CN102752462B - Method and system for recommending telecommunications service - Google Patents

Abstract

The invention provides a method and a system for recommending telecommunications service. The method comprises the following steps of: determining a drainage density factor of a user and each contact of the user according to a communication record of the user, wherein the drainage density factor is used for representing communication frequency of the user and each contact; obtaining service information of each contact, wherein the service information is used for representing a service order situation and/or service use condition of the contact; and recommending the service to the user according to the drainage density factor of the user and each contact and the service information of each contact. The system comprises a drainage density module, a service information module and a service recommending module. The method has the beneficial effects that the success rate of recommending the telecommunications service to the user is improved according to the drainage density of the user and the contact and the preference of the user contact to the telecommunications service.

Description

Telecommunication service recommendation method and system

Technical field

The present invention relates to wireless communication technology, relate in particular to a kind of telecommunication service recommendation method and system.

Background technology

Present stage China's telecommunication service faces the market pressure of homogeneity competition, makes to develop new user more and more difficult; Telecom operators just progressively attract user's crude management from the competitive methods such as Price-dependent war in the past, to Focus service, fine management, change, and the numerous and confused data digging system of building oneself is assisted and carried out precision marketing.

In prior art, analysis and the excavation of the data such as the basic document based on telecommunication user, geographical position, therefrom find service operation rule, and then carry out telecommunication service and recommend, yet this method may cause recommending the success rate of telecommunication service not high to user.

Summary of the invention

The invention provides a kind of telecommunication service recommendation method and system, to improve the success rate of user being recommended to telecommunication service.

For realizing object of the present invention, on the one hand, the invention provides a kind of telecommunication service recommendation method, comprising:

According to user's communications records, determine each contact person's of described user and described user the density factor, the described density factor is for representing described user and the frequent degree of communicating by letter of each contact person;

Obtain described each contact person's business information, described business information is for representing described contact person's service order situation and/or business service condition;

According to described user and described each contact person's the density factor and the described described business information respectively contacting, for described user recommends business.

On the other hand, the present invention also provides a kind of telecommunication service commending system, comprising:

Density module, for according to user's communications records, determines each contact person's of described user and described user the density factor, and the described density factor is for representing described user and the frequent degree of communicating by letter of each contact person;

Business information module, for obtaining described each contact person's business information, described business information is for representing described contact person's service order situation and/or business service condition;

Business recommended module, for according to described user and described each contact person's the density factor and the described described business information respectively contacting, for described user recommends business.

Beneficial effect of the present invention is, the fancy grade to telecommunication service according to user and its contact person's density degree and user contact person improves the success rate that this user is recommended to telecommunication service.

Accompanying drawing explanation

Fig. 1 is the flow chart of the embodiment of telecommunication service recommendation method of the present invention;

Fig. 2 is the flow chart of the another embodiment of telecommunication service recommendation method of the present invention;

Fig. 3 is the flow chart of the another embodiment of telecommunication service recommendation method of the present invention;

Fig. 4 is the flow chart of the another embodiment of telecommunication service recommendation method of the present invention;

Fig. 5 is the structural representation of the embodiment of telecommunication service commending system of the present invention.

Embodiment

Below in conjunction with accompanying drawing and embodiment, the present invention will be further described.

Fig. 1 is the flow chart of the embodiment of telecommunication service recommendation method of the present invention, and the present embodiment specifically can carry by implementing one or more network elements that telecommunication service recommends, or is carried by network element independently; Can realize with software or example, in hardware, also can realize in the mode of software and combination of hardware, for example: can be implemented by telecommunication service commending system.

The concrete steps of the telecommunication service recommendation method of the present embodiment are as follows:

S10: according to user's communications records, determine each contact person's of user and this user the density factor, this density factor is for representing the frequent degree of communicating by letter of this each contact person of user and its.

Above-mentioned communications records can include but not limited to following any one or much information: voice call recorded information, visual telephone recorded information, note recorded information, multimedia message recorded information and email record information.

The density factor is for representing the frequent degree of communicating by letter of each contact person of user and its, user communicates by letter with its contact person their close relation is frequently described, influencing each other can be larger, common topic is more, the probability with identical hobby is very high, this user can to need to order the probability of same electrical communication service also very high.

Therefore, can, according to the frequent degree of each shown contact person of user's above-mentioned various message registrations and telex network, determine each contact person's of this user the density factor.

S11: obtain this each contact person's of user business information, this business information is for representing each contact person's service order situation and/or business service condition.

