A personalized interpersonal relationship push method, device and system based on big data

Abstract

The embodiment of the present invention discloses a method, device and system for personalized interpersonal relationship push based on big data, wherein the method comprises: setting a social recommendation cycle, obtaining historical social data and historical transaction data of a user within the social recommendation cycle, inputting the historical social data and historical transaction data into a preset credit scoring model to obtain the social credit level of the user, and inputting the historical social data and historical transaction data into a preset classification model to obtain the social tag and transaction tag of the user; the present invention recommends friends to the user based on the social credit level of the user and the tag set corresponding to the user, which can not only personalize the interpersonal relationship push for the user, but also ensure the social security of the user, and fully improve the social experience of the user.

Description

Technical Field

[0001] The present invention relates to the field of big data processing technology, and in particular to a method, device and system for pushing personalized interpersonal relationships based on big data. Background Art

[0002] In recent years, with the rapid development of Internet technology, the use of social platforms has become more and more common in people's daily lives, making people more and more accustomed to speaking, chatting and making friends through social networks. In this case, the social network dating model was born; in order to allow users to make like-minded friends, the current social network dating model also provides a friend recommendation function, specifically analyzing the interests of users from the personal display information posted by users, so as to recommend friends with the same interests and hobbies to users. However, in online dating, users are usually unable to identify the identities of recommended friends, resulting in the risk of privacy leakage and fraud in dating. In the prior art, the usual practice is to divide users into credit levels, not recommend users with lower credit levels to users, or restrict the social activities of low-credit users, such as posting, making friends, etc. However, the credit of users will change at any time with the social activities of users. The above practices cannot guarantee the social security of high-credit users, and also limit the social experience of low-credit users. Summary of the invention

[0003] Therefore, the main purpose of the present invention is to provide a personalized interpersonal relationship push method, device and system based on big data, so as to achieve the comprehensive, accurate and objective quantification of the trust relationship between the recommended user and the neighbor user in the social network, recommend friends to the user based on the user's dynamic credit level, set different friend recommendation strategies for users of different levels, fully guarantee the social experience of users at all levels, and promote users to improve their credit level, thereby improving the security and stability of the social environment.

[0004] In a first aspect, an embodiment of the present invention provides a method for pushing personalized interpersonal relationships based on big data, the method comprising the following steps:

[0005] Step 1: Preset a fixed time interval as a social recommendation cycle, obtain the user's historical social data and historical transaction data within the current social recommendation cycle, and pre-process the historical social data and the historical transaction data;

[0006] Step 2: Input the pre-processed historical social data and the historical transaction data into a preset credit scoring model to calculate the social credit score of the user;

[0007] Step 3: Obtaining the social credit level of the user based on the social credit score of the user;

[0008] Step 4: Input the preprocessed historical social data and the historical transaction data into a preset classification model respectively to obtain the social tag and transaction tag of the user, and let the user set the interest tag by himself, and the social tag, the transaction tag and the interest tag constitute a tag set;

[0009] Step 5: Recommending friends to the current user according to the social credit level and the tag set corresponding to the user;

[0010] Step 6: Acquire feedback data of the user in the current social recommendation cycle, obtain an adjustment coefficient of the social credit level of the user based on the feedback data, and adjust the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient;

[0011] In step 5, making friend recommendations for the current user according to the social credit level and the tag set corresponding to the user specifically includes:

[0012] Assume that the user's social credit level is , ,

[0013] like , recommending the social credit level to the user and X users whose tag set is exactly the same as that of the user;

[0014] like , recommending the social credit level to the user and Y users whose tag set is exactly the same as that of the user;

[0015] X and Y are both preset natural numbers greater than 1, and Y>X;

[0016] In step 6, obtaining an adjustment coefficient of the social credit level of the user based on the feedback data, and adjusting the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient, specifically includes:

[0017] The feedback data is quantified to obtain an adjustment coefficient ; The feedback data includes positive feedback data and negative feedback data;

[0018] In the next social recommendation cycle, repeat steps 1 to 4 to obtain the social credit rating in the next social recommendation cycle. and a new set of labels, and then adjust the coefficients according to the Get the user's final social credit rating in the next social recommendation cycle , based on the The new tag set is used to make friend recommendations for the user in the next social recommendation cycle.

