Number evaluation method and device

Abstract

The present disclosure is about a method and device for evaluating a number. The method includes: obtaining various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a parameter value corresponding to the call parameter type; then obtaining the theoretical recommended value corresponding to each call parameter type of the number to be called and the weight value coefficient corresponding to each call parameter type according to each parameter value corresponding to each call parameter type and a preset recommendation table; finally, obtaining the real-time total call recommendation value of the number to be called according to the theoretical recommended value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type, wherein data sources of multiple call parameter types can be used to analyze the number to be called from multiple dimensions, rather than relying on a single data source, so that different recommended values ​​can be given according to different business needs of the enterprise, helping the enterprise to reduce disturbing behavior to users.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a number evaluation method and device. Background Art

[0002] Currently, some number service providers use known (managed by operators) complainant numbers as blacklists before outbound calls. These complainant numbers are filtered out before outbound calls. This measure is relatively backward and the data feedback is not real-time enough, which will also cause losses to the normal business operations. There is an urgent need for a set of pre-emptive, real-time, wide-coverage, and accurate data algorithm products to solve the current number outbound call problem. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the embodiments of the present disclosure provide a number evaluation method and device. The technical solution is as follows:

[0004] According to a first aspect of the present disclosure, there is provided

[0005] Acquire various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a value corresponding to the call parameter type;

[0006] According to the parameter values ​​corresponding to the call parameter types and the preset recommendation table, the theoretical recommended values ​​corresponding to the call parameter types of the to-be-called number and the weight value coefficients corresponding to the call parameter types are obtained; the preset recommendation table includes: the corresponding relationship between the parameter values ​​corresponding to the call parameter types, the theoretical recommended values ​​of the parameter values ​​corresponding to the call parameter types and the weight value coefficients of the parameter values ​​corresponding to the call parameter types;

[0007] The real-time total call recommendation value of the to-be-called number is obtained according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type.

[0008] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: in the present disclosure, various call parameter information of the number to be called will be obtained; the call parameter information includes: a call parameter type and a parameter value corresponding to the call parameter type; then, according to the various parameter values ​​corresponding to the various call parameter types and a preset recommendation table, the theoretical recommended values ​​corresponding to the various call parameter types of the number to be called and the weight value coefficients corresponding to the various call parameter types are obtained; finally, according to the theoretical recommended values ​​corresponding to the various call parameter types and the weight value coefficients corresponding to the various call parameter types, the real-time total call recommendation value of the number to be called is obtained, wherein data sources of various call parameter types can be used to analyze the number to be called from multiple dimensions instead of relying on a single data source, so that different recommendation values ​​can be given according to the different business needs of the enterprise, helping the enterprise to reduce disturbance to users.

[0009] In one embodiment, before obtaining the call parameter information of the number to be called, the method further includes:

[0010] Obtain call parameter information corresponding to each sample number;

[0011] Training a support vector machine model according to the call parameter information corresponding to each sample number to obtain a total recommendation value corresponding to each call parameter type;

[0012] Obtaining weight value coefficients corresponding to each of the call parameter types;

[0013] The preset recommendation table is obtained at least according to the total recommendation values ​​corresponding to the various call parameter types and the weight value coefficients corresponding to the various call parameter types.

[0014] In one embodiment, the obtaining the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types includes:

[0015] Dividing all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type;

[0016] Obtaining a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located;

[0017] Dividing all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types;

[0018] The weight value coefficient corresponding to the parameter value of the call parameter type is obtained according to the first preset rule and the number of the de-weighting interval in which the parameter value of the call parameter type is located.

[0019] In one embodiment, the step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes:

[0020] Obtain a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type;

[0021] Obtaining a total weight coefficient; the total weight coefficient is the product of the weight coefficients corresponding to each call parameter type;

[0022] The product of the total theoretical recommendation value and the total weight value coefficient is obtained as the real-time total call recommendation value.

[0023] In one embodiment, the call parameter types include: business type, density of marketing calls made by the enterprise, complaints in the region where the called user is located, whether the called user itself has a negative mark, service mark produced by the called user, whether the called user has a good number and the degree of the good number, number activity of the called user, degree of connection between the called user and the enterprise's business industry, degree of connection between the called user and the enterprise's business itself, geographical relationship between the called user and the enterprise's business, degree of sensitivity of the called user to unfamiliar numbers, the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and registration status of rejected calls.

[0024] In one embodiment, the call parameter type is the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and the DNS registration status;

[0025] The obtaining, according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located, a theoretical recommended value corresponding to the parameter value of the call parameter type includes:

[0026] If the number is in the blacklist database of the enterprise itself, or the number is connected with the complaint agency, or the number is registered in the call rejection service, the corresponding theoretical interval is the first interval;

[0027] If the number is not in the blacklist of the enterprise, or the number has no connection with the complaint agency, or the number is not registered in the call rejection service, the corresponding theoretical interval is the second interval;

[0028] The theoretical recommended value corresponding to the first interval is 0, and the theoretical recommended value corresponding to the second interval is the total recommended value corresponding to the call parameter type;

[0029] The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes:

[0030] If the theoretical recommendation value corresponding to the enterprise's own blacklist database, whether the called user has contact with the complaint agency, or the DNS registration status is 0, the real-time total call recommendation value of the to-be-called number is 0.

[0031] According to a second aspect of an embodiment of the present disclosure, there is provided a number evaluation device, comprising:

[0032] A first acquisition module is used to acquire various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a value corresponding to the call parameter type;

[0033] The second acquisition module is used to obtain the theoretical recommended values ​​corresponding to the various call parameter types of the to-be-called number and the weight value coefficients corresponding to the various call parameter types according to the various parameter values ​​corresponding to the various call parameter types and the preset recommendation table; the preset recommendation table includes: the corresponding relationship between the various parameter values ​​corresponding to the various call parameter types, the theoretical recommended values ​​of the various parameter values ​​corresponding to the various call parameter types and the weight value coefficients of the various parameter values ​​corresponding to the various call parameter types;

[0034] The third acquisition module is used to acquire the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type.

[0035] In one embodiment, the apparatus further comprises:

[0036] A third acquisition module, used for acquiring the call parameter information corresponding to each sample number before the first acquisition module acquires the call parameter information of the number to be called;

[0037] A fourth acquisition module, used for training a support vector machine model according to the call parameter information corresponding to each sample number, and obtaining a total recommendation value corresponding to each call parameter type;

[0038] A fifth acquisition module, used to obtain a weight value coefficient corresponding to each of the call parameter types;

[0039] The sixth acquisition module is used to acquire the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types.

[0040] In one embodiment, the sixth acquisition module includes:

[0041] A first division submodule, used to divide all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type;

[0042] A first acquisition submodule is used to obtain a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type and the theoretical number of intervals in which the parameter value of the call parameter type is located;

[0043] A second division submodule, configured to divide all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types;

[0044] The second acquisition submodule is used to acquire the weight value coefficient corresponding to the parameter value of the call parameter type according to the first preset rule and the number of the de-weighting interval where the parameter value of the call parameter type is located.

[0045] In one embodiment, the third acquisition module includes:

[0046] The fourth acquisition submodule is used to acquire a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type;

[0047] The fifth acquisition submodule is used to acquire a total weight value coefficient; the total weight value coefficient is the product of the weight value coefficients corresponding to each call parameter type;

[0048] The sixth acquisition submodule is used to obtain the product of the total theoretical recommendation value and the total weight value coefficient as the real-time total call recommendation value.