Above-mentioned business information can include but not limited to following any one or much information: order business type, service order time started, business usage frequency and business complains records.

By each contact person's of user business information, can obtain the hobby place of each contact person to each telecommunication service.

S12: according to this user and its each contact person's the density factor and each contact person's business information, for this user recommends business.

Because user communicates by letter their close relation is frequently described with its contact person, influencing each other can be larger, and common topic is more, and the probability with identical hobby is very high, this user may to order the probability of same electrical communication service also very high.

Therefore, can know the demand of user to telecommunication service by inference, thereby recommend telecommunication service to this user according to the above-mentioned density factor and business information, can precisely recommend.

In the present embodiment, by the degree that influences each other between the known user of the above-mentioned density factor and its contact person.By each contact person's of user business information, can obtain the hobby place of each contact person to each telecommunication service.In conjunction with the above-mentioned density factor and business information, can know the demand of user to telecommunication service by inference, thereby recommend telecommunication service to this user, improve the success rate that this user is recommended to telecommunication service.

Fig. 2 is the flow chart of the another embodiment of telecommunication service recommendation method of the present invention, as shown in Figure 2, the present embodiment is on the basis of above-described embodiment, S12 is according to this user and its each contact person's the density factor and each contact person's business information, for this user business of recommending specifically comprises the following steps:

S121: judging whether this user has ordered k class business, is to carry out S126; No, carry out S122;

S122: according to Recommend (S _i, k)=F ₄(R _i, P _ik, N _ik), obtain this user's i contact person S _ito this user, recommend the recommendation degree of k class business.

Wherein, S _ibe only one in order to represent this user contact person's sign, in order to shine upon this contact person's name and whole telephone numbers corresponding to this name; R _ifor this user and its i contact person S _ithe density factor; P _iki the contact person S for this user _ithe positive feedback factor to k class business, represents the degree of this contact person to k class business positive feedback, i.e. P _ikshow more greatly contact person S _ihigher to k class business fancy grade; N _iki the contact person S for this user _ipassive feedback factor to k class business, represents passiveness feedback degree, the i.e. N of this contact person to k class business _ikshow more greatly contact person S _imore passive to the business recommended attitude of k class; F ₄represent Recommend (S _i, k) and R _i, P _ikbe directly proportional, with N _ikrelation is inversely proportional to.

F ₄represented to affect telecommunication service recommendation degree Recommend (S _i, principal element R k) _i, P _ikand N _ikwith recommendation degree Recommend (S _i, relation k), Recommend (S _i, k) with user and its i contact person S _idensity factor R _ibe directly proportional, with user's i contact person S _ipositive feedback factor P to k class business _ikbe directly proportional, with user's i contact person S _ipassive feedback factor N to k class business _ikbe inversely proportional to.Illustrate between contact person and user, contact closer, this contact person order and the recommendation degree of the business of hobby higher.

S123: obtain each contact person of this user and recommend the recommendation degree Recommend (S of k class business to it _i, the k) maximum in (0≤i≤N).

Wherein, N represents contact person's sum of this user;

S124: judge whether this maximum is greater than default recommendation degree threshold value, no, carry out S126; To carry out S125;

S125: recommend k class business to this user;

S126: do not recommend k class business to this user, flow process finishes.

First judge whether this user has ordered k class business, if do not ordered, relatively each contact person of user recommends the maximum of recommendation degree and the size of default recommendation degree threshold value of k class business to this user, if the maximum of recommendation degree is less than default recommendation degree threshold value, illustrate that neither one contact person recommends the recommendation degree of k class business to meet recommendation requirement to this user, this business is not suitable for recommending to this user; If the maximum of recommendation degree is greater than default recommendation degree threshold value, illustrating in whole contact persons of this user has at least a contact person to recommend the recommendation degree of k class business to meet recommendation requirement to this user, this business is applicable to recommending to this user, to this user, recommends this business.

In the present embodiment, according to user contact person, to the positive feedback factor of business and passive feedback factor, can know the fancy grade of this contact person to this business, and obtain the recommendation degree of this contact person to this business according to the density factor between this contact person and this user.If have at least a contact person to recommend the recommendation degree of this business to meet recommendation requirement to this user in whole contact persons of this user, this business is applicable to recommending to this user.Thereby to this user, recommend telecommunication service, can precisely recommend, improve the success rate that user is recommended to telecommunication service.