[0019] Optionally, in the above method embodiments of the present invention, the method further comprises:

[0020] In step 1, preprocessing the historical social data and the historical transaction data specifically includes:

[0021] The historical social data and the historical transaction data are subjected to data normalization processing and data dimension reduction processing based on principal component analysis.

[0022] Optionally, in the above method embodiments of the present invention, the method further comprises:

[0023] In step 2, the preset credit scoring model is a trained neural network model.

[0024] Optionally, in the above method embodiments of the present invention, the method further comprises:

[0025] In step 3, obtaining the social credit level of the user based on the social credit score of the user specifically includes:

[0026] Presets social credit levels, 1, 2, ..., Each social credit level corresponds to a social credit score range, and the social credit level of the user is obtained according to the social credit score range in which the credit score of the user is located. is a positive integer not less than 10.

[0027] Optionally, in the above method embodiments of the present invention, the method further comprises:

[0028] If the social credit rating of the user is greater than the social credit rating in the previous social recommendation cycle in P consecutive social recommendation cycles, the user is marked as a potential user, and in the P+1th social recommendation cycle, the adjusted social credit rating of the user is increased by 1 as the final social credit rating of the user;

[0029] If the social credit rating of the user is lower than the social credit rating of the previous social recommendation cycle in P consecutive social recommendation cycles, the user is marked as a warning user, and no friends are recommended to the user in the P+1th social recommendation cycle, and the user cannot actively add friends;

[0030] If the user's social credit level is N for P consecutive social recommendation cycles, the user is marked as a high-quality user and is displayed preferentially on the friend recommendation page and friend list;

[0031] If the social credit level of the user is 1 for P consecutive social recommendation cycles, the user will be marked as a dishonest user and will be prohibited from using social functions.

[0032] In a second aspect, an embodiment of the present invention provides a device for pushing personalized interpersonal relationships based on big data, the device comprising the following modules:

[0033] Data collection module: used to obtain the user's historical social data and historical transaction data within the current social recommendation cycle;

[0034] Data processing module: used for preprocessing the historical social data and the historical transaction data;

[0035] Personalization module: including a preset credit scoring model and a classification model, used to input the pre-processed historical social data and the historical transaction data into the preset credit scoring model, calculate the social credit score of the user, and obtain the social credit level of the user based on the social credit score of the user; also used to input the pre-processed historical social data and the historical transaction data into the preset classification model respectively, obtain the social tag and transaction tag of the user, and let the user set the interest tag by himself, and the social tag, the transaction tag and the interest tag constitute a tag set;

[0036] Personalized recommendation module: used to recommend friends to the current user according to the social credit level and the tag set corresponding to the user;

[0037] Feedback module: used to obtain feedback data of the user in the current social recommendation cycle, obtain an adjustment coefficient of the social credit level of the user based on the feedback data, and adjust the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient;

[0038] The obtaining an adjustment coefficient of the social credit level of the user based on the feedback data, and adjusting the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient, specifically includes:

[0039] The feedback data is quantified to obtain an adjustment coefficient ; The feedback data includes positive feedback data and negative feedback data;

[0040] In the next social recommendation cycle, according to the social credit level in the next social recommendation cycle and a new set of labels, and then adjust the coefficients according to the Get the user's final social credit rating in the next social recommendation cycle .

[0041] Optionally, in the above method embodiments of the present invention, the device further comprises:

[0042] The preset credit scoring model is a trained neural network model;

[0043] The preprocessing of the historical social data and the historical transaction data specifically includes:

[0044] The historical social data and the historical transaction data are subjected to data normalization processing and data dimension reduction processing based on principal component analysis.

[0045] In a third aspect, an embodiment of the present invention provides a personalized interpersonal relationship push system based on big data, and the system is used to implement the above-mentioned personalized interpersonal relationship push method based on big data.

[0046] In a fourth aspect, an embodiment of the present invention provides a personalized interpersonal relationship push device based on big data, the device comprising: a memory and a processor, the memory and the processor being coupled; the memory storing program instructions, and when the program instructions are executed by the processor, the device executes the above-mentioned personalized interpersonal relationship push method based on big data.

[0047] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium, including a computer program. When the computer program runs on an electronic device, the electronic device executes the above-mentioned personalized interpersonal relationship push method based on big data.