[0049] In one embodiment, the call parameter types include: business type, density of marketing calls made by the enterprise, complaints in the region where the called user is located, whether the called user itself has a negative mark, service mark produced by the called user, whether the called user has a good number and the degree of the good number, number activity of the called user, degree of connection between the called user and the enterprise's business industry, degree of connection between the called user and the enterprise's business itself, geographical relationship between the called user and the enterprise's business, degree of sensitivity of the called user to unfamiliar numbers, the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and registration status of rejected calls.

[0050] In one embodiment, the call parameter type is the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and the DNS registration status;

[0051] The first acquisition submodule includes:

[0052] A first determination subunit is used for, if the number is in the blacklist database of the enterprise itself, or the number is connected with the complaint agency, or the number is registered in the call rejection, then the corresponding theoretical interval is the first interval;

[0053] A second determination subunit is used for determining that if the number is not in the blacklist database of the enterprise itself, or the number has no contact with the complaint agency, or the number is not registered in the call rejection, the corresponding theoretical interval is the second interval;

[0054] The theoretical recommended value corresponding to the first interval is 0, and the theoretical recommended value corresponding to the second interval is the total recommended value corresponding to the call parameter type;

[0055] The third acquisition module includes:

[0056] The determination submodule is used for determining that if the theoretical recommended value corresponding to the enterprise's own blacklist database, whether the called user has contact with the complaint agency, or the DNS registration status is 0, the real-time total call recommendation value of the to-be-called number is 0.

[0057] According to a third aspect of an embodiment of the present disclosure, there is provided a number evaluation device, including:

[0058] processor;

[0059] a memory for storing processor-executable instructions;

[0060] Wherein, the processor is configured to:

[0061] Acquire various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a value corresponding to the call parameter type;

[0062] According to the parameter values ​​corresponding to the call parameter types and the preset recommendation table, the theoretical recommended values ​​corresponding to the call parameter types of the to-be-called number and the weight value coefficients corresponding to the call parameter types are obtained; the preset recommendation table includes: the corresponding relationship between the parameter values ​​corresponding to the call parameter types, the theoretical recommended values ​​of the parameter values ​​corresponding to the call parameter types and the weight value coefficients of the parameter values ​​corresponding to the call parameter types;

[0063] The real-time total call recommendation value of the to-be-called number is obtained according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type.

[0064] According to a fourth aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium on which computer instructions are stored, and when the instructions are executed by a processor, the steps of any one of the methods described in the first aspect are implemented.

[0065] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0067] Figure 1 The figure is a flowchart of a method for evaluating a number according to an exemplary embodiment.

[0068] Figure 2 The figure is a flowchart of a method for evaluating a number according to an exemplary embodiment.

[0069] Figure 3 is a flow chart of a method for evaluating a number according to a first exemplary embodiment.

[0070] Figure 4 The figure is a block diagram of a number evaluation device according to an exemplary embodiment.

[0071] Figure 5 is a block diagram of a device 80 for evaluating a number according to an exemplary embodiment. DETAILED DESCRIPTION

[0072] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0073] Since its establishment in 2012, Telangbang has been deeply involved in the communications industry and has been committed to combating harassment and fraudulent calls. It has rich experience in communications data processing and leading technology. Based on the current status of the communications industry, Telangbang has joined forces with call centers, outbound call companies, Internet security software companies, mobile smart terminals and other industry companies to jointly initiate the establishment of an "Intelligent Anti-Harassment Service Platform" (hereinafter referred to as the "Platform") to seek a balance between the protection of individual user rights and the legitimate promotion of marketing companies' services, and to provide companies with a reasonable communications solution.

[0074] The intelligent anti-harassment service platform can predict the risk index of user numbers that the company is about to call out through big data algorithm analysis and data mining, reduce the number of calls to user numbers that are unintentional, highly sensitive, suspected of being invalid, etc., and improve the efficiency of corporate number operations.

[0075] Currently, some number service providers use known (managed by operators) complainant numbers as blacklists before outbound calls. These complainant numbers are filtered out before outbound calls. This measure is relatively backward and the data feedback is not real-time enough, which will also cause losses to the normal business operations. There is an urgent need for a set of pre-emptive, real-time, wide-coverage, and accurate data algorithm products to solve the current number outbound call problem.

[0076] First, let’s introduce the nouns in this disclosure:

[0077] Number: phone number;

[0078] Number segment: mobile phone number or landline number, minus the last two digits;

[0079] Number location: The fixed area where the number is distributed. This area is the number location;

[0080] Life mark: The mark type is one of courier delivery and taxi driver;

[0081] Negative tags: Tags include intermediary, sales, harassment calls, recruitment headhunters, and suspected fraud;

[0082] Dialing density: the number of calls divided by the time in a given time period;

[0083] Good number: also known as "lucky number", generally refers to a number composed of consecutive identical Arabic numerals that is easy to remember;

[0084] The degree of beauty of the number: the degree of ease of remembering;

[0085] Number activity: the number of calls in a given time period;

[0086] Demotion: The final recommendation value will be multiplied by a percentage value less than 1 and greater than 0;

[0087] Weighting: The final recommendation value will be multiplied by a percentage greater than 1;

[0088] Square: It is an operation. For example, the square of a represents a×a, which is abbreviated as a2;

[0089] Normal distribution: also known as Gaussian distribution (English: Gaussian distribution) and regular distribution, is a very common continuous probability distribution.

[0090] Support Vector Machine (SVM): It is a generalized linear classifier that performs binary classification of data in a supervised learning manner. Its decision boundary is the maximum-margin hyperplane solved for the learning samples.

[0091] Complaints Authority: Complaints handling agency;

[0092] DNC registration: No incoming call registration is accepted.

[0093] Figure 1 is a flow chart of a method for evaluating a number according to an exemplary embodiment. Figure 1 As shown, the method includes the following steps S101-S103:

[0094] In step S101, various call parameter information of the number to be called is obtained; the call parameter information includes: a call parameter type and a parameter value corresponding to the call parameter type;

[0095] For example, the call parameter types include: business type, the time when the enterprise makes marketing calls, complaints in the region where the called user is located, whether the called user has a negative mark, the called user produces a service mark, whether the called user has a good number and the degree of the good number, the activity of the called user's number, the degree of connection between the called user and the enterprise's business industry, the degree of connection between the called user and the enterprise's business itself, the geographical relationship between the called user and the enterprise's business, the called user's sensitivity to unfamiliar numbers, the enterprise's own blacklist library, whether the called user has any connection with the complaint agency, and the DNS registration status.

[0096] In step S102, the theoretical recommended values ​​and weight value coefficients corresponding to the various call parameter types of the to-be-called number are obtained according to the various parameter values ​​corresponding to the various call parameter types of the to-be-called number and the preset recommendation table; the preset recommendation table includes: the corresponding relationship between the various parameter values ​​corresponding to the various call parameter types, the theoretical recommended values ​​of the various parameter values ​​corresponding to the various call parameter types, and the weight value coefficients of the various parameter values ​​corresponding to the various call parameter types;

[0097] In step S103, the real-time total call recommendation value of the number to be called is obtained according to the theoretical recommendation value and the weight value coefficient corresponding to each call parameter type of the number to be called.

[0098] In one embodiment, Figure 2 As shown, the above step S103 includes the following sub-steps 1031-1033:

[0099] In step S1031, a total theoretical recommended value is obtained; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type;

[0100] In step S1023, a total weight coefficient is obtained; the total weight coefficient is the product of the weight coefficients corresponding to each call parameter type;

[0101] In step S1033, the product of the total theoretical recommendation value and the total weight value coefficient is obtained as the real-time total call recommendation value.