Fig. 3 is the flow chart of the another embodiment of telecommunication service recommendation method of the present invention, as shown in Figure 3, the present embodiment is on the basis of above-described embodiment, and S10, according to user's communications records, determines that each contact person's of user and this user the density factor specifically comprises the following steps:

S101: respectively in obtaining communication record this user as total stoichiometric number of calling party with as total stoichiometric number of callee;

Above-mentioned communications records can include but not limited to following one or more information: voice call, visual telephone, note, multimedia message, Email etc.

Above-mentioned user as total stoichiometric number of calling party refer to this user initiatively set up with its contact person between total stoichiometric number of communicating by letter, above-mentioned user as total stoichiometric number of callee refer to this user's contact person initiatively set up with this user between total stoichiometric number of communicating by letter.

Above-mentioned total stoichiometric number can be definite according to different communication modes, and for example, voice call and visual telephone record can be usingd talk times as total stoichiometric number, or using the duration of call as total stoichiometric number; Note and multimedia message record can be usingd number as total stoichiometric number, or using length as total stoichiometric number.

S102: according to $R_i=F_1\left(\underset{j}{\mathrm{\Σ}}(\frac{c_{\mathrm{ij}1}\×{\mathrm{\α}}_{j1}+c_{\mathrm{ij}2}\×{\mathrm{\α}}_{j2}}{C_j}\×{\mathrm{\λ}}_j)\right),$ Determine this user and its i contact person's the density factor;

Wherein, R _ii the contact person S that represents described user and described user _ithe density factor, c _ij1represent the total stoichiometric number as calling party of described user in the j class communications records with described i contact person, c _ij2represent the total stoichiometric number as callee of described user in the j class communications records with described i contact person, α _j1the caller weight factor that represents j class communications records, α _j2the called weight factor that represents j class communications records, and α _j1+ α _j2=1,0≤α _j1≤ 1,0≤α _j2≤ 1, α _j1with α _j2can be according to c _ij1, c _ij1the ratio of carrying out total stoichiometric number that j class communicates by letter that accounts for respectively this user and its i contact person is determined, or is artificially set; C _jthe total stoichiometric number that represents described user's j class communications records, and C _i> 0; If C _j=0, λ _jthe weight factor that represents j class communications records, and 0≤λ≤1, F ₁represent R _iwith c _ij1, c _ij2, α _j1, α _j2and λ _jbe directly proportional, with C _jbe inversely proportional to, λ _jthe ratio that can account for total stoichiometric number of these user's all categories communications records according to total stoichiometric number of this user's j class communications records is determined, or is artificially set according to traffic category.

The present embodiment is according to the communications records between user and its contact person, obtain the density factor between this user and its contact person, represent the frequent degree of communicating by letter of this each contact person of user and its, the density factor more illustrates that this user and its contact person contact closer, between them, interactional degree is just higher, and they more likely have identical business demand.Therefore, determine user and its each contact person's the density factor, to determine accordingly the recommendation degree of telecommunication service, thereby improve the success rate that this user is recommended to telecommunication service.

Fig. 4 is the flow chart of the another embodiment of telecommunication service recommendation method of the present invention, and as shown in Figure 4, on the basis of the embodiment shown in Fig. 1 and Fig. 2, the business information that S11 obtains this each contact person of user specifically comprises the following steps:

S111: according to P _ik=F ₂(T _ik, B _ik), the positive feedback factor of i the contact person who obtains this user to k class business.

Wherein, P _ikrepresent the positive feedback factor, T _iki the contact person S that represents this user _iorder the duration of k class business, definite by current system time and k class service order time started; B _iki the contact person S that represents this user _iuse the frequency of k class business; F ₁represent P _ikwith T _ik, B _ikproportional.Be contact person S _ithe time of ordering k class business is longer, uses the more frequent of k class business, this positive feedback factor P _iklarger, show contact person S _idesirability to k class business is higher.

S112: according to the passive feedback factor of i the contact person who obtains this user to k class business;

Wherein, N _ikrepresent passive feedback factor, DN _ikxthis user's i contact person S _ithe x class of k class business is complained to number of times, ω _xrepresenting the penalty coefficient that x class is complained, is preset value, F ₃represent N _ikwith DN _ikxand ω _xbe directly proportional.Passive feedback factor N _ikeach class that is k class business is complained the summation of number of times and such complaint penalty coefficient product, shows contact person S _icomplaint number of times to k class business is more, and this contact person is more passive to the recommendation attitude of such business.

It should be noted that, step S111 and S112 in the present embodiment do not have sequencing.

