Marriage object recommendation method and apparatus

A recommendation method and object technology, applied in the field of data processing, can solve problems such as the inability to determine a marriage partner

Active Publication Date: 2018-05-25
TENCENT TECH (SHENZHEN) CO LTD
7 Cites 10 Cited by

AI-Extracted Technical Summary

Problems solved by technology

With the increase of registered users on the platform, the number of marriage and love objects stored in the database has increased sharply, and the number of eligible marriage and love objects screened out according to the bas...
View more

Method used

In the embodiment of the present application, the application server 20 can filter and obtain candidate marriage and love objects according to the basic requirements of target users for marriage and love objects, and further obtain the credit points of target users and each candidate marriage and love objects from the credit database 40, and determine The credit score matching degree of the target user and each of the candidate marriage and love objects is used as the first matching degree, and the ranking results of each of the candidate marriage and love objects on the marriage and love object recommendation page are determined according to the first matching degree, and then sent to the application client In the recommended page of marriage and love objects sent, highly matching marriage and love partners can be ranked first, and target users can first check marriage and love partners with a higher degree of matching with their own credit scores, which improves the probability of target users quickly finding their favorite marriage and love partners.
In the method provided by the embodiment of the present application, after obtaining the candidate marriage and love objects according to the target user's basic requirements for the marriage and love objects, further obtain the credit points of the target users and each candidate marriage and love objects, and determine the relationship between the target user and each described The credit score matching degree of the candidate marriage and love object is used as the first matching degree, and the ranking result of each candidate marriag...
View more

Abstract

Embodiments of the invention provide a marriage object recommendation method and apparatus. According to the method, screening is performed according to basic requirement conditions of a target user to a marriage object to obtain candidate marriage objects; further, credit scores of the target user and the candidate marriage objects are obtained; a credit score matching degree of the target user and each candidate marriage object is determined as a first matching degree; according to the first matching degree, a sorting result of the candidate marriage objects in a marriage object recommendation page is determined; in the marriage object recommendation page sent to the user, the marriage objects with high matching degrees can be sorted in the front; and the target user can preferentially look up the marriage object with higher degree of matching with the own credit score, so that the probability of quickly finding the favorite marriage object by the target user is increased.

Application Domain

Data processing applicationsSpecial data processing applications

Technology Topic

Data miningData science +1

Image

  • Marriage object recommendation method and apparatus
  • Marriage object recommendation method and apparatus
  • Marriage object recommendation method and apparatus

Examples

  • Experimental program(1)