[0102] In the present disclosure, various call parameter information of the number to be called will be obtained; the call parameter information includes: a call parameter type and a parameter value corresponding to the call parameter type; then, according to the various parameter values ​​corresponding to the various call parameter types and a preset recommendation table, the theoretical recommended values ​​corresponding to the various call parameter types of the number to be called and the weight value coefficients corresponding to the various call parameter types are obtained; finally, according to the theoretical recommended values ​​corresponding to the various call parameter types and the weight value coefficients corresponding to the various call parameter types, the real-time total call recommendation value of the number to be called is obtained, wherein data sources of various call parameter types can be used to analyze the calling number from multiple dimensions instead of relying on a single data source, so that different recommendation values ​​can be given according to the different business needs of the enterprise, helping the enterprise to reduce disturbance to users.

[0103] In one possible implementation, Figure 3 As shown, before step S101, the method further includes the following steps 104-107:

[0104] In step S104, call parameter information corresponding to each sample number is obtained;

[0105] In step S105, a support vector machine model is trained according to the call parameter information corresponding to each sample number to obtain a total recommendation value corresponding to each call parameter type;

[0106] In the present disclosure, the above-mentioned call parameter types are used as features, and the support vector machine model in machine learning is used to train sample data. The final model weight coefficient obtained is the ratio of the recommended values ​​of each dimension; then, based on the given preset total recommended value, the total recommended value corresponding to each of the call parameter types can be obtained.

[0107] For example, the recommended value ratios corresponding to the various call parameter types are shown in Table 1:

[0108] Table 1 Recommended value ratios corresponding to various call parameter types

[0109] Call parameter type Recommended value ratio Business Type 11 The time when companies make marketing calls 18 Complaints in the region where the called user is located 12 Whether the called user has a negative tag 4 The called user produces a service tag, 1 Whether the called user has a premium number and the degree of premium number 16 Activity of the called user's number 6 The degree of connection between the called user and the enterprise's business industry 15 The degree of connection between the called user and the enterprise's business 18 Geographical relationship between the called user and the enterprise business 2 The sensitivity of the called user to unfamiliar numbers 9 Enterprise's own blacklist database 20 The degree of contact between the called user and the complaint agency 20 dnc registration status 20

[0110] For example, if the preset total recommended value is 860 points, the total recommended values ​​corresponding to the call parameter types are shown in Table 2:

[0111] Table 2 Total recommended values ​​corresponding to each call parameter type

[0112] Call parameter type Total recommended value Business Type 55 The time when companies make marketing calls 90 Complaints in the region where the called user is located 60 Whether the called user has a negative tag 20 The called user produces a service tag, 5 Whether the called user has a premium number and the degree of premium number 80 Activity of the called user's number 30 The degree of connection between the called user and the enterprise's business industry 75 The degree of connection between the called user and the enterprise's business 90 Geographical relationship between the called user and the enterprise business 10 The sensitivity of the called user to unfamiliar numbers 45 Enterprise's own blacklist database 100 The degree of contact between the called user and the complaint agency 100 dnc registration status 100

[0113] In step S106, the weight value coefficient corresponding to each call parameter type is obtained according to the call parameter information corresponding to each sample number;

[0114] Specifically, step S106 includes the following sub-steps A1-A2:

[0115] A1, dividing all parameter values ​​corresponding to each call parameter type into multiple weight reduction intervals, each weight reduction interval including parameter values ​​corresponding to multiple call parameter types;

[0116] A2: Obtain a weight value coefficient corresponding to the parameter value of the call parameter type according to the first preset rule and the number of the weight reduction interval in which the parameter value of the call parameter type is located.

[0117] Taking the density of marketing calls made by enterprises as an example, as shown in Table 3, all parameter values ​​corresponding to the density of marketing calls made by enterprises are divided into 5 weight reduction intervals. The first preset rule is: weight value coefficient = 1-10%*weight reduction interval*2. Then, as shown in Table 3, when the parameter values ​​of the call parameter type are 0 and 1, the corresponding weight reduction interval is 0, and the corresponding weight reduction coefficient is 100%, and so on.

[0118] Table 3 Weight coefficient table of the density of marketing calls made by enterprises

[0119] scope Demotion range Weight coefficient 0～1 0 1-10％*0*2＝100％ 1～2 1 1-10％*1*2＝80％ 2～4 2 1-10％*2*2＝60％ 4～7 3 1-10％*3*2＝40％ 7～+∞ 4 1-10％*4*2＝20％

[0120] In step S107, a preset recommendation table is obtained according to at least the total recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to the parameter value of each call parameter type.

[0121] In one embodiment, the above step S107 includes the following sub-steps B1-B2:

[0122] B1, dividing all parameter values ​​corresponding to each call parameter type into multiple theoretical intervals, each theoretical interval including parameter values ​​corresponding to multiple call parameter types;

[0123] B2, obtaining a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located.

[0124] Take the complaint situation in the called user's region as an example, as shown in Table 4, the total recommended value of the complaint situation in the called user's region is 60; according to the formula: theoretical recommended value = total recommended value - (total recommended value / total theoretical interval number) * the theoretical interval number of the current parameter value, then, as shown in Table 4, if this area is within the statistical time period, such as a week, if the complaint record in the user's region is 8, then the theoretical interval falls in the 3 interval, and the theoretical recommended value is 24.

[0125] Table 4 Theoretical recommended values ​​of complaints in the region where the called user is located

[0126] scope Theoretical interval Theoretical recommended value 0～1 0 60–(60 / 5)*0＝60 1～3 1 60–(60 / 5)*1＝48 3～7 2 60–(60 / 5)*2＝36 7～16 3 60–(60 / 5)*3＝24 16～+∞ 4 60–(60 / 5)*4＝12

[0127] In one embodiment, the call parameter type is the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and the DNS registration status;

[0128] The obtaining, according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located, a theoretical recommended value corresponding to the parameter value of the call parameter type includes:

[0129] If the number is in the blacklist of the enterprise itself, or the number is connected with the complaint agency, or the number is registered in the DNC, the corresponding theoretical interval is the first interval;

[0130] If the number is not in the blacklist of the enterprise, or the number has no connection with the complaint agency, or the number is not registered in the DNC, the corresponding theoretical interval is the second interval;

[0131] The theoretical recommended value corresponding to the first interval is 0, and the theoretical recommended value corresponding to the second interval is the total recommended value corresponding to the call parameter type;

[0132] The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes:

[0133] If the theoretical recommendation value corresponding to the enterprise's own blacklist database, whether the called user has contact with the complaint agency, or the DNS registration status is 0, the real-time total call recommendation value of the to-be-called number is 0.

[0134] The following is a detailed analysis of each call parameter type:

[0135] 1. Business Type

[0136] Note: In this disclosure, the business types are divided into 1: notification and return visit, 2: debt collection, 3: voice verification code, 4: finance, 5: real estate, 6: insurance, 7: securities, 8: education and training, 9: financial management and loan, 10: recruitment, 11: stock, 12: health care, 13: travel, 14: online shopping, 15: food and medicine. Then, according to the degree of direct connection with money, the deeper the degree of direct connection with money, the lower the recommended value. The total recommended value of this call parameter type is currently 55;

[0137] For example: If the customer is notifying for a follow-up visit, a theoretical recommendation value of 55 will be given. If it is a stock, a theoretical recommendation value of 5.5 will be given.

