duration abnormal category weight, call time abnormal category weight and traffic usage abnormal category weight of the target phone number.

[0119] The analysis module 806 may be used to perform hierarchical analysis based on the target category information of the target phone number and the multiple call information to obtain the abnormal index weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal index weight.

[0120] The anomaly index weight may include a plurality of weights of a plurality of anomaly index levels relative to the target telephone number, respectively.

[0121] Multiple abnormal indicator levels include blacklist level, graylist level and whitelist level. The abnormality level of the blacklist level is higher than that of the graylist level, and the whitelist level is not abnormal.

[0122] The analysis module 806 can also be used to: obtain a comparison matrix of multiple call information based on the target category information of the target phone number, the comparison matrix of multiple call information is used to indicate the importance of each two call information in the multiple call information relative to the target phone number; obtain a comparison matrix of multiple abnormal indicator levels based on the multiple call information, the comparison matrix of multiple abnormal indicator levels is used to indicate the importance of each two abnormal indicator levels in the multiple abnormal indicator levels relative to various call information; obtain multiple weights of the multiple abnormal indicator levels relative to the target phone number based on the comparison matrix of the multiple call information and the comparison matrix of the multiple abnormal indicator levels.

[0123] The analysis module 806 can also be used to compare the abnormal category weights of each two call information among multiple call information based on the abnormal category weights of the outgoing number, call duration, call time and traffic usage of the target phone number, to obtain a comparison matrix of the multiple call information.

[0124] The analysis module 806 can also be used to: perform a consistency check on the comparison matrices of multiple call information according to the target category information of the target telephone number; if the comparison matrices of the multiple call information pass the consistency check, obtain the normalized feature vectors of the comparison matrices of the multiple call information; perform a consistency check on the comparison matrices of multiple abnormal indicator levels according to the multiple call information; if the comparison matrices of multiple abnormal indicator levels pass the consistency check, obtain the normalized feature vectors of the comparison matrices of multiple abnormal indicator levels; obtain multiple weights of multiple abnormal indicator levels relative to the target telephone number according to the normalized feature vectors of the comparison matrices of the multiple call information and the normalized feature vectors of the comparison matrices of multiple abnormal indicator levels.

[0125] The analysis module 806 can also be used to: obtain a comparison matrix of multiple call information based on target category information of the target phone number and preset sensitive words; and obtain a comparison matrix of multiple abnormal indicator levels based on multiple call information and preset sensitive words.

[0126] The analysis module 806 can also be used to: match the call text according to preset sensitive words to obtain the sensitivity information of the call text; obtain a comparison matrix of multiple call information according to the target category information of the target phone number and the sensitivity information of the call text; obtain a comparison matrix of multiple abnormal indicator levels according to the sensitivity information of multiple call information and call text.

[0127] The specific implementation of each module in the device provided by the embodiment of the present disclosure can refer to the content of the above method, which will not be repeated here.

[0128] Fig. 9 The following is a schematic diagram showing the structure of an electronic device in an embodiment of the present disclosure. It should be noted that: Fig. 9 The device shown is only an example of a computer system, which should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0129] like Fig. 9 As shown, the device 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage part 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the device 900 are also stored. The CPU 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0130] The following components are connected to the I / O interface 905: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed, so that a computer program read therefrom is installed into the storage section 908 as needed.

[0131] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part 909, and / or installed from a removable medium 911. When the computer program is executed by a central processing unit (CPU) 901, the above-mentioned functions defined in the system of the present disclosure are executed.

[0132] It should be noted that the computer-readable medium shown in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0133] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0134] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes an acquisition module, a processing module and an analysis module. The names of these modules do not constitute a limitation on the modules themselves in some cases, for example, the data acquisition module may also be described as a "module for acquiring call information from a connected server".

[0135] As another aspect, the present disclosure further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes:

[0136] Acquire multiple call information of the target phone number; classify the target phone number according to the multiple call information to obtain target category information of the target phone number; perform hierarchical analysis based on the multiple call information and the target category information of the target phone number to obtain the abnormal index weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal index weight.

[0137] The exemplary embodiments of the present disclosure are specifically shown and described above. It should be understood that the present disclosure is not limited to the detailed structures, configurations or implementations described herein; on the contrary, the present disclosure is intended to cover various modifications and equivalent configurations included in the spirit and scope of the appended claims.

Claims

1. An information processing method, characterized in that: include: Get various call information of the target phone number; Classify the target phone number according to the multiple call information to obtain target category information of the target phone number; Performing hierarchical analysis according to the multiple call information and the target category information of the target phone number to obtain the abnormal index weight of the target phone number, so as to perform abnormal processing on the target phone number according to the abnormal index weight; in the hierarchical analysis, the target phone number is the highest level, the multiple call information of the target phone number is the multiple factors of the middle level, and the abnormal index level of the target phone number is the lowest level; The abnormality index weight includes a plurality of weights of a plurality of abnormality index levels respectively relative to the target telephone number; Performing hierarchical analysis according to the multiple call information and the target category information of the target phone number to obtain the abnormal index weight of the target phone number includes: Obtaining a comparison matrix of the multiple types of call information according to the target category information of the target phone number, wherein the comparison matrix of the multiple types of call information is used to indicate the importance of each two types of call information in the multiple types of call information relative to the target phone number; Obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information, wherein the comparison matrix of the multiple abnormal indicator levels is used to indicate the importance of each two abnormal indicator levels in the multiple abnormal indicator levels relative to various call information; Based on the normalized feature vector of the comparison matrix of the multiple call information and the normalized feature vector of the comparison matrix of the multiple abnormal indicator levels, multiple weights of the multiple abnormal indicator levels relative to the target telephone number are obtained, wherein the normalized feature vector of each layer is used as the weight vector of the layer, and the weights of each layer are multiplied according to each abnormal indicator level of the lowest layer to obtain multiple weights of the multiple abnormal indicator levels relative to the target telephone number.