Example Embodiment

[0029] The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of this application.
[0030] figure 1 The implementation system architecture diagram of the method for recommending marriage and love objects provided by this embodiment of the application, refer to figure 1 The system architecture may include: an application client 10, an application server 20, an attribute information database 30, and a credit database 40. Among them, the attribute information database can also be called the marriage and love object database.
[0031] The application client 10 may be loaded on a user device such as a smart phone, a tablet computer, or a notebook computer, and communicate with the application server 20. The user can access the application server 20 through the application client 10 and request to display the marriage partner recommendation page.
[0032] Optionally, the application client 10 may be a separately developed client that cooperates with the application server 20; the user may download the application client locally to the user device through the application market (APP STORE) or the official website of the application. The application client can also exist in the form of a browser.
[0033] The application server 20 is a service device in a server cluster to which the network application belongs, and is set up on the network side. In the embodiment of the present application, the application server 20 is mainly used for data processing, displaying the information of the marriage partner recommendation page according to the request initiated by the client, determining the candidate marriage partner to be recommended, and the sort order of each candidate marriage partner.
[0034] The application server 20 belongs to a device in the marriage partner recommendation platform.
[0035] The attribute information database 30 is a database to which the network application belongs, and records attribute information of users registered on the marriage and love object recommendation platform and basic requirements of the users for the marriage and love objects. Among them, the attribute information includes information such as the user's age, occupation, income, and residence. The basic requirements are as follows: age over 25, income over 10,000, and residence in Beijing, etc. The application server 20 can read the target user’s basic requirements for the marriage partner from the attribute information database when the target user requests to display the marriage partner recommendation page, and read the stored attribute information of each marriage partner, and filter them to meet the basic requirements. Candidate marriage partner.
[0036] The attribute information database 30 also belongs to equipment in the marriage and love object recommendation platform.
[0037] The credit database 40 is a database that records the credit score of each user. The user’s credit score refers to the score that represents the user’s creditworthiness measured based on the user’s historical time period of financial, online social and other behavioral data; the user’s credit score can reflect the probability that the user will repay on time in a period of time in the future , Or the probability of debt default, etc. The higher the user's credit score, the higher the user's credit level.
[0038] The user's credit score can be trained to predict the user's credit model, and the user's behavioral data in the dimensions of finance and social networking can be imported into the model, and the user's credit score can be calculated.
[0039] The credit database 40 can be independent of network applications. The application server 20 can obtain the credit score of each user through the open interface of the credit database 40; that is, the application server 20 can obtain the credit score of each user calculated in the credit database 40 , There is no need to perform specific calculation of the credit score of each user; optionally, the credit investigation database 40 may be a bank credit investigation database or another credit investigation database that is open to query the user's credit score.
[0040] Optionally, the credit database 40 may also be a database to which the network application belongs. The application server 20 may adjust the user's basic credit score according to the historical communication behavior data between the user and the marriage partner, and store the adjusted credit score in Credit information database 40. At the same time, the application server 20 can retrieve the user's credit score from the credit database 40.
[0041] In this embodiment of the application, the application server 20 can screen and obtain candidate marriage partners according to the basic requirements of the target user for the marriage partner, and further obtain the target user and the credit scores of each candidate marriage partner from the credit database 40 to determine the target user and The credit score matching degree of each candidate marriage partner is used as the first matching degree, and the ranking result of each candidate marriage partner on the marriage partner recommendation page is determined according to the first matching degree, and then the marriage partner is sent to the application client. In the object recommendation page, the marriage partner with a high degree of matching can be ranked first, and the target user can first check the marriage partner with a higher degree of matching his credit score, which improves the probability that the target user can quickly find the favorite marriage partner.
[0042] based on figure 1 System architecture shown, figure 2 Shows the signaling flow of the method for recommending a marriage partner provided by the embodiment of the present application, refer to figure 2 , The process can include:
[0043] Step S10: The application client sends to the application server a request to display the recommended page of the marriage partner, and the request includes the basic requirements of the target user for the marriage partner;
[0044] Among them, the basic requirements include but are not limited to: age conditions, income conditions, occupational conditions, residential conditions, etc.
[0045] Optionally, the application client can be operated by the target user.
[0046] The application client can send to the application server a request to display the marriage partner recommendation page based on the target user's operation; for example, when the application client detects an instruction to enter the marriage partner recommendation page triggered by the target user's operation, the application client The client can send a request to display a marriage partner recommendation page to the application server. The marriage partner recommendation page can display information about several marriage partners recommended by the application server for the target user.
[0047] It can be understood that the manner in which the application client sends a request to display the recommendation page of the marriage partner to the application server based on the operation of the target user is not limited to the above description.