[0048] The present invention adopts a personalized interpersonal relationship push method based on big data, obtains users' historical social data and historical transaction data through big data, objectively and impartially divides users into social credit levels, and recommends friends to users through friend push strategies corresponding to social credit levels, thereby creating a safe and stable dating environment for users with high credit, and encouraging users with low credit to work hard to improve their credit levels, thereby improving the dating experience of the entire social platform.

[0049] The present invention displays the dynamic changes of the user's social credit level by setting a social recommendation cycle, and comprehensively, timely and accurately reflects the user's real credit.

[0050] The present invention obtains the social tags and transaction tags corresponding to the user through historical social data and historical transaction data, and the user presets interest tags. The objective and subjective methods are used to set user tags. The tag-matched friends recommended to the user based on the above multiple tags have the same or similar social behaviors as the user and have similar interests with the user, which fully guarantees the user's social experience.

[0051] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and other purposes, features and advantages of the present invention will become more apparent by describing the embodiments of the present invention in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0053] Figure 1 It is a flowchart of a method provided by an exemplary embodiment of the present invention.

[0054] Figure 2 It is a schematic diagram of the structure of a device provided by an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0055] Below, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described here.

[0056] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0057] Those skilled in the art can understand that the terms "first" and "second" in the embodiments of the present invention are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor indicate the necessary logical order between them.

[0058] It should also be understood that, in the embodiments of the present invention, “plurality” may refer to two or more than two, and “at least one” may refer to one, two or more than two.

[0059] It should also be understood that any component, data or structure mentioned in the embodiments of the present invention can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0060] Embodiment 1

[0061] like Figure 1 As shown, this embodiment provides a personalized interpersonal relationship push method based on big data, including the following steps:

[0062] S1: Preset a fixed time interval as a social recommendation cycle, obtain the user's historical social data and historical transaction data within the current social recommendation cycle, and pre-process the historical social data and the historical transaction data. Among them, the user's historical social data refers to data related to the user's social behavior, such as the user's activity on the social platform, the number of posts, the number of friends, the community influence, the number of complaints from other users, etc., which can be used to characterize the user's social behavior. The user's historical transaction data refers to data related to the user's transaction behavior, such as the user's historical consumption amount, transaction frequency, credit data, etc., which can be used to characterize the user's transaction behavior. The above historical social data and historical transaction data can be obtained from various social platforms and transaction websites by user ID.

[0063] Data preprocessing is due to the different dimensions and numerical ranges of data from different sources. Data of different dimensions need to be normalized to the same level. Otherwise, some factors will have too much influence, which will affect the evaluation results. The methods for preprocessing users' historical social data and historical transaction data mainly include data normalization and data dimensionality reduction based on principal component analysis.

[0064] Data normalization refers to scaling the data so that it falls into a small specific interval. Since the units of measurement of each feature in the feature vector are different, in order to be able to participate in the evaluation calculation, the indicator needs to be normalized and its value mapped to a certain value interval through function transformation.

[0065] Principal components analysis (PCA) is a technique used to explore the structure of high-dimensional data. When extracting and processing features, problems involving high-dimensional feature vectors are often prone to the curse of dimensionality and the problem of strong feature correlation. As the dimension of the data set increases, the number of samples required for algorithm learning increases exponentially. In some applications, encountering such big data is very disadvantageous, and learning from large data sets requires more memory and processing power. In addition, as the dimension increases, the sparsity of the data increases. It is more difficult to explore the same data set in a high-dimensional vector space than in an equally sparse data set. PCA is often used for the exploration and visualization of high-dimensional data sets. It can also be used for data compression, data preprocessing, etc. PCA can also synthesize high-dimensional variables that may be correlated into linearly independent low-dimensional variables, called principal components. The new low-dimensional data set will retain the variables of the original data as much as possible. PCA projects the data into a low-dimensional subspace to achieve dimensionality reduction. For example, the dimensionality reduction of a two-dimensional data set is to project the points into a line, and each sample of the data set can be represented by one value, not two values. A three-dimensional data set can be reduced to two dimensions, that is, mapping the variables into a plane. In general, high-dimensional correlated data can be reduced to low-dimensional non-correlated data through mapping.