[0138] The specific scoring rules are shown in Table 5:

[0139] Table 5

[0140] Business Type Theoretical range Theoretical recommended value Notification return visit, voice verification code 0 55–(55 / 10)*0＝55 Recruitment 1 55–(55 / 10)*1＝49.5 Real estate, education and training 2 55–(55 / 10)*2＝44 Tourism, Food and Medicine 3 55–(55 / 10)*3＝38.5 Health care, 4 55–(55 / 10)*4＝33 Online shopping 5 55–(55 / 10)*5＝27.5 Debt Collection 6 55–(55 / 10)*6＝22 Finance, Insurance, 7 55–(55 / 10)*7＝16.5 Financial Loans 8 55–(55 / 10)*8＝11 Stocks and securities 9 55–(55 / 10)*9＝5.5

[0141] It is worth noting that the weight coefficient corresponding to the business type is 1.

[0142] 2. The density of marketing calls made by enterprises

[0143] Note: Within a specific period of time (one week), the higher the number of times a similar business enterprise calls the same user, the lower the recommended value, and this call parameter type is a downgraded call parameter type. Once triggered, the overall score will be downgraded. Within a unit of time (3 hours), the more the maximum number of calls, the more downgraded the score. The downgrade coefficient (weight value coefficient) = 1-10%*downgrade interval*2. The current total recommended value for this call parameter type is 90;

[0144] For example: within the statistical period, such as a week, all the marketing numbers of such enterprises have a dial count of 4, which falls within the theoretical interval of 3, and the corresponding theoretical recommendation value is 36. If the maximum number of dials per unit time is 3, the downgrade interval is 2, and the weight coefficient is 60%. If the total recommended value is finally calculated to be 500, the final recommended value is 500*60%=300.

[0145] The specific scoring rules are shown in Table 6 and Table 7:

[0146] Table 6

[0147] scope Theoretical interval Theoretical recommended value 0～1 0 90–(90 / 5)*0＝90 1～2 1 90–(90 / 5)*1＝72 2～4 2 90–(90 / 5)*2＝64 4～8 3 90–(90 / 5)*3＝36 8～16 4 90–(90 / 5)*4＝18 8～+∞ 5 90–(90 / 5)*5＝0

[0148] The weight reduction coefficient rules are shown in Table 7:

[0149] Table 7

[0150] scope Demotion range Weight coefficient 0～1 0 1-10％*0*2＝100％ 1～2 1 1-10％*1*2＝80％ 2～4 2 1-10％*2*2＝60％ 4～7 3 1-10％*3*2＝40％ 7～+∞ 4 1-10％*4*2＝20％

[0151] 3. Complaints in the region where the called user is located

[0152] Note: Based on the called user's number segment, the number of complaints in the user's area is counted. The more complaints there are, the less recommended it is. The total recommended value for this call parameter type is currently 60.

[0153] For example: If the number of complaints recorded in the user's area during the statistical time period, such as a week, is 8, it falls within the theoretical interval of 3, and the theoretical recommended value is 24.

[0154] The specific scoring rules are shown in Table 8:

[0155] Table 8

[0156] scope Theoretical interval Theoretical recommended value 0～1 0 60–(60 / 5)*0＝60 1～3 1 60–(60 / 5)*1＝48 3～7 2 60–(60 / 5)*2＝36 7～16 3 60–(60 / 5)*3＝24 16～+∞ 4 60–(60 / 5)*4＝12

[0157] It is worth noting that the weight coefficient corresponding to the complaint situation in the region where the called user is located is 1.

[0158] 4. Does the called user have a negative label?

[0159] Note: Statistics on the negative marks of users. If the marks are more serious, the less recommended it is. And this call parameter type is a downgraded call parameter type. Once triggered, the overall score will be downgraded. The downgrade coefficient (weight value coefficient) = interval * 3%. The total recommended value of this call parameter type is 20.

[0160] For example: If the number of negative user marks in this area is 5 during the statistical time period, such as 3 months, it falls in the theoretical interval 2, the theoretical recommendation value is 12, and the overall score is reduced by 2*3%=6%. If the last calculated minutes are 500, after the reduction adjustment it is 500*(1-6%)=470.

[0161] The specific scoring rules are shown in Table 9:

[0162] Table 9

[0163] scope Theoretical interval Theoretical recommended value 0～1 0 20–(20 / 5)*0＝20 1～4 1 20–(20 / 5)*1＝16 4～9 2 20–(20 / 5)*2＝12 9～16 3 20–(20 / 5)*3＝8 16～25 4 20–(20 / 5)*4＝4 25～+∞ 5 20–(20 / 5)*5＝0

[0164] 5. The called user generates a service tag

[0165] Note: Statistics on the user's production service tag. The more relevant it is to the enterprise business, the less recommended it is. The total recommended value for this call parameter type is currently 5.

[0166] For example: If the number of user production service tags in this area is 5 during the statistical time period, such as 3 months, it falls in the theoretical interval 2 and the theoretical recommended value is 3.

[0167] The specific scoring rules are shown in Table 10:

[0168] Table 10

[0169] scope Theoretical range Theoretical recommended value 0～1 0 5–(5 / 5)*0＝5 1～4 1 5–(5 / 5)*1＝4 4～9 2 5–(5 / 5)*2＝3 9～16 3 5–(5 / 5)*3＝2 16～+∞ 4 5–(5 / 5)*4＝1

[0170] It is worth noting that the weight coefficient corresponding to the service mark produced by the called user is 1.

[0171] 6. Whether the called user has a premium number and the degree of premium number

[0172] Note: Determine whether the user has a good number and the degree of goodness of the number. If it is a good number, the more good the number is, the less recommended it is. In addition, this call parameter type is a downgraded call parameter type. Once triggered, the overall score will be downgraded. The downgrade coefficient (weight value coefficient) = interval * 15%. The current total recommended value for this call parameter type is 80.

[0173] For example: If the user’s number quality is 3, it falls in the theoretical range 3, and the theoretical recommendation value is 40. The overall recommendation value is reduced by 3*15%=45%. If the final calculated recommendation value is 500, after the reduction adjustment it becomes 500*(1–45%)=275.

[0174] The specific scoring rules are shown in Table 11:

[0175] Table 11

[0176]

[0177]

[0178] The rules for the degree of beauty are shown in Table 12:

[0179] Table 12

[0180]

[0181] 7. Activity of the called user's number

[0182] Note: Calculate the normal distribution position of user activity among all users. The closer it is to the sides, the less recommended it is. The total recommended value for this call parameter type is currently 30.

[0183] For example: If the number of users deviating from the center in this area during the statistical time period, such as 1 week, is 5, then it falls in the theoretical interval 3, and the theoretical recommended value is 15.

[0184] The specific scoring rules are shown in Table 13:

[0185] Table 13

[0186] Off-center range Theoretical interval Theoretical recommended value 0～1 0 30–(30 / 6)*0＝30 1～2 1 30–(30 / 6)*1＝25 2～4 2 30–(30 / 6)*2＝20 4～8 3 30–(30 / 6)*3＝15 8～16 4 30–(30 / 6)*4＝10 16～+∞ 5 1–(30 / 6)*5＝5

[0187] It is worth noting that the weight coefficient corresponding to the called user's number activity is 1.

[0188] 8. Degree of connection between the called user and the enterprise's business industry

[0189] Note: Determine the degree of connection between the user and the enterprise business industry. The lower the degree of connection and the longer the contact time, the less recommended it is. This call parameter type is a downgraded call parameter type. It is triggered in one section and the overall score is downgraded. The downgrade coefficient (weight value coefficient) = (maximum value of the interval – interval) * 8%. The current total recommended value for this call parameter type is 75.