[0048] Step S11: The application server selects candidate marriage partners that meet the basic requirements in the attribute information database;
[0049] Specifically, the application server accesses the attribute information database, and selects candidate marriage partners whose attribute information meets the basic requirements based on the attribute information of each marriage object recorded in the attribute information database.
[0050] Step S12: The application server obtains the credit score of the target user and each of the candidate marriage partners;
[0051] Among them, the user's credit score indicates the degree of sincerity of the user's blind date. The higher the credit score, the higher the sincerity of users on blind dates.
[0052] The application server can access the credit database, and search for the corresponding credit score based on the target user and each candidate's name, ID number, and mobile phone number.
[0053] Optionally, the user's credit score stored in the credit database may be provided by a third-party platform, or may be the credit score adjusted by the marriage partner recommendation platform of this application.
[0054] Step S13: The application server determines the credit score matching degree of the target user and each of the candidate romantic objects according to the credit scores of the target user and each of the candidate romantic objects, as the first matching degree;
[0055] Specifically, if the credit scores of the target user and the candidate marriage partner are closer, it means that the credit scores of the target user and the candidate marriage partner are higher.
[0056] In this embodiment, it is believed that the easier it is for the target user to associate with the romantic partner whose credit score is close to him, the easier it is to match successfully in the end.
[0057] When determining the credit score matching degree between the target user and each of the candidate marriage partners, the credit scores can be matched according to a preset matching strategy. In an optional implementation manner, the target user and the candidate marriage partner are determined The process of matching the credit score of the object can be as follows:
[0058]
[0059] Among them, m1 is the matching degree of the credit scores of the target user and the candidate marriage partner; s1 and s2 respectively represent the credit scores of the target user and the candidate marriage partner; T is a set constant.
[0060] According to the above formula, it can be seen that when the credit scores of the target user and the candidate marriage partner are closer, the matching degree of the credit scores of the two is higher.
[0061] It is understandable that, in addition to the above-mentioned credit score matching calculation method, this application can also use other methods to calculate the credit score matching between the target user and the candidate marriage partner, which is not strictly limited in this application.
[0062] Optionally, the application server may also obtain the attribute information of the target user and each of the candidate marriage partners, and then determine the target user and each of the candidate marriage partners according to the attribute information of the target user and each candidate marriage partner. The matching degree of attribute information of candidate marriage and love objects is described. On this basis, the application server takes the credit score matching degree and the attribute information matching degree together as the first matching degree.
[0063] Among them, the attribute information includes but is not limited to the following: age, income, face value, constellation, personality, residence, interest, work industry.
[0064] In this embodiment, the above-mentioned several types of attribute information are used to respectively introduce the calculation process of the matching degree of each attribute information:
[0065] 1) Age matching:
[0066]
[0067] Among them, m2 is the age matching degree of the target user and the candidate marriage partner; α1 and α2 are the ages of the target user and the candidate marriage partner respectively; a'is the set age constant, and a'represents the optimal age difference between men and women.
[0068] 2) Income matching degree:
[0069]
[0070] Among them, m3 is the income matching degree between the target user and the candidate marriage partner; β1 is the income of the male boy among the target user and the candidate marriage partner, and β2 is the income of the female student among the target user and the candidate marriage partner.
[0071] It can be seen from the above formula that the greater the income of boys than girls, the more matching the two.
[0072] 3) Matching degree of appearance:
[0073]
[0074] Among them, m4 is the matching degree of the face value of the target user and the candidate marriage partner; χ1 and χ2 are the face values ​​of the target user and the candidate marriage partner respectively; avg is the average value of the face value of the target user and the candidate marriage partner.
[0075] It can be seen from the above formula that when the variance of the face value of the target user and the candidate marriage partner is smaller, the closer their face value is, the higher the matching degree.
[0076] 4) Constellation matching degree:
[0077] Pre-construct the constellation matching rule table, such as Sagittarius and Virgo are inappropriate. Furthermore, the constellation matching degree between the target user and the candidate marriage partner is determined according to the constellation matching rule table.
[0078] 5) Personality matching:
[0079] The way of determining the degree of personality matching is similar to that of constellation matching. By constructing a personality matching rule table, the personality matching degree of the target user and the candidate marriage partner is determined.
[0080] 6) Matching degree of residence:
[0081] The matching degree of residence is relatively simple. If the target user and the candidate marriage partner have the same residence, the matching degree of residence is determined to be 1, otherwise, it is 0.
[0082] 7) Interest matching degree:
[0083]
[0084] Among them, m5 is the interest matching degree of the target user and the candidate marriage partner; δ1j represents the degree of interest j of the target user δ1; δ2j represents the degree of interest j of the candidate marriage partner δ2; the total number of interests is n.
[0085] 8) Matching degree of work industry:
[0086]
[0087] Among them, m6 is the degree of match between the target user and the work industry of the candidate marriage partner; ε1 represents the degree of interest of the target user in the work industry of the candidate marriage partner, and ε2 represents the degree of interest the candidate marriage partner is in the work industry of the target user.
