TW201740337A - Friend pairing algorithm including collecting real time information from terminal and historic information stored on social media - Google Patents

Abstract

The invention relates to a friend pairing algorithm. The friend pairing algorithm includes: (A) receiving information on multiple users from multiple terminals by a friend pairing server unit; (B) accessing member information associated with multiple users stored on a social media server unit with the friend pairing server unit and generating characteristic information corresponding to the users based on the member information; (C) generating multiple filtering conditions corresponding to the users according to the characteristic information and the user information with the friend pairing server unit; and (D) generating multiple recommended friend lists related to users based on the filtering conditions with friend pairing server unit for users to access through the terminals.

Description

Dating pairing calculation method

The invention relates to a dating pair calculation algorithm, in particular to a dating matching calculation method for recommending friends according to the membership information of the social network website.

There are a number of dating sites that offer dating and matching services for users to use as a platform to meet new people through dating sites. The general dating site only recommends the user based on the personal data entered by the user when registering. The inventor of the case, Min Sihe, if he can develop a new method of dating matching calculations, can collect more information about registered members, and generate recommendations based on the more information, which will be more suitable for users. Object.

Accordingly, it is an object of the present invention to provide a dating matching calculation method that can recommend a more suitable object for a user.

Therefore, the friend pairing calculation method of the present invention is implemented by a community website server unit, a friend matching server unit and a plurality of terminal devices, and the community website server unit stores the member data of the plurality of users. The dating pairing calculation method includes: (A) the friend pairing server unit receives a plurality of personal data from the terminal devices and respectively corresponding to the user; (B) the friend pairing server unit accesses the community The member data stored by the server server unit, and generating a plurality of characteristic data corresponding to the user according to the member data; (C) the friend matching server unit according to the characteristic data and the personal data Generating a plurality of screening conditions respectively corresponding to the users, the screening conditions for screening the users to be recommended from the users; and (D) the dating matching server unit generating according to the screening conditions A plurality of friend recommendation lists respectively corresponding to the users are accessed by the users via the terminal devices.

In some implementations, between step (B) and step (C), the method further comprises: (E) determining, by the friend matching server unit, each of the users through the friend-matching server unit and at least another user Whether the interaction conforms to a predetermined condition that the user interacts well with the at least one other user. In this embodiment, the step (C) is when the friend pairing server unit determines that the interaction of one of the users with the at least another user through the dating server unit meets the predetermined condition, the friend pairing server unit Adding the at least another user to a good interaction list corresponding to the one of the users, and generating a correspondence according to the feature data of the at least another user in the interaction good list and the personal data The screening condition of one of the users. In this embodiment, the step (D) is that the friend matching server unit generates a friend recommendation list corresponding to the one of the users according to the screening condition, for the one of the users to save via one of the terminal devices. take.

In some implementations, after the step (D), the method further includes: (F) the friend matching server unit determines that the user interacts with the user in the friend recommendation list through the dating server unit Whether the meeting condition is met; and (G) when the friend matching server unit determines that the user interacts with the user in the friend recommendation list through the friend matching server unit within a predetermined time The predetermined condition, the friend matching server unit adds the user in the friend recommendation list to an interaction unsatisfactory list corresponding to the one of the users, and performs step (C). In this embodiment, in step (C), the friend matching server unit generates the screening condition according to the feature data of the interactive good list and the user in the interactive unsatisfactory list and the personal data. .

In some implementations, the member profile of each user stored in the community website server unit includes a record of the user's favorite.

In some implementations, the member profile of each user stored in the community website server unit includes a fan group that the user joins.

In some implementations, the member profile of each user stored in the community website server unit includes the user's interest profile.

The effect of the present invention is that the friend matching server unit generates the friend recommendation list according to the personal data received from the terminal device and the member information stored by the community website server unit, thereby Can recommend more suitable objects for users.

Referring to FIG. 1 and FIG. 2, it is an embodiment of the dating matching calculation method of the present invention. The friend pairing calculation method of the present invention is implemented by a community website server unit 1, a friend pairing server unit 2, and a plurality of terminal devices 3. The dating matching server unit 2 is connected to the social networking server unit 1 and the terminal device 3 via the communication network 200. The social network server unit 1 of the present embodiment is a server unit of a Facebook® website that is currently very popular and has a large number of users. The terminal device 3 can be, for example, a smart phone, a tablet computer, a notebook computer, or a desktop computer. A computing device that connects to the Internet. The plurality of users 100 can connect the social network server unit 1 and the dating server unit 2 using the terminal device 3. The users 100 of the embodiment are registered as members of one of the community websites constructed by the social network server unit 1, and the social network server unit 1 stores the member information of the users 100. Each member profile includes a record of the user's 100 praise, a fan group to which the user 100 joins, interest data of the user 100, and other basic information of the user 100 (eg, favorite books, schools, occupations, etc.) .