[0190] For example: If the user’s industry connection level is 3, it falls in the theoretical interval 2, and the theoretical recommended value is 25. The overall score is downgraded by (6–2)*8%=32%. If the final calculated score is 500, after the downgrade adjustment it becomes 500*(1-32%)=320.

[0191] The specific scoring rules are shown in Table 14:

[0192] Table 14

[0193] Industry connections Theoretical interval Theoretical recommended value 0～1 0 (75 / 6)*0＝0 1～2 1 (75 / 6)*1＝12.5 2～4 2 (75 / 6)*2＝25 4～8 3 (75 / 6)*3＝37.5 8～16 4 (75 / 6)*4＝50 16～32 5 (75 / 6)*5＝62.5 32～+∞ 6 (75 / 6)*6＝75

[0194] 9. The degree of connection between the called user and the enterprise business itself

[0195] Note: Determine the degree of contact between the user and the enterprise business itself. The lower the degree of contact and the longer the contact time, the less recommended it is. However, if the recommended value of the previous call parameter type (the degree of contact between the called user and the enterprise business industry) is higher, it is more recommended. Weighting coefficient (weight value coefficient) = (1 + (the degree of contact between the called user and the enterprise business industry) interval * 15% * (maximum value of interval - interval)) square root, recommended value = original recommended value * recommended coefficient, the current total recommended value of this call parameter type is 90.

[0196] For example: If the degree of connection between the user and the enterprise business is 3, it falls in the theoretical interval 2, and the theoretical recommended value is 30. The interval of the previous call parameter type (the degree of connection between the called user and the enterprise business industry) is 4, then the weighting coefficient is 1+4*15%*square root of (6-2)=1.84. If the final calculated total recommended value is 500, then the final recommended value is 500*1.84=920.

[0197] The specific scoring rules are shown in Table 15:

[0198] Table 15

[0199] Degree of business connection Theoretical interval Theoretical recommended value 0～1 0 (90 / 6)*0＝0 1～2 1 (90 / 6)*1＝15 2～4 2 (90 / 6)*2＝30 4～8 3 (90 / 6)*3＝45 8～16 4 (90 / 6)*4＝60 16～32 5 (90 / 6)*5＝75 32～+∞ 6 (90 / 6)*6＝90

[0200] 10. Geographical relationship between the called user and the enterprise business

[0201] Note: Determine the geographical relationship between the user and the enterprise. If the enterprise business is life-related, the farther the geographical relationship, the less recommended it is. In addition, this call parameter type is a downgraded call parameter type. It is triggered in one section, and finally the overall score is downgraded. The downgrade coefficient (weight value coefficient) = interval * 1%. The current total recommended value for this call parameter type is 10.

[0202] For example, if the user’s geographical connection level is 3, it falls in the theoretical interval 2, and the theoretical recommendation value is 6 points. The overall score is reduced by 2*1%=2%. If the final calculated score is 500, after the reduction adjustment it is 500*(1–2%)=490.

[0203] The specific scoring rules are shown in Table 16:

[0204] Table 16

[0205] Geographical distance Theoretical interval Theoretical recommended value 0～1 0 10–(10 / 5)*0＝10 1～2 1 10–(10 / 5)*1＝8 2～4 2 10–(10 / 5)*2＝6 4～8 3 10–(10 / 5)*3＝4 8～16 4 10–(10 / 5)*4＝2 16～+∞ 5 1–(10 / 5)*5＝0

[0206] 11. How sensitive are called users to unfamiliar numbers?

[0207] Description: Determine the user's sensitivity to unfamiliar numbers. It is believed that the more homely the user is, the more sensitive he is. If the user is more sensitive, the call is less recommended. In addition, this call parameter type is a downgraded call parameter type. Once triggered, the overall score will be downgraded. The downgrade coefficient (weight value coefficient) = 1-interval*2%. The total recommended value of this call parameter type is 45.

[0208] For example: If the user’s sensitivity is 3, it falls in the theoretical range 2, and the theoretical recommended value is 30. The overall score is reduced by 2*2%=4%. If the final calculated score is 500, after the reduction adjustment it becomes 500*(1–4%)=480.

[0209] The specific scoring rules are shown in Table 17:

[0210] Table 17

[0211] Sensitivity Theoretical interval Theoretical recommended value 0～1 0 45–(45 / 6)*0＝45 1～2 1 45–(45 / 6)*1＝37.5 2～4 2 45–(45 / 6)*2＝30 4～8 3 45–(45 / 6)*3＝22.5 8～16 4 45–(45 / 6)*4＝15 16～32 5 45–(45 / 6)*5＝7.5 32～+∞ 6 1–(45 / 6)*6＝0

[0212] 12. Enterprise’s own blacklist database

[0213] Note: Determine whether the user is in the blacklist of the enterprise. Once confirmed, it is not recommended, and this call parameter type is a veto call parameter type. Once triggered, the final overall score will be 0. The current total recommended value of this call parameter type is 100.

[0214] For example: If the user is not in the database, the recommendation value is 100. If the user is in the database, the overall recommendation value is 0. If the final calculated recommendation value is 500, but the score of this call parameter type is 0, the overall recommendation is rejected and the recommendation value is 0.

[0215] The specific scoring rules are shown in Table 18:

[0216] Table 18

[0217] Is it in the blacklist database? Theoretical interval Theoretical recommended value exist 0 0 Not Available 1 100

[0218] 13. Degree of contact between the called user and the complaint agency

[0219] Description: Determine whether the user has contact with the complaint agency within a period of time. Once confirmed, it is not recommended, and this call parameter type is a veto call parameter type. Once triggered, the final overall score will be 0. The current total recommended value of this call parameter type is 100.

[0220] For example: If the user contacts the complaint agency, the recommendation value will be reset to 0. If the final calculated recommendation value is 500, but the recommendation value of this call parameter type is 0, the entire request will be rejected and the recommendation value will be reset to 0.

[0221] The specific scoring rules are shown in Table 19:

[0222] Table 19

[0223] Contacting a Complaints Agency Theoretical interval Theoretical recommended value have 0 0 No 1 100

[0224] 14. DNC registration status

[0225] Description: Determine whether the user has registered in the dnc within a period of time. Once confirmed, it is not recommended, and this call parameter type is a veto call parameter type. Once triggered, the final overall score will be 0. The current total recommended value of this call parameter type is 100.

[0226] For example: If the user has registered in dnc, the recommended value is 0. If the final calculated recommended value is 500, but the value of this call parameter type is 0, the whole call is rejected and the recommended value is set to 0.

[0227] The specific scoring rules are shown in Table 20:

[0228] Table 20

[0229] dnc registration Theoretical interval Theoretical recommended value have 0 0 No 1 100

[0230] In combination with the above embodiments, the role of each call parameter type in the final obtained real-time total call recommendation value can be obtained as shown in Table 21:

[0231] Table 21

[0232] serial number Call parameter type illustrate 1 Business Type 2 The time when companies make marketing calls Demotion Dimension 3 Complaints in the region where the called user is located 4 Whether the called user has a negative tag Demotion Dimension 5 The called user produces a service tag, 6 Whether the called user has a premium number and the degree of premium number Demotion Dimension 7 Activity of the called user's number 8 The degree of connection between the called user and the enterprise's business industry Demotion Dimension 9 The degree of connection between the called user and the enterprise's business Weighted Dimensions 10 Geographical relationship between the called user and the enterprise business Demotion Dimension 11 The sensitivity of the called user to unfamiliar numbers Demotion Dimension 12 Enterprise's own blacklist database Veto Dimension 13 The degree of contact between the called user and the complaint agency Veto Dimension 14 dnc registration status Veto Dimension

[0233] The method of the present disclosure is described in detail below through specific examples.