2. The method according to claim 1, characterized in that The multiple call information includes outgoing number information, call duration information, call time information and traffic usage information; The target category information of the target phone number includes an abnormal category weight of an outgoing number, an abnormal category weight of a call duration, an abnormal category weight of a call time, and an abnormal category weight of a flow usage; Classifying the target phone number according to the multiple call information to obtain target category information of the target phone number includes: Obtaining a call feature vector of the target phone number according to the outgoing number information, the call duration information, the call time information and the traffic usage information; Inputting the call feature vector of the target phone number into a convolutional neural network to obtain the outgoing number abnormality category weight, call duration abnormality category weight, call time abnormality category weight and traffic usage abnormality category weight of the target phone number; Obtaining a comparison matrix of the plurality of call information according to the target category information of the target phone number includes: According to the outgoing number abnormality category weight, call duration abnormality category weight, call time abnormality category weight and traffic usage abnormality category weight of the target phone number, the abnormal category weights of each two call information among the multiple call information are compared to obtain a comparison matrix of the multiple call information.

3. The method according to claim 1, characterized in that Also includes: Performing consistency check on the comparison matrix of the plurality of call information according to the target category information of the target phone number; If the comparison matrix of the multiple call information passes the consistency check, obtaining a normalized feature vector of the comparison matrix of the multiple call information; Performing consistency check on the comparison matrices of the multiple abnormal indicator levels according to the multiple call information; If the comparison matrices of the multiple abnormal indicator levels pass the consistency check, a normalized eigenvector of the comparison matrices of the multiple abnormal indicator levels is obtained.

4. The method according to claim 1, characterized in that: The multiple call information includes call voice information; The method further comprises: Get preset sensitive words; Obtaining a comparison matrix of the plurality of call information according to the target category information of the target phone number includes: Obtaining a comparison matrix of the multiple call information according to the target category information of the target phone number and the preset sensitive words; Obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information includes: A comparison matrix of the multiple abnormal indicator levels is obtained according to the multiple call information and the preset sensitive words.

5. The method according to claim 4, characterized in that The call voice information includes a call text obtained by performing natural language processing on the call voice; Obtaining a comparison matrix of the multiple call information according to the target category information of the target phone number and the preset sensitive words includes: Matching the call text according to the preset sensitive words to obtain sensitivity information of the call text; Obtaining a comparison matrix of the multiple call information according to the target category information of the target phone number and the sensitivity information of the call text; Obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information and the preset sensitive words includes: A comparison matrix of the multiple abnormal indicator levels is obtained according to the sensitivity information of the multiple call information and the call text.

6. The method according to claim 1, characterized in that The multiple abnormal indicator levels include a blacklist level, a graylist level and a whitelist level. The abnormality degree of the blacklist level is higher than that of the graylist level, and the whitelist level is a non-abnormal level.

7. An information processing device, characterized in that: include: An acquisition module is used to obtain various call information of the target phone number; A processing module, configured to classify the target telephone number according to the multiple call information to obtain target category information of the target telephone number; An analysis module is used to perform hierarchical analysis according to the multiple call information and the target category information of the target phone number, obtain the abnormal index weight of the target phone number, and perform abnormal processing on the target phone number according to the abnormal index weight; in the hierarchical analysis, the target phone number is the highest level, the multiple call information of the target phone number is the multiple factors of the middle level, and the abnormal index level of the target phone number is the lowest level; the abnormal index weight includes multiple weights of multiple abnormal index levels respectively relative to the target phone number; The analysis module is further used to obtain a comparison matrix of the multiple call information according to the target category information of the target phone number, wherein the comparison matrix of the multiple call information is used to indicate the importance of each two call information in the multiple call information relative to the target phone number; Obtaining a comparison matrix of the multiple abnormal indicator levels according to the multiple call information, wherein the comparison matrix of the multiple abnormal indicator levels is used to indicate the importance of each two abnormal indicator levels in the multiple abnormal indicator levels relative to various call information; Based on the normalized feature vector of the comparison matrix of the multiple call information and the normalized feature vector of the comparison matrix of the multiple abnormal indicator levels, multiple weights of the multiple abnormal indicator levels relative to the target telephone number are obtained, wherein the normalized feature vector of each layer is used as the weight vector of the layer, and the weights of each layer are multiplied according to each abnormal indicator level of the lowest layer to obtain multiple weights of the multiple abnormal indicator levels relative to the target telephone number.

8. An information processing device, comprising: A memory, a processor, and executable instructions stored in the memory and executable in the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the executable instructions.

9. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the executable instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images