The flow of the friend pairing calculation method of the present invention firstly, as shown in step S01, the friend pairing server unit 2 receives a plurality of personal data from the terminal devices 3 and respectively corresponding to the user 100. That is to say, the user 100 connects to a dating partner website constructed by the dating match server unit 2 through the terminal device 3, and registers as a member of the dating partner website. When registering, each user 100 needs to input the personal data, which includes gender, height, size, age, region, marriage needs, living habits and the like. Each user 100 also needs to enter an account number on the social networking site, and agree to the statement that the friend matching server unit 2 accesses the member information corresponding to the account.

Then, as shown in step S02, the friend matching server unit 2 accesses the member materials stored in the community website server unit 1, and generates a plurality of feature data corresponding to the user 100 according to the member data. The feature information includes, for example, the user 100 joining the XXX fan group.

Then, as shown in step S03, the friend matching server unit 2 determines whether one of the users 100 interacts with at least one other user 100 through the dating server unit 2, and represents one of the users 100 and the At least another user 100 interacts with a good predetermined condition, and the at least one other user 100 has not been added to a well-matched list corresponding to the one of the users 100. In this embodiment, the predetermined condition includes that one of the users 100 and the at least another user 100 add friends to each other on the dating site, express mutual interest or transmit messages to each other more than a certain number of sentences. If the answer of step S03 is YES, step S04 is performed. If the result of step S03 is NO, step S02 is executed to update the feature data of the user 100.

Step S04 is that the friend matching server unit 2 adds the at least another user 100 to the interactive good list corresponding to the one of the users 100.

Then, as shown in step S05, the friend matching server unit 2 generates, according to the interactive good list of one of the users 100 and the feature data and personal data of the user 100 in an interactive unsatisfied list. One of the screening criteria for the user 100. The screening criteria are used to screen out from the users 100 the user 100 to be recommended to the one of the users 100. How the user 100 in the unsuccessful list is added is explained in steps S07 and S08.

Then, as shown in step S06, the friend matching server unit 2 generates a friend recommendation list corresponding to the one of the users 100 according to the screening condition, for the one of the users 100 to access via one of the terminal devices 3. .

For example, if the users 100b, 100c, and 100d join the XXX fan group on the social networking site, and if the friend matching server unit 2 determines that the interaction between the user 100a and the user 100b meets the predetermined condition, the friend pairing servo The unit 2 will add the user 100b to the interactive list of users 100a. Since the feature data of the user 100b includes the user 100b joining the XXX fan group, the screening condition of the user 100a generated by the friend matching server unit 2 includes joining the XXX fan group. Since the feature data of the users 100c and 100d includes the XXX fan group, the friend matching server unit 2 puts the users 100c and 100d into the friend recommendation list of the user 100a.

In the foregoing steps S02 to S06, the friend matching server list generates a friend recommendation list according to the member information of the social network website in addition to the personal data of the user 100, and in the case where the data is richer. The friend matching server can recommend a more suitable object for the user 100. In addition, since the present embodiment compares the interaction state with the prior art, that is, the feature data is analyzed for the well-interacted object to generate a friend recommendation list, thereby further improving the suitability of the recommended object.

After step S06 is performed, next, as shown in step S07, the friend matching server unit 2 determines that one of the users 100 passes the friend matching server unit 2 and the friend recommendation list within a predetermined time (for example, within one month). Whether the interaction of the user 100 in the meeting meets the predetermined condition, if not, step S08 is performed, and if yes, step S09 is performed.

Step S08 is that the friend matching server unit 2 joins the user 100 who does not interact well with the one of the users 100 in the friend recommendation list to join the interaction list of the one of the users 100. Step S09 is a friend matching server unit 2 that joins the user 100 who interacts well with one of the users 100 in the friend recommendation list to the interactive good list of the one of the users 100. After step S08 or step S09 is performed, step S02 is performed to update the feature data of the user 100. When the process proceeds to step S05 again, the friend matching server unit 2 generates a screening condition according to the interaction unsatisfactory list, that is, the embodiment further tracks the one after providing the user 100 with the friend recommendation list. If the user 100 interacts with the user 100 in the friend recommendation list, if the interaction is not good, the screening condition can be corrected through the interaction unsatisfactory list, so that the screening condition is more and more consistent with the preference of one of the users 100/ A requirement to further recommend a more suitable object for one of the users 100.