In the present embodiment, according to the duration of contact person's order business and business usage frequency, can obtain the positive feedback factor of this contact person to this business, this factor shows that more greatly this contact person is higher to the fancy grade of this business; The penalty coefficient that the number of times of according to contact person, each class of business being complained again and each default class are complained, can obtain the passive feedback factor of this contact person to this business, this factor shows that more greatly this contact person is lower to the fancy grade of this business, to user, recommends the probability of this business lower.By knowing the fancy grade of user contact person to each business, to obtain the demand of user to all kinds of business, realize the accurate recommendation of business.

It should be noted that, in above-described embodiment, the business of using due to user and contact thereof may adhere to different operator's operations separately, therefore the present invention is in specific implementation process, when the order of statistical service and service condition, can only counting user and its contact person in same carrier network between communications records and service order, service condition.

Fig. 5 is the structural representation of the embodiment of telecommunication service commending system of the present invention, and as shown in Figure 5, the telecommunication service commending system of the present embodiment comprises: density module 11, business information module 12 and business recommended module 13;

Density module 11, for according to user's communications records, determines each contact person's of described user and described user the density factor, and the described density factor is for representing described user and the frequent degree of communicating by letter of each contact person;

Business information module 12, for obtaining described each contact person's business information, described business information is for representing described contact person's service order situation and/or business service condition;

Business recommended module 13, for according to described user and described each contact person's the density factor and the described described business information respectively contacting, for described user recommends business.

Optionally, business recommended module 13 is specifically for according to Recommend (S _i, k)=F ₄(R _i, P _ik, N _ik), obtain this user's i contact person S _ito this user, recommend the recommendation degree of k class business; Wherein, R _ii the contact person S for this user and this user _ithe density factor; P _iki the contact person S for this user _ithe positive feedback factor to k class business, represents the degree of this contact person to k class business positive feedback; N _iki the contact person S for this user _ipassive feedback factor to k class business, represents the positive feedback degree of this contact person to k class business; F ₄represent Recommend (S _i, k) and R _i, P _ikbe directly proportional, with N _ikrelation is inversely proportional to; Obtain each contact person of this user and to this user, recommend the recommendation degree Recommend (S of k class business _i, the k) maximum in (0≤i≤N), wherein N represents contact person's sum of this user; Judging whether this maximum is greater than default recommendation degree threshold value, is to this user, to recommend k class business; No, to this user, do not recommend k class business.

As a kind of embodiment, density module 11 is specifically for obtaining respectively user described in described communications records as total stoichiometric number of calling party with as total stoichiometric number of callee; According to determine described user with i contact person's the density factor; Wherein, R _ii the contact person S that represents described user and described user _ithe density factor, c _ij1represent the total stoichiometric number as calling party of described user in the j class communications records with described i contact person, c _ij2represent the total stoichiometric number as callee of described user in the j class communications records with described i contact person, α _j1the caller weight factor that represents j class communications records, α _j2the called weight factor that represents j class communications records, and α _j1+ α _j2=1,0≤α _j1≤ 1,0≤α _j2≤ 1; F ₁represent R _iwith c _ij1, c _ij2, α _j1, α _j2and λ _jbe directly proportional, with C _jbe inversely proportional to; C _jthe total stoichiometric number that represents described user's j class communications records, and C _j> 0; If C _j=0, λ _jthe weight factor that represents j class communications records, and 0≤λ≤1.

Preferably, business information module 12 is specifically for according to P _k=F ₂(T _ik, B _ik), the positive feedback factor of i the contact person who obtains described user to k class business; Wherein, P _ikrepresent the described positive feedback factor, T _ikrepresent that described user's i contact person orders the duration of k class business, definite by current system time and k class service order time started; B _iki the contact person who represents described user uses the frequency of k class business, F ₂represent P _ikwith T _ik, B _ikproportional; According to the passive feedback factor of i the contact person who obtains described user to k class business; Wherein, N _ikrepresent described passive feedback factor, DN _ikxdescribed user's i contact person complains number of times, ω to the x class of k class business _xrepresent the penalty coefficient that x class is complained, F ₃represent N _ikwith DN _ikxand ω _xbe directly proportional.

Telecommunication service commending system embodiment provided by the invention, corresponding with the embodiment of telecommunication service recommendation method provided by the invention, actuating equipment for telecommunication service recommendation method provided by the invention, the operation that in its concrete structure and structure, each several part is carried out can, referring to embodiment of the method, not repeat them here.