[0234] 1. Training of total recommended values ​​for each parameter type:

[0235] Using the above Table 21 as the feature, the support vector machine model in machine learning is used to train the actual data. The final model weight coefficient is the ratio of the recommended values ​​of each parameter type.

[0236] The total recommended values ​​for each parameter type are shown in Table 22:

[0237] Table 22

[0238]

[0239]

[0240] 2. Adjustment of weight values ​​of each parameter type:

[0241] Although the total recommended values ​​of each parameter type have been trained, in line with the principle of minimizing disturbance to users and the principle of people-oriented, through actual observation and verification, the weights of each dimension are manually adjusted as shown in Table 23:

[0242] Table 23

[0243]

[0244]

[0245] Assume that a certain user, the company's business type is real estate, the time interval value for the enterprise to make marketing calls is 2, the maximum number of calls is 1, the number of complaints in the called user's region is 3, whether the called user has a negative mark is 1, the called user's production service mark is 0, the degree to which the called user is a good number is 1, the called user's number activity deviation from the center value is 1, the called user's industry connection degree is 3, the called user's connection degree with the enterprise's business itself is 1, the called user's geographical relationship with the enterprise's business is 3, the called user's sensitivity to unfamiliar numbers is 1, the enterprise's own blacklist database does not have him, the called user's connection degree with the complaint agency is 0, and the dnc registration situation is not; then the calculation of each score is as shown in the following table 24:

[0246] Table 24

[0247]

[0248]

[0249] The final total recommended value is calculated as:

[0250] (44+64+36+16+5+66.67+25+25+15+6+37.5+100+100+100)*80%*97%*85%*64%*158%*97%*98%=408.43.

[0251] The algorithm disclosed in the present invention can be applied to an intelligent anti-harassment service platform. The algorithm is based on the user's willingness to answer marketing calls, and uses AI + big data analysis technology to conduct in-depth mining and learning of different outbound call scenarios. It pre-screens and intercepts user numbers that the company is about to call out, helps companies actively respond to the call of relevant departments, strictly regulates outbound calling behaviors, effectively reduces harassment, and purifies the user's calling environment.

[0252] It can be seen from the above embodiments that the present disclosure calculates the theoretical recommended value corresponding to each call parameter type, and finally performs an overall weight adjustment to give a final total call recommended value, so that enterprises can reduce harassment to users based on the total call recommended value and their own business.

[0253] The present disclosure has the following beneficial effects:

[0254] 1. Utilize multiple data sources, multi-dimensional analysis, and multiple cross-validations. Instead of relying on a single data source, different recommended values ​​are given based on the different business needs of the enterprise to help the enterprise reduce disruption to users.

[0255] 2. Each recommended value relies on big data analysis and machine learning model training and is continuously improved.

[0256] 3. Starting from the user's perspective, in the spirit of respecting the user's perspective to the greatest extent, on the basis of machine learning, manual correction is introduced, and the adjustment of weight coefficients is added, especially the veto dimension, to protect users from being disturbed to the greatest extent.

[0257] The following are embodiments of the apparatus of the present disclosure, which can be used to execute embodiments of the method of the present disclosure.

[0258] Figure 4 is a block diagram of a number evaluation device according to an exemplary embodiment, which can be implemented as part or all of an electronic device through software, hardware, or a combination of both. Figure 4 As shown, the number of evaluation devices includes:

[0259] The first acquisition module 11 is used to obtain various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a value corresponding to the call parameter type;

[0260] The second acquisition module 12 is used to obtain the theoretical recommended values ​​corresponding to the various call parameter types of the to-be-called number and the weight value coefficients corresponding to the various call parameter types according to the various parameter values ​​corresponding to the various call parameter types and the preset recommendation table; the preset recommendation table includes: the corresponding relationship between the various parameter values ​​corresponding to the various call parameter types, the theoretical recommended values ​​of the various parameter values ​​corresponding to the various call parameter types and the weight value coefficients of the various parameter values ​​corresponding to the various call parameter types;

[0261] The third acquisition module 13 is used to acquire the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type.

[0262] In one embodiment, the apparatus further comprises:

[0263] The third acquisition module is used to acquire the call parameter information corresponding to each sample number before the first acquisition module 11 acquires the call parameter information of the number to be called;

[0264] A fourth acquisition module, used for training a support vector machine model according to the call parameter information corresponding to each sample number, and obtaining a total recommendation value corresponding to each call parameter type;

[0265] A fifth acquisition module, used to obtain a weight value coefficient corresponding to each of the call parameter types;

[0266] The sixth acquisition module is used to acquire the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types.

[0267] In one embodiment, the sixth acquisition module includes:

[0268] A first division submodule, used to divide all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type;

[0269] A first acquisition submodule is used to obtain a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type and the theoretical number of intervals in which the parameter value of the call parameter type is located;

[0270] A second division submodule, configured to divide all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types;

[0271] The second acquisition submodule is used to acquire the weight value coefficient corresponding to the parameter value of the call parameter type according to the first preset rule and the number of the de-weighting interval where the parameter value of the call parameter type is located.

[0272] In one embodiment, the third acquisition module 13 includes:

[0273] The fourth acquisition submodule is used to acquire a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type;

[0274] The fifth acquisition submodule is used to acquire a total weight value coefficient; the total weight value coefficient is the product of the weight value coefficients corresponding to each call parameter type;

[0275] The sixth acquisition submodule is used to obtain the product of the total theoretical recommendation value and the total weight value coefficient as the real-time total call recommendation value.

[0276] In one embodiment, the call parameter types include: business type, density of marketing calls made by the enterprise, complaints in the region where the called user is located, whether the called user itself has a negative mark, service mark produced by the called user, whether the called user has a good number and the degree of the good number, number activity of the called user, degree of connection between the called user and the enterprise's business industry, degree of connection between the called user and the enterprise's business itself, geographical relationship between the called user and the enterprise's business, degree of sensitivity of the called user to unfamiliar numbers, the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and registration status of rejected calls.

[0277] In one embodiment, the call parameter type is the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and the DNS registration status;

[0278] The first acquisition submodule includes:

[0279] A first determination subunit is used for, if the number is in the blacklist database of the enterprise itself, or the number is connected with the complaint agency, or the number is registered in the call rejection, then the corresponding theoretical interval is the first interval;

[0280] A second determination subunit is used for determining that if the number is not in the blacklist database of the enterprise itself, or the number has no contact with the complaint agency, or the number is not registered in the call rejection, the corresponding theoretical interval is the second interval;

[0281] The theoretical recommended value corresponding to the first interval is 0, and the theoretical recommended value corresponding to the second interval is the total recommended value corresponding to the call parameter type;

[0282] The third acquisition module 13 includes:

[0283] The determination submodule is used for determining that if the theoretical recommended value corresponding to the enterprise's own blacklist database, whether the called user has contact with the complaint agency, or the DNS registration status is 0, the real-time total call recommendation value of the to-be-called number is 0.