In summary, the embodiment of the friend pairing calculation method of the present invention generates a friend recommendation based on the member data stored in the social network server unit 1 not only according to the personal data received from the terminal device 3 but also by the friend matching server unit 2. a list, so that a more suitable object can be recommended for the user 100; furthermore, the friend matching server unit 2 generates a friend recommendation list according to the well-organized list, and can further improve the suitability of the recommended object by analyzing the well-interactive object; The friend matching server unit 2 generates a friend recommendation list according to the interaction unsatisfactory list, and can track the interaction between the user 100 and the recommended object, thereby correcting the screening condition and the friend recommendation list, thereby enabling the user 100 to use the learning mechanism. A more suitable object is recommended, so that the object of the present invention can be achieved.

However, the above is only the embodiment of the present invention, and the scope of the invention is not limited thereto, and all the equivalent equivalent changes and modifications according to the scope of the patent application and the patent specification of the present invention are still The scope of the invention is covered.

1‧‧‧Community website server unit
2‧‧‧Dating pairing server unit
3‧‧‧ Terminal devices
100, 100a, 100b, 100c, 100d‧‧‧ users
200‧‧‧Communication network
S01~S09‧‧‧ Process steps

Other features and effects of the present invention will be apparent from the embodiments of the present invention. FIG. 1 is a schematic diagram of a hardware connection relationship, illustrating a hardware for implementing an embodiment of the dating pairing calculation method of the present invention; And Figure 2 is a flow chart illustrating the flow steps of the embodiment.

S01~S09‧‧‧ Process steps

Claims (6)

A friend pairing calculation method is implemented by a community website server unit, a friend matching server unit and a plurality of terminal devices, wherein the community website server unit stores member information of a plurality of users, and the friends The pairing calculation method includes: (A) the friend pairing server unit receives a plurality of personal data from the terminal devices and respectively corresponding to the user; (B) the friend matching server unit accesses the social network server The member information stored in the unit, and generating a plurality of characteristic data corresponding to the user according to the member data; (C) the friend matching server unit generates multiple based on the characteristic data and the personal data Corresponding to the screening conditions of the user, the screening conditions are used to screen out the users to be recommended from the users; and (D) the friend matching server unit generates multiple differences according to the screening conditions. A list of friend recommendations for the user should be accessed by the users via the terminal devices. The method for calculating the dating pairing method according to claim 1, further comprising: (e) the dating pairing server unit determines that each of the users matches the server unit through the friend pairing device and the step (C) Whether the interaction of the other user meets a predetermined condition that the user interacts well with the at least one other user; wherein, step (C) is when the dating partner server unit determines that one of the users is paired by the friend The interaction of the server unit with at least one other user meets the predetermined condition, and the friend-matching server unit adds the at least another user to a good interaction list corresponding to the one of the users, and according to the interaction The feature data of the at least one other user in the list and the personal data generate a screening condition corresponding to the one of the users; wherein the step (D) is generated by the friend matching server unit according to the screening condition A list of friend recommendations corresponding to the one of the users is accessed by one of the users via one of the terminal devices. The method for calculating the dating pairing method according to claim 2, further comprising, after the step (D): (F) the friend matching server unit determines that the one of the users passes the friend matching server unit and the friend recommendation list Whether the user's interaction meets the predetermined condition; and (G) when the friend matching server unit determines that the one of the users passes the friend matching server unit and the user in the friend recommendation list within a predetermined time The interaction does not meet the predetermined condition, and the friend matching server unit adds the user in the friend recommendation list to an interaction unsatisfactory list corresponding to the one of the users, and performs step (C); wherein, in the step In (C), the friend matching server unit generates the screening condition according to the feature data of the interactive good list and the user in the unsatisfactory interactive list and the personal data. The method for calculating a friend-matching calculation method according to claim 3, wherein the member profile of each user stored in the server unit of the community website includes a record of the user's favorite. The friend pairing calculation method according to claim 3, wherein the member profile of each user stored in the community website server unit includes a fan group that the user joins. The friend pairing calculation method according to claim 3, wherein the member data of each user stored in the server unit of the community website includes the user's interest data.