Telecommunication service commending system provided by the invention, by the degree that influences each other between density factor acquisition user and its contact person.By each contact person's of user business information, obtain the hobby place of each contact person to each telecommunication service.In conjunction with the above-mentioned density factor and business information, can know the demand of user to telecommunication service by inference, thereby recommend telecommunication service to this user, improve the success rate that this user is recommended to telecommunication service.

One of ordinary skill in the art will appreciate that: all or part of step that realizes above-mentioned each embodiment of the method can complete by the relevant hardware of program command.Aforesaid program can be stored in a computer read/write memory medium.This program, when carrying out, is carried out the step that comprises above-mentioned each embodiment of the method; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CDs.

Finally it should be noted that: each embodiment, only in order to technical scheme of the present invention to be described, is not intended to limit above; Although the present invention is had been described in detail with reference to aforementioned each embodiment, those of ordinary skill in the art is to be understood that: its technical scheme that still can record aforementioned each embodiment is modified, or some or all of technical characterictic is wherein equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.

Claims (6)

1. a telecommunication service recommendation method, is characterized in that, comprising:

According to user's communications records, determine each contact person's of described user and described user the density factor, the described density factor is for representing described user and the frequent degree of communicating by letter of its each contact person;

Obtain described each contact person's business information, described business information is for representing described contact person's service order situation and/or business service condition;

According to described user and described each contact person's the density factor and described each contact person's described business information, for described user recommends business;

Described according to user's communications records, determine and comprise each contact person's of described user and described user the density factor:

Obtain respectively user described in described communications records as total stoichiometric number of calling party with as total stoichiometric number of callee;

According to $R_i=F_1\left(\underset{j}{\mathrm{\Σ}}(\frac{c_{\mathrm{ij}1}\×{\mathrm{\α}}_{j1}+c_{\mathrm{ij}2}\×{\mathrm{\α}}_{j2}}{C_j}\×{\mathrm{\λ}}_j)\right),$ Determine described user and i contact person's the density factor;

Wherein, R _ii the contact person S that represents described user and described user _ithe density factor, c _ij1represent the total stoichiometric number as calling party of described user in the j class communications records with described i contact person, c _ij2represent the total stoichiometric number as callee of described user in the j class communications records with described i contact person, α _j1the caller weight factor that represents j class communications records, α _j2the called weight factor that represents j class communications records, and α _j1+ α _j2=1,0≤α _j1≤ 1,0≤α _j2≤ 1; C _jthe total stoichiometric number that represents described user's j class communications records, and C _j>0; If C _j=0, $\frac{c_{\mathrm{ij}1}\×{\mathrm{\α}}_{j1}+c_{\mathrm{ij}2}\×{\mathrm{\α}}_{j2}}{C_j}=0;$ λ _jthe weight factor that represents j class communications records, and $\underset{j}{\mathrm{\Σ}}{\mathrm{\λ}}_j=1,$ 0≤λ≤1; F ₁represent R _iwith c _ij1, c _ij2, α _j1, α _j2and λ _jbe directly proportional, with C _jbe inversely proportional to.

2. method according to claim 1, is characterized in that, described according to described user and described each contact person's the density factor and described each contact person's described business information, for described user recommends business, comprising:

Judging whether described user has ordered k class business, is to described user, not recommend k class business; No, according to Recommend (S _i, k)=F ₄(R _i, P _ik, N _ik), obtain described user's i contact person S _ito described user, recommend the recommendation degree of k class business; Wherein, R _ii the contact person S for described user and described user _ithe density factor; Described P _iki the contact person S for described user _ithe positive feedback factor to k class business, represents the degree of described contact person to k class business positive feedback; Described N _iki the contact person S for described user _ipassive feedback factor to k class business, represents the positive feedback degree of described contact person to k class business; F ₄represent Recommend (S _i, k) and R _i, P _ikbe directly proportional, with N _ikrelation is inversely proportional to;

Obtain each contact person of described user and to described user, recommend the recommendation degree Recommend (S of k class business _i, k), the maximum in 0≤i≤N, wherein N represents contact person's sum of described user;

Judge whether described maximum is greater than default recommendation degree threshold value, no, to described user, do not recommend k class business; To described user, to recommend k class business.