[0284] According to a third aspect of an embodiment of the present disclosure, there is provided a number evaluation device, including:

[0285] processor;

[0286] a memory for storing processor-executable instructions;

[0287] The processor is configured as follows:

[0288] Acquire various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a value corresponding to the call parameter type;

[0289] According to the parameter values ​​corresponding to the call parameter types and the preset recommendation table, the theoretical recommended values ​​corresponding to the call parameter types of the to-be-called number and the weight value coefficients corresponding to the call parameter types are obtained; the preset recommendation table includes: the corresponding relationship between the parameter values ​​corresponding to the call parameter types, the theoretical recommended values ​​of the parameter values ​​corresponding to the call parameter types and the weight value coefficients of the parameter values ​​corresponding to the call parameter types;

[0290] The real-time total call recommendation value of the to-be-called number is obtained according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type.

[0291] The above processor may also be configured to:

[0292] In one embodiment, before obtaining the call parameter information of the number to be called, the method further includes:

[0293] Obtain call parameter information corresponding to each sample number;

[0294] Training a support vector machine model according to the call parameter information corresponding to each sample number to obtain a total recommendation value corresponding to each call parameter type;

[0295] Obtaining weight value coefficients corresponding to each of the call parameter types;

[0296] The preset recommendation table is obtained at least according to the total recommendation values ​​corresponding to the various call parameter types and the weight value coefficients corresponding to the various call parameter types.

[0297] In one embodiment, the obtaining the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types includes:

[0298] Dividing all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type;

[0299] Obtaining a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located;

[0300] Dividing all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types;

[0301] The weight value coefficient corresponding to the parameter value of the call parameter type is obtained according to the first preset rule and the number of the de-weighting interval in which the parameter value of the call parameter type is located.

[0302] In one embodiment, the step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes:

[0303] Obtain a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type;

[0304] Obtaining a total weight coefficient; the total weight coefficient is the product of the weight coefficients corresponding to each call parameter type;

[0305] The product of the total theoretical recommendation value and the total weight value coefficient is obtained as the real-time total call recommendation value.

[0306] In one embodiment, the call parameter types include: business type, density of marketing calls made by the enterprise, complaints in the region where the called user is located, whether the called user itself has a negative mark, service mark produced by the called user, whether the called user has a good number and the degree of the good number, number activity of the called user, degree of connection between the called user and the enterprise's business industry, degree of connection between the called user and the enterprise's business itself, geographical relationship between the called user and the enterprise's business, degree of sensitivity of the called user to unfamiliar numbers, the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and registration status of rejected calls.

[0307] In one embodiment, the call parameter type is the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and the DNS registration status;

[0308] The obtaining, according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located, a theoretical recommended value corresponding to the parameter value of the call parameter type includes:

[0309] If the number is in the blacklist database of the enterprise itself, or the number is connected with the complaint agency, or the number is registered in the call rejection service, the corresponding theoretical interval is the first interval;

[0310] If the number is not in the blacklist of the enterprise, or the number has no connection with the complaint agency, or the number is not registered in the call rejection service, the corresponding theoretical interval is the second interval;

[0311] The theoretical recommended value corresponding to the first interval is 0, and the theoretical recommended value corresponding to the second interval is the total recommended value corresponding to the call parameter type;

[0312] The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes:

[0313] If the theoretical recommendation value corresponding to the enterprise's own blacklist database, whether the called user has contact with the complaint agency, or the DNS registration status is 0, the real-time total call recommendation value of the to-be-called number is 0.

[0314] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0315] Figure 5 1 is a block diagram of a device 80 for evaluating a number according to an exemplary embodiment, and the device is applicable to a terminal device. For example, the device 80 may be a computer or the like.

[0316] The device 80 may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output (I / O) interface 812 , a sensor component 814 , and a communication component 816 .

[0317] The processing component 802 generally controls the overall operation of the device 80, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0318] The memory 804 is configured to store various types of data to support operations on the device 80. Examples of such data include instructions for any application or method operating on the device 80, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0319] The power supply component 806 provides power to the various components of the device 80. The power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 80.

[0320] The multimedia component 808 includes a screen that provides an output interface between the device 80 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the device 80 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0321] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the device 80 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0322] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: home button, volume button, start button, and lock button.

[0323] The sensor assembly 814 includes one or more sensors for providing various aspects of the status assessment of the device 80. For example, the sensor assembly 814 can detect the open / closed state of the device 80, the relative positioning of components, such as the display and keypad of the device 80, and the sensor assembly 814 can also detect the position change of the device 80 or a component of the device 80, the presence or absence of user contact with the device 80, the orientation or acceleration / deceleration of the device 80, and the temperature change of the device 80. The sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0324] The communication component 816 is configured to facilitate wired or wireless communication between the device 80 and other devices. The device 80 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0325] In an exemplary embodiment, the device 80 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above methods.

[0326] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, and the instructions can be executed by the processor 820 of the device 80 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0327] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by the processor of the device 80, enables the device 80 to perform the above-mentioned number evaluation method, the method comprising:

[0328] Acquire various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a value corresponding to the call parameter type;

[0329] According to the parameter values ​​corresponding to the call parameter types and the preset recommendation table, the theoretical recommended values ​​corresponding to the call parameter types of the to-be-called number and the weight value coefficients corresponding to the call parameter types are obtained; the preset recommendation table includes: the corresponding relationship between the parameter values ​​corresponding to the call parameter types, the theoretical recommended values ​​of the parameter values ​​corresponding to the call parameter types and the weight value coefficients of the parameter values ​​corresponding to the call parameter types;

[0330] The real-time total call recommendation value of the to-be-called number is obtained according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type.

[0331] The above processor may also be configured to:

[0332] In one embodiment, before obtaining the call parameter information of the number to be called, the method further includes:

[0333] Obtain call parameter information corresponding to each sample number;

[0334] Training a support vector machine model according to the call parameter information corresponding to each sample number to obtain a total recommendation value corresponding to each call parameter type;

[0335] Obtaining weight value coefficients corresponding to each of the call parameter types;

[0336] The preset recommendation table is obtained at least according to the total recommendation values ​​corresponding to the various call parameter types and the weight value coefficients corresponding to the various call parameter types.

[0337] In one embodiment, the obtaining the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types includes:

[0338] Dividing all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type;

[0339] Obtaining a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located;

[0340] Dividing all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types;

[0341] The weight value coefficient corresponding to the parameter value of the call parameter type is obtained according to the first preset rule and the number of the de-weighting interval in which the parameter value of the call parameter type is located.

[0342] In one embodiment, the step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes:

[0343] Obtain a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type;

[0344] Obtaining a total weight coefficient; the total weight coefficient is the product of the weight coefficients corresponding to each call parameter type;

[0345] The product of the total theoretical recommendation value and the total weight value coefficient is obtained as the real-time total call recommendation value.

[0346] In one embodiment, the call parameter types include: business type, density of marketing calls made by the enterprise, complaints in the region where the called user is located, whether the called user itself has a negative mark, service mark produced by the called user, whether the called user has a good number and the degree of the good number, number activity of the called user, degree of connection between the called user and the enterprise's business industry, degree of connection between the called user and the enterprise's business itself, geographical relationship between the called user and the enterprise's business, degree of sensitivity of the called user to unfamiliar numbers, the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and registration status of rejected calls.

[0347] In one embodiment, the call parameter type is the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and the DNS registration status;

[0348] The obtaining, according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located, a theoretical recommended value corresponding to the parameter value of the call parameter type includes:

[0349] If the number is in the blacklist database of the enterprise itself, or the number is connected with the complaint agency, or the number is registered in the call rejection service, the corresponding theoretical interval is the first interval;

[0350] If the number is not in the blacklist of the enterprise, or the number has no connection with the complaint agency, or the number is not registered in the call rejection service, the corresponding theoretical interval is the second interval;

[0351] The theoretical recommended value corresponding to the first interval is 0, and the theoretical recommended value corresponding to the second interval is the total recommended value corresponding to the call parameter type;

[0352] The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes:

[0353] If the theoretical recommendation value corresponding to the enterprise's own blacklist database, whether the called user has contact with the complaint agency, or the DNS registration status is 0, the real-time total call recommendation value of the to-be-called number is 0.