[0088] The foregoing embodiment only exemplifies several ways of calculating the matching degree of attribute information. In addition to the foregoing attributes, other attribute information can also be used to calculate the matching degree.
[0089] After determining the matching degree of each attribute information of the target user and the candidate marriage partner, the first matching degree may be determined according to the matching degree of each attribute information and the credit score matching degree.
[0090] There are many ways to determine the first matching degree. An optional method is to set a weight for each attribute information matching degree and credit score matching degree, and then to weight the attribute information matching degree and credit score matching degree. Plus, the result is regarded as the first matching degree. In addition, other methods may also be used to determine the first degree of matching.
[0091] Step S14: The application server determines the ranking result of each candidate marriage partner on the marriage partner recommendation page at least according to the first matching degree between the target user and each candidate marriage partner;
[0092] Specifically, the application server may sort the candidate marriage partners in the descending order of the first matching degree, and the sorting result is used as the sort order in the marriage partner recommendation page.
[0093] It is understandable that, according to this sorting order, the target user can first browse the candidate marriage partners with a higher matching degree of credit score and attribute information, and the probability of a successful match will be higher in the end.
[0094] Step S15: Send the marriage partner recommendation page with the ranking result to the application client.
[0095] The method provided by the embodiment of this application, after obtaining the candidate marriage partner according to the basic requirements of the target user for the marriage partner, further obtain the credit score of the target user and each candidate marriage partner, and determine the target user and each candidate marriage partner The credit score matching degree of is used as the first matching degree. According to the first matching degree, the ranking result of each candidate marriage partner on the marriage partner recommendation page is determined, and then in the marriage partner recommendation page sent to the user, the highest matching degree The marriage partner can be ranked first, and the target user can firstly check the marriage partner with a higher degree of matching his credit score, which improves the probability that the target user can quickly find the favorite marriage partner.
[0096] Optionally, this application can set different recommendation strategies for new users and old users.
[0097] Among them, new users refer to users who meet the conditions for setting new users; old users refer to users who meet the conditions for setting old users.
[0098] The new user conditions may include: users within a set period of time after registration, such as users within 3 months after registration as new users; or, the number of times the request to display the recommendation page of the marriage partner is within the set number of times. In this application, the interface of the application client may display a refresh button. Each time the user clicks the refresh button, it can be regarded as sending a request to the application server to display the recommendation page of the marriage partner.
[0099] Correspondingly, the old user conditions may include: users after a set period of time after registration, such as users after 3 months after registration as old users; or, the number of requests to display the recommended page of marriage partners exceeds the set number of times.
[0100] Recommended strategies for new users:
[0101] If it is determined that the target user is a new user, the application server determines the ranking result of each candidate marriage partner on the marriage partner recommendation page at least according to the first matching degree between the target user and each candidate marriage partner , Which can specifically include:
[0102] The candidate marriage partners are sequentially displayed on the marriage partner recommendation page in the descending order of the first matching degree.
[0103] Further, the marriage partner recommendation page may be sent to the application client, and the application client may display it.
[0104] For new users, this application pushes all eligible candidate marriage partners to the application client for display. The display mode of the application client can include:
[0105] All candidate marriage and love objects are displayed on one page, and the sorting order of the candidate marriage and love objects is in descending order according to the first matching degree. In addition, the application client can divide all candidate marriage and love objects into several consecutive groups according to the sort order, and then display the candidate marriage and love objects in different groups through multiple pages. For example, when the user clicks to turn the page, the next set of candidate marriage partners is displayed.
[0106] Of course, when the candidate marriage partners in each group are displayed, they are also displayed in the descending order of the first matching degree.
[0107] Further, introduce the recommendation strategy of old users:
[0108] If it is determined that the target user is an old user, the application server may further determine a second matching degree between the target user and each candidate marriage partner, and the second matching degree is based on the relationship between the target user and the candidate marriage partner. Communication behavior data is determined.
[0109] Among them, the communication behavior data between the target user and the candidate marriage partner refers to whether the target user clicks on the candidate marriage partner, whether to communicate with each other, and the communication time, etc., reflecting the frequent degree of interaction between the target user and the candidate marriage partner.
[0110] It is understandable that only when the target user is an old user, the marriage partner recommendation platform can collect enough communication behavior data between the target user and the candidate marriage partner, and then can determine the target user and the candidate marriage partner based on the communication behavior data. Two matching degrees.
[0111] On this basis, the application server determines the ranking result of each candidate in the marriage partner recommendation page at least according to the first degree of matching between the target user and each of the candidate marriage partners, which can be referred to image 3 As shown, the process can include:
[0112] Step S300: Determine the display probability of each candidate marriage partner according to the first matching degree and the second matching degree between the target user and each candidate marriage partner;
[0113] Specifically, in this embodiment, the application server combines the first matching degree and the second matching degree between the target user and the candidate marriage partner to determine the display probability of each candidate marriage partner.
[0114] In an optional implementation manner, the process of determining the display probability of each candidate marriage partner may refer to the following formula:
[0115] P=x*M1+(1-x)*M2
[0116] Among them, P represents the display probability of the candidate marriage partner; M1 represents the first matching degree, M2 represents the second matching degree; x is a fixed constant, and the value range is [0, 1].
[0117] It is understandable that in addition to determining the display probability P according to the above formula, the present application may also use other methods to determine the display probability P by combining the first matching degree and the second matching degree. The specific determination method is not strictly limited in this application.
[0118] Step S310, according to the display probability of each candidate marriage partner, select a target number of candidate marriage partners from each candidate partner;
[0119] Specifically, according to the display probability of each candidate marriage partner, a target number of candidate marriage partners is randomly determined and displayed, and the remaining candidate marriage partners are not displayed in this recommendation process, that is, they are not pushed to the target user.
[0120] It is understandable that the higher the display probability of the candidate marriage partner, the higher the probability of being selected, and the higher the probability of recommending to the target user in this recommendation process.
[0121] In step S320, the selected target number of candidate marriage partners are sequentially displayed on the marriage partner recommendation page in descending order of display probability.
[0122] Specifically, by displaying the selected target number of candidate marriage partners in the marriage partner recommendation page, the marriage partner recommendation page is further sent to the target user.
[0123] In this embodiment, for the case that the target user is an old user, the second matching degree between the target user and each candidate marriage partner is further determined, and the display probability of each candidate marriage partner is determined based on the first matching degree and the second matching degree. Display probability to select the target number of candidate marriage partners, display them on the marriage partner recommendation page, and recommend them to the target users.
[0124] According to the operation mode of this embodiment, every time an old user requests a marriage partner recommendation page, he can see a new candidate marriage partner, avoiding being too monotonous.
[0125] Optionally, this embodiment of the application introduces the process by which the above-mentioned application server determines the second degree of matching between the target user and each candidate marriage partner, see Figure 4 As shown, the process can include:
[0126] Step S400, among the candidate marriage partners, determine the candidate marriage partners that have been displayed on the marriage partner recommendation page before sending the marriage partner recommendation page to the application client this time, and the candidate marriage partner that has not yet been in the marriage partner recommendation page Candidates for marriage and love that have not been shown on the previous page;
[0127] Specifically, the target user may request the marriage partner recommendation page from the application server multiple times. When receiving the current marriage partner recommendation page request sent by the target user, the application server determines the candidate marriage partner that has been displayed on the marriage partner recommendation page sent to the target user before this time as the displayed candidate marriage partner . At the same time, the candidate marriage partner that has not been displayed on the marriage partner recommendation page is determined as the undisplayed candidate marriage partner.
[0128] Step S410: It is determined in this recommendation process that the target user's recognition of the undisplayed candidate marriage partner is the first matching degree between the target user and the undisplayed candidate marriage partner;
[0129] Specifically, this application further defines the user's recognition of the candidate marriage partner Wi. The application server determines in this recommendation process that the target user’s recognition of the undisplayed candidate marriage partner is the first degree of matching between the target user and the undisplayed candidate marriage partner.
[0130] That is, if the candidate marriage partner has not been displayed, the recognition degree of the target user is equal to the first matching degree of the target user.
[0131] Step S420: According to the recognition of the displayed candidate marriage partner by the target user in the previous recommendation process, and the historical communication behavior data between the target user and the displayed candidate marriage partner, determine the current recommendation process. Recognition of marriage partner;
[0132] Specifically, for the candidate marriage partner that has been displayed, the way in which the target user's approval degree is determined in this recommendation process is different from that of the candidate marriage partner that is not displayed.
[0133] In this step, the application server determines the recommendation process according to the recognition of the target user to the displayed candidate marriage partner and the historical communication behavior data between the target user and the displayed candidate marriage partner during the previous recommendation process of this recommendation process , The target user’s recognition of the displayed candidate marriage partner.
[0134] In an optional implementation manner, the process of determining the recognition of the target user for the displayed candidate marriage partner may refer to the following:
[0135] Wi=prev_Wi*exp(gama*Xi/m)
[0136]
[0137] Among them, Wi represents this recommendation process, the target user's recognition of the candidate marriage partner i; prev_Wi represents the previous recommendation process, the target user's recognition of the candidate marriage partner i; gama is a preset fixed constant, the value range is [0,1]; m is the total number of requests made by the target user from the first request to this request to display the recommendation page of the marriage partner. It can be set that the target user clicks the refresh button once, which is regarded as a request; i Indicates the degree of interest of the target user in the candidate marriage partner i, and Pi represents the display probability of the candidate marriage partner i.