3. method according to claim 2, is characterized in that, described in obtain described each contact person's business information, described business information, for representing described contact person's service order situation and/or business service condition, specifically comprises:

According to P _ik=F ₂(T _ik, B _ik), the positive feedback factor of i the contact person who obtains described user to k class business;

Wherein, P _ikrepresent the described positive feedback factor, T _ikrepresent that described user's i contact person orders the duration of k class business, definite by current system time and k class service order time started; B _iki the contact person who represents described user uses the frequency of k class business, F ₂represent P _ikwith T _ik, B _ikproportional;

According to the passive feedback factor of i the contact person who obtains described user to k class business;

Wherein, N _ikrepresent described passive feedback factor, DN _ikxdescribed user's i contact person complains number of times, ω to the x class of k class business _xrepresent the penalty coefficient that x class is complained, F ₃represent N _ikwith DN _ikxand ω _xbe directly proportional.

4. a telecommunication service commending system, is characterized in that, comprising:

Density module, for according to user's communications records, determines each contact person's of described user and described user the density factor, and the described density factor is for representing described user and the frequent degree of communicating by letter of its each contact person;

Business information module, for obtaining described each contact person's business information, described business information is for representing described contact person's service order situation and/or business service condition;

Business recommended module, for according to described user and described each contact person's the density factor and described each contact person's described business information, for described user recommends business;

Described density module specifically for:

Obtain respectively user described in described communications records as total stoichiometric number of calling party with as total stoichiometric number of callee;

According to $R_i=F_1\left(\underset{j}{\mathrm{\Σ}}(\frac{c_{\mathrm{ij}1}\×{\mathrm{\α}}_{j1}+c_{\mathrm{ij}2}\×{\mathrm{\α}}_{j2}}{C_j}\×{\mathrm{\λ}}_j)\right),$ Determine described user and i contact person's the density factor;

Wherein, R _ii the contact person S that represents described user and described user _ithe density factor, c _ij1represent the total stoichiometric number as calling party of described user in the j class communications records with described i contact person, c _ij2represent the total stoichiometric number as callee of described user in the j class communications records with described i contact person, α _j1the caller weight factor that represents j class communications records, α _j2the called weight factor that represents j class communications records, and α _j1+ α _j2=1,0≤α _j1≤ 1,0≤α _j2≤ 1; C _jthe total stoichiometric number that represents described user's j class communications records, and C _j>0; If C _j=0, $\frac{c_{\mathrm{ij}1}\×{\mathrm{\α}}_{j1}+c_{\mathrm{ij}2}\×{\mathrm{\α}}_{j2}}{C_j}=0;$ λ _jthe weight factor that represents j class communications records, and $\underset{j}{\mathrm{\Σ}}{\mathrm{\λ}}_j=1,$ 0≤λ≤1; F ₁represent R _iwith c _ij1, c _ij2, α _j1, α _j2and λ _jbe directly proportional, with C _jbe inversely proportional to.

5. system according to claim 4, is characterized in that, described business recommended module specifically for:

Judging whether described user has ordered k class business, is to described user, not recommend k class business; No, according to Recommend (S _i, k)=F ₄(R _i, P _ik, N _ik), obtain described user's i contact person S _ito described user, recommend the recommendation degree of k class business; Wherein, R _ii the contact person S for described user and described user _ithe density factor; Described P _iki the contact person S for described user _ithe positive feedback factor to k class business, represents the degree of described contact person to k class business positive feedback; Described N _iki the contact person S for described user _ipassive feedback factor to k class business, represents the positive feedback degree of described contact person to k class business; F ₄represent Recommend (S _i, k) and R _i, P _ikbe directly proportional, with N _ikrelation is inversely proportional to;

Obtain each contact person of described user and to described user, recommend the recommendation degree Recommend (S of k class business _i, k), the maximum in 0≤i≤N, wherein N represents contact person's sum of described user;

Judge whether described maximum is greater than default recommendation degree threshold value, no, to described user, do not recommend k class business; To described user, to recommend k class business.

6. system according to claim 5, is characterized in that, described business information module specifically for:

According to P _ik=F ₂(T _ik, B _ik), the positive feedback factor of i the contact person who obtains described user to k class business;

Wherein, P _ikrepresent the described positive feedback factor, T _ikrepresent that described user's i contact person orders the duration of k class business, definite by current system time and k class service order time started; B _iki the contact person who represents described user uses the frequency of k class business, F ₂represent P _ikwith T _ik, B _ikproportional;

According to the passive feedback factor of i the contact person who obtains described user to k class business;

Wherein, N _ikrepresent described passive feedback factor, DN _ikxdescribed user's i contact person complains number of times, ω to the x class of k class business _xrepresent the penalty coefficient that x class is complained, F ₃represent N _ikwith DN _ikxand ω _xbe directly proportional.