[0354] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0355] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for evaluating a number, characterized in that: include: Obtain various call parameter information of the number to be called; The call parameter information includes: a call parameter type and a value corresponding to the call parameter type; Obtain theoretical recommended values ​​and weight value coefficients corresponding to each call parameter type of the number to be called according to each parameter value corresponding to each call parameter type of the number to be called and a preset recommendation table; the preset recommendation table includes: corresponding relationships between each parameter value corresponding to each call parameter type, theoretical recommended values ​​of each parameter value corresponding to each call parameter type, and weight value coefficients of each parameter value corresponding to each call parameter type; Obtaining a real-time total call recommendation value of the number to be called according to the theoretical recommendation value and weight value coefficient corresponding to each call parameter type of the number to be called; Before obtaining the call parameter information of the number to be called, the method further includes: Obtain call parameter information corresponding to each sample number; Training a support vector machine model according to the call parameter information corresponding to each sample number to obtain a total recommendation value corresponding to each call parameter type; Obtaining weight value coefficients corresponding to each of the call parameter types; Acquire the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types; The obtaining the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types includes: Dividing all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type; Obtaining a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located; Dividing all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types; Obtaining a weight value coefficient corresponding to the parameter value of the call parameter type according to a first preset rule and the number of weight reduction intervals in which the parameter value of the call parameter type is located; The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes: Obtain a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type; Obtaining a total weight coefficient; the total weight coefficient is the product of the weight coefficients corresponding to each call parameter type; The product of the total theoretical recommendation value and the total weight value coefficient is obtained as the real-time total call recommendation value.

2. The method according to claim 1, characterized in that The call parameter types include: business type, density of marketing calls made by the enterprise, complaints in the region where the called user is located, whether the called user has a negative mark, service mark produced by the called user, whether the called user has a good number and the degree of the good number, number activity of the called user, degree of connection between the called user and the enterprise's business industry, degree of connection between the called user and the enterprise's business itself, geographical relationship between the called user and the enterprise's business, degree of sensitivity of the called user to unfamiliar numbers, the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and registration status of do not call.

3. The method according to claim 2, characterized in that The call parameter types include the enterprise's own blacklist database, whether the called user has contact with the complaint agency, and the DNS registration status; The obtaining, according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located, a theoretical recommended value corresponding to the parameter value of the call parameter type includes: If the number is in the blacklist database of the enterprise itself, or the number is connected with the complaint agency, or the number is registered in the call rejection service, the corresponding theoretical interval is the first interval; If the number is not in the blacklist of the enterprise, or the number has no connection with the complaint agency, or the number is not registered in the call rejection service, the corresponding theoretical interval is the second interval; The theoretical recommended value corresponding to the first interval is 0, and the theoretical recommended value corresponding to the second interval is the total recommended value corresponding to the call parameter type; The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes: If the theoretical recommendation value corresponding to the enterprise's own blacklist database, whether the called user has contact with the complaint agency, or the DNS registration status is 0, the real-time total call recommendation value of the to-be-called number is 0.

4. A number evaluation device, characterized in that: include: The first acquisition module is used to obtain various call parameter information of the number to be called; The call parameter information includes: a call parameter type and a value corresponding to the call parameter type; The second acquisition module is used to obtain the theoretical recommended values ​​corresponding to the various call parameter types of the to-be-called number and the weight value coefficients corresponding to the various call parameter types according to the various parameter values ​​corresponding to the various call parameter types and the preset recommendation table; the preset recommendation table includes: the corresponding relationship between the various parameter values ​​corresponding to the various call parameter types, the theoretical recommended values ​​of the various parameter values ​​corresponding to the various call parameter types and the weight value coefficients of the various parameter values ​​corresponding to the various call parameter types; A third acquisition module is used to obtain the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type; Before acquiring the call parameter information of the number to be called, the first acquisition module is further used to: Obtain call parameter information corresponding to each sample number; Training a support vector machine model according to the call parameter information corresponding to each sample number to obtain a total recommendation value corresponding to each call parameter type; Obtaining weight value coefficients corresponding to each of the call parameter types; Acquire the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types; The obtaining the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types includes: Dividing all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type; Obtaining a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located; Dividing all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types; Obtaining a weight value coefficient corresponding to the parameter value of the call parameter type according to a first preset rule and the number of weight reduction intervals in which the parameter value of the call parameter type is located; The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes: Obtain a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type; Obtaining a total weight coefficient; the total weight coefficient is the product of the weight coefficients corresponding to each call parameter type; The product of the total theoretical recommendation value and the total weight value coefficient is obtained as the real-time total call recommendation value.

5. A number evaluation device, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: Acquire various call parameter information of the number to be called; the call parameter information includes: a call parameter type and a value corresponding to the call parameter type; According to the parameter values ​​corresponding to the call parameter types and the preset recommendation table, the theoretical recommended values ​​corresponding to the call parameter types of the to-be-called number and the weight value coefficients corresponding to the call parameter types are obtained; the preset recommendation table includes: the corresponding relationship between the parameter values ​​corresponding to the call parameter types, the theoretical recommended values ​​of the parameter values ​​corresponding to the call parameter types and the weight value coefficients of the parameter values ​​corresponding to the call parameter types; Obtaining a real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type; Before obtaining the call parameter information of the number to be called, the method further includes: Obtain call parameter information corresponding to each sample number; Training a support vector machine model according to the call parameter information corresponding to each sample number to obtain a total recommendation value corresponding to each call parameter type; Obtaining weight value coefficients corresponding to each of the call parameter types; Acquire the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types; The obtaining the preset recommendation table at least according to the total recommendation value corresponding to each of the call parameter types and the weight value coefficient corresponding to each of the call parameter types includes: Dividing all parameter values ​​corresponding to each of the call parameter types into a plurality of theoretical intervals, each of the theoretical intervals including a plurality of parameter values ​​corresponding to the call parameter type; Obtaining a theoretical recommended value corresponding to the parameter value of the call parameter type according to the total recommended value corresponding to the call parameter type, the total theoretical number of intervals corresponding to the call parameter type, and the theoretical number of intervals in which the parameter value of the call parameter type is located; Dividing all the parameter values ​​corresponding to each of the call parameter types into a plurality of de-weighting intervals, each of the de-weighting intervals including a plurality of parameter values ​​corresponding to the call parameter types; Obtaining a weight value coefficient corresponding to the parameter value of the call parameter type according to a first preset rule and the number of weight reduction intervals in which the parameter value of the call parameter type is located; The step of obtaining the real-time total call recommendation value of the to-be-called number according to the theoretical recommendation value corresponding to each call parameter type and the weight value coefficient corresponding to each call parameter type includes: Obtain a total theoretical recommended value; the total theoretical recommended value is the sum of the theoretical recommended values ​​corresponding to each call parameter type; Obtaining a total weight coefficient; the total weight coefficient is the product of the weight coefficients corresponding to each call parameter type; The product of the total theoretical recommendation value and the total weight value coefficient is obtained as the real-time total call recommendation value.

6. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instruction is executed by a processor, the steps of the method described in any one of claims 1 to 3 are implemented.

Images