[0138] Wherein, the degree of the target user's interest in the candidate marriage partner i can be determined by the historical communication behavior data between the target user and the candidate marriage partner. For example, the longer the communication time between the two and the more the number of times, the higher the degree of interest.
[0139] Step S430: Determine the second match between the target user and each candidate marriage partner according to the target user’s recognition of each candidate marriage partner and the total number of requests from the target user to display the marriage partner recommendation page from the first request to this request degree.
[0140] In an optional implementation manner, the process of determining the second matching degree between the target user and the candidate marriage partner may refer to the following:
[0141]
[0142] Among them, M2i represents the second matching degree between the target user and the candidate marriage partner i; Wi target user's recognition of the candidate marriage partner i, the number of candidate marriage partners is n; m is the target user from the first request to this request Shows the total number of requests for the recommendation page of the marriage partner, which can also be called the number of refreshes of the target user; alpha is a set constant, and the value range is [0, 1].
[0143] In the above formula, One meaning is that the target user's recognition of the i-th candidate marriage partner is relative to the relative recognition of all candidate marriage partners.
[0144] It is understandable that as the number of requests from target users increases, the greater m is, The smaller, The smaller the value, The greater the role played, the more important the target user’s recognition of the candidate marriage partner. This is consistent with the actual situation, because as the number of target user requests increases, the more the target user’s communication behavior data collected by the dating partner recommendation platform, the more accurate the target user’s recognition of the candidate marriage partner based on this determination becomes. high,
[0145] Optionally, the application server can also adjust the user's credit score, such as reducing the credit score of a user who has fraudulent marriages, harassing the romantic partner, or insulting, so as to adjust the user's ranking order in the romantic partner recommendation page . Next, taking the adjustment of the credit score of the target user as an example, this embodiment of the application describes in detail the process of adjusting the credit score of the target user by the application server, see Figure 5 As shown, the process can include:
[0146] Step S500: Collect historical communication behavior data between the target user and the candidate marriage partner, and the number of times the target user has been reported as having bad behavior;
[0147] Specifically, the candidate marriage partner is in the communication process of the target user. If the target user is found to have bad behavior, the target user can be reported to the marriage partner recommendation platform. At the same time, the marriage partner recommendation platform can collect historical communication behavior data between target users and candidate marriage partners.
[0148] Step S510: Determine the first degree of bad behavior of the target user according to the historical communication behavior data of the target user and the candidate marriage partner, and the pre-trained bad behavior degree prediction model;
[0149] Specifically, the present application may pre-train the bad behavior degree prediction model, so as to use the collected historical communication behavior data of the target user and the candidate marriage partner to determine the first bad behavior degree of the target user.
[0150] In an optional implementation manner, the training process of the bad behavior degree prediction model may include:
[0151] S1. Obtain positive and negative sample data, where the positive sample data includes normal user dialogue content, and the negative sample data includes the dialogue content of a user who is reported to have bad behavior.
[0152] S2, according to the feature template, extract the set features from the positive and negative sample data respectively, and use the extracted features to train the machine learning model to obtain the trained bad behavior degree prediction model.
[0153] Among them, feature templates include but are not limited to the following:
[0154] A. Describe the degree of harassment
[0155] Whether to insult the other party (whether it contains common abusive expressions); whether to entangle the other party (after being rejected by the other party, keep sending messages).
[0156] B. Portray the degree of insincerity
[0157] Keep in touch with many people of the opposite sex at the same time; After recommending a suitable person, and after in-depth communication, they still keep close contact with other marriage partners; After the two enter the certification stage, the number of times they change their minds.
[0158] C. Describe the degree of fraudulent marriage
[0159] The degree of asking for gifts; the degree of urgency; the frequency and amount of other requests for money (such as frequent requests for gifts).
[0160] D. Describe the degree of seduction
[0161] Whether to deliberately show off personal behavior data such as wealth and status; whether to pretend to be oneself; whether to deliberately seek sympathy.
[0162] After extracting the features, use GBDT (Gradient Boosting Decision Tree, a machine learning method) or other methods to train the model to obtain the trained bad behavior degree prediction model.
[0163] After the trained bad behavior degree prediction model is obtained, the first degree of bad behavior of the target user can be determined by inputting the collected historical communication behavior data of the target user and the candidate marriage partner into the prediction model.
[0164] Step S520: Determine the second degree of bad behavior of the target user according to the number of times the target user has been reported as having bad behavior;
[0165] It can be understood that the execution order of step S410 and step S420 is not limited, and the two can be reversed or executed simultaneously.
[0166] Step S530: Determine the total degree of bad behavior of the target user according to the first degree of bad behavior and the second degree of bad behavior;
[0167] Specifically, weights can be set for the first degree of bad behavior and the second degree of bad behavior respectively, and then the total bad behavior degree of the target user is determined by a weighted summation method. It is necessary to ensure that the total bad behavior degree does not exceed 1.
[0168] Step S540: Determine the adjusted credit score of the target user according to the credit score of the target user and the total degree of bad behavior.
[0169] Optionally, this application can set a credit decay model to describe the user's credit score. The credit decay model can be as follows:
[0170] S′=S*e -beta*η
[0171] Among them, S′ represents the adjusted credit score of the target user, S represents the initial credit score of the target user, η represents the total degree of bad behavior of the target user, and beta is a preset constant with a value greater than 1, which is used to control the attenuation rate. The larger the value of beta, the faster the decay. That is to say, if the target user has a bad behavior, his credit score will decrease a lot; the smaller the value of beta, the slower the decay, that is, the target user has a bad behavior. Its credit score has decreased slightly.
[0172] It should be noted that the timing for the application server to adjust the user's credit score may be to uniformly adjust the credit score of each registered user every set period. It may also be that when it is found that the number of times a user has been complained reaches the set threshold, the credit score of the user is adjusted.
[0173] By adjusting the user's credit score, it is ensured that when a marriage partner is recommended after adjustment, the sort order is determined according to the user's adjusted credit score.
[0174] The following describes the device for recommending a marriage partner provided by the embodiments of the present application. The device for recommending a marriage partner described below can be cross-referenced with the method for recommending a marriage partner described above. The device for recommending a romantic partner described below can be considered as a functional module structure that the application server needs to set in order to implement the method for recommending a romantic partner provided by the embodiment of the present application.
[0175] Image 6 This is a structural block diagram of the device for recommending marriage and love objects provided in this embodiment of the application. The device can be applied to an application server. Refer to Image 6 , The device may include:
[0176] The request information obtaining unit 11 is configured to obtain information sent by the application client requesting to display the recommended page of the marriage partner, the information including the basic requirements of the target user corresponding to the application client for the marriage partner;
[0177] The candidate marriage partner selection unit 12 is configured to select the candidate marriage partner that meets the basic requirements in the marriage partner database;
[0178] The credit score acquiring unit 13 is configured to acquire the credit scores of the target user and each of the candidate marriage partners, wherein the user's credit score indicates the degree of sincerity of the user for blind date;
[0179] The credit score matching degree determining unit 14 is configured to determine the credit score matching degree of the target user and each of the candidate marriage partners according to the credit scores of the target user and each of the candidate marriage partners, as the first matching degree ;
[0180] The ranking result determining unit 15 is configured to determine the ranking result of each candidate marriage partner on the marriage partner recommendation page at least according to the first degree of matching between the target user and each candidate marriage partner;
[0181] The recommendation page sending unit 16 is configured to send the marriage and love object recommendation page with the ranking result to the application client.
[0182] The marriage partner recommendation device provided by the embodiment of the application, after obtaining the candidate marriage partner according to the basic requirements of the target user for the marriage partner, further obtains the credit score of the target user and each candidate marriage partner, and determines the target user and each said The credit score matching degree of the candidate marriage partner is used as the first matching degree, and the ranking result of each candidate marriage partner on the marriage partner recommendation page is determined according to the first matching degree, and then in the marriage partner recommendation page sent to the user, high Matching marriage partners can be ranked first, and the target user can first check the marriage partner with a higher degree of matching his credit score, which improves the probability of the target user quickly finding the favorite marriage partner.
[0183] Optionally, the device of this application may also include:
[0184] The attribute information obtaining unit is configured to obtain attribute information of the target user and each candidate marriage partner;
[0185] The attribute information matching degree determining unit is configured to determine the attribute information matching degree of the target user and each candidate marriage partner according to the attribute information of the target user and each of the candidate marriage partners, and match the attribute information The degree of matching with the credit score is used together as the first degree of matching.
[0186] Wherein, the attribute information may include any one or more of age, income, face value, constellation, personality, place of residence, interest, and work industry.
[0187] Optionally, this application can provide different recommendation strategies for the target user being a new user or an old user:
[0188] If the target user is a user who meets the conditions for setting a new user, the ranking result determining unit may include:
[0189] The first sorting result determination subunit is used to display each of the candidate marriage and love objects in the order of the first matching degree in descending order on the marriage and love object recommendation page.
[0190] If the target user is a user who meets the set old user conditions, the device of this application may further include:
[0191] The second matching degree determining unit is configured to determine a second matching degree between the target user and each candidate marriage partner, and the second matching degree is based on communication behavior data between the target user and the candidate marriage partner Determined. Based on this, the ranking result determining unit may include:
[0192] The second ranking result determination subunit is used to determine the display probability of each candidate marriage partner according to the first matching degree and the second matching degree between the target user and each candidate marriage partner;
[0193] The third sorting result determination subunit is used to select a target number of candidate marriage partners from each candidate marriage partner according to the display probability of each candidate marriage partner;
[0194] The fourth sorting result determination subunit is used to display the selected target number of candidate marriage partners in descending order of display probability on the marriage partner recommendation page.
[0195] Optionally, the process for the second matching degree determining unit to determine the second matching degree between the target user and each candidate marriage partner may specifically include:
[0196] Among the candidate marriage partners, it is determined that before sending the marriage partner recommendation page to the application client this time, the candidate marriage partners that have been displayed on the marriage partner recommendation page, and those that have not yet been displayed on the marriage partner recommendation page Does not show candidate marriage partners;
[0197] Determine this recommendation process, the target user's recognition of the undisplayed candidate marriage partner is the first degree of matching between the target user and the undisplayed candidate marriage partner;
[0198] According to the target user’s recognition of the displayed candidate marriage partner in the previous recommendation process, and the historical communication behavior data between the target user and the displayed candidate marriage partner, determine the recommendation process, the target user’s opinion of the displayed candidate marriage partner Recognition
[0199] According to the target user's recognition of each candidate marriage partner, and the total number of requests of the target user from the first request to this request to display the marriage partner recommendation page, the second matching degree between the target user and each candidate marriage partner is determined.
[0200] Optionally, the device of this application may also include:
[0201] A historical communication behavior data collection unit, configured to collect historical communication behavior data between the target user and the candidate marriage partner, and the number of times the target user has been reported as having bad behavior;
[0202] The first bad behavior degree determining unit is configured to determine the first bad behavior degree of the target user according to the historical communication behavior data of the target user and the candidate marriage partner, and a pre-trained bad behavior degree prediction model;
[0203] The second degree of bad behavior determining unit is configured to determine the second degree of bad behavior of the target user according to the number of times the target user has been reported as having bad behavior;
[0204] A total bad behavior degree determining unit, configured to determine the total bad behavior degree of the target user according to the first bad behavior degree and the second bad behavior degree;
[0205] The credit score adjustment unit is configured to determine the credit score of the target user after adjustment according to the credit score of the target user and the total degree of bad behavior.
[0206] Optionally, the device of the present application may further include: a model training unit for training to obtain a bad behavior degree prediction model; the model training unit may include:
[0207] The sample data acquisition unit is used to acquire positive and negative sample data, where the positive sample data includes normal user dialogue content, and the negative sample data includes the dialogue content of a user who is reported to have bad behavior;
[0208] The feature training unit is used to extract set features from the positive and negative sample data according to the feature template, and use the extracted features to train the machine learning model to obtain the trained bad behavior degree prediction model.
[0209] An embodiment of the present application also provides an application server, which may include the aforementioned device for recommending a marriage partner.
[0210] Figure 7 Shows the optional hardware structure of the application server, refer to Figure 7 , The application server may include: a processor 1, a communication interface 2, a memory 3 and a communication bus 4;
[0211] The processor 1, the communication interface 2, and the memory 3 communicate with each other through the communication bus 4;
[0212] Optionally, the communication interface 2 may be an interface of a communication module, such as an interface of a GSM module;
[0213] The processor 1 may be a central processing unit CPU, or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.
[0214] The memory 3 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
[0215] Among them, the processor 1 is specifically used for:
[0216] Acquiring the information sent by the application client requesting to display the recommendation page of the marriage partner, the information including the basic requirements of the target user corresponding to the application client for the marriage partner;
[0217] Select candidate marriage partners that meet the basic requirements in the marriage partner database;
[0218] Obtaining credit scores of the target user and each of the candidate marriage partners, where the user's credit score indicates the degree of sincerity of the user for blind date;
[0219] According to the credit scores of the target user and each of the candidate marriage partners, determine the credit score matching degree between the target user and each candidate marriage partner as the first matching degree;
[0220] At least according to the first degree of matching between the target user and each of the candidate marriage partners, determining a ranking result of each candidate marriage partner on the marriage partner recommendation page;
[0221] Sending the marriage partner recommendation page with the ranking result to the application client.
[0222] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.
[0223] Professionals can further realize that the units and algorithm steps of the examples described in the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of both, in order to clearly illustrate the possibilities of hardware and software. Interchangeability, in the above description, the composition and steps of each example have been described generally in terms of function. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
[0224] The steps of the method or algorithm described in combination with the embodiments disclosed herein can be directly implemented by hardware, a software module executed by a processor, or a combination of the two. The software module can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or all areas in the technical field. Any other known storage media.
[0225] The above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined in this document can be implemented in other embodiments without departing from the core idea or scope of the application. Therefore, this application will not be limited to the embodiments shown in this text, but should conform to the widest scope consistent with the principles and novel features disclosed in this text.

PUM

no PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products