[0066] In addition, since the user's social behavior and transaction behavior are continuous, the user's credit rating may also continue to change. A fixed time interval can be preset as a social recommendation cycle. During the current social recommendation cycle, the user's credit score is calculated based on the acquired historical data. A social recommendation cycle can be one week or one month, which is not specifically limited here.

[0067] S2: Inputting the pre-processed historical social data and the historical transaction data into a preset credit scoring model to calculate the social credit score of the user.

[0068] The preset credit scoring model can be obtained by training a common neural network model based on the relevant data of multiple sample users and the labeled credit scores of the multiple sample users. The specific implementation method can be similar to that in the related technology and will not be repeated here.

[0069] S3: Obtaining the social credit level of the user based on the social credit score of the user.

[0070] N social credit levels are preset, namely 1, 2, ..., N, each social credit level corresponds to a social credit score range, and the social credit level of the user is obtained according to the social credit score range in which the social credit score of the user falls, and N is a positive integer not less than 10.

[0071] For example, the user's social credit score can be set to 1-100, the user's social credit level can be set to 1-10, the social credit score corresponding to level 1 is 1-10, the social credit score corresponding to level 2 is 11-20, and so on. The higher the social credit level, the better the credit of the user, and he is a high-quality user on the social platform. The lower the social credit level, the worse the credit of the user, and there is a certain risk in social behavior.

[0072] S4: inputting the preprocessed historical social data and the historical transaction data into a preset classification model respectively to obtain the social tag and transaction tag of the user, and allowing the user to set interest tags by himself, and the social tags, the transaction tags and the interest tags constitute a tag set;

[0073] In the related art, it is usually to use big data technology to input the information related to the user into the classification model to obtain the label category that can characterize the characteristics of the user group, and recommend friends to the user based on the obtained label category. The disadvantage of this is that the friends recommended to the user are all friends with similar behaviors to the user, and the user cannot contact people of other categories; and recommending friends to the user through the labels set by the user himself may cause the user to meet friends with the same interests and hobbies as himself, but different values ​​or consumption habits, which will also bring a bad friendship experience to the user. In this application, by inputting the historical social data and the historical transaction data into the preset classification model respectively, the social label corresponding to the user's social behavior and the transaction label corresponding to the user's transaction behavior are obtained, which are used as objective labels, and then the user sets the interest label by himself, which is used as a subjective label, and the label set is obtained by combining the objective label and the subjective label, so as to recommend friends to the user, which can not only meet the user's interests, but also be in sync with the user, thereby improving the user's friendship experience. The interest labels set by the user can be movies, music, pets, star chasing, art, reading, etc.

[0074] The preset classification model can be obtained by training a common neural network model based on the relevant data of multiple sample users and the labels annotated by the multiple sample users. The specific implementation method can be similar to that in the related technology and will not be repeated here.

[0075] S5: Recommending friends to the current user according to the social credit level and the tag set corresponding to the user.

[0076] Different friend push strategies can be set for different social credit levels. For users with higher social credit levels, friends with similarly high credit levels can be recommended to them, providing them with a safe social environment. For users with lower social credit levels, their friendship behaviors on social platforms can be restricted, thereby motivating users to work hard to improve their social credit levels.

[0077] Assume that the user's social credit level is , ,

[0078] like , recommending the social credit level to the user and X users whose tag set is exactly the same as that of the user;

[0079] like , recommending the social credit level to the user and Y users whose tag set is exactly the same as that of the user;

[0080] X and Y are both preset natural numbers greater than 1, and Y>X.

[0081] In addition, when recommending friends for a user based on the user's social credit level and tag set, if the number of qualified friends to be recommended is insufficient, we can consider recommending users who meet the social credit level requirements, whose interest tags match the user, and whose social tags or transaction tags match the user as friends to be recommended.

[0082] S6: Acquire feedback data of the user in the current social recommendation cycle, obtain an adjustment coefficient of the social credit level of the user based on the feedback data, and adjust the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient.

[0083] Specifically, the user's feedback data in the current social recommendation cycle is used to characterize the user's social behavior in the current social recommendation cycle, and may include positive feedback data, such as the number of friends actively added by the user in the current social recommendation cycle, the number of friends passively added, the average daily online time, the proportion of users with high social credit levels among the new friends, etc. It may also include negative feedback data, such as the number of times the user has been deleted or reported by friends in the current social recommendation cycle.

[0084] The method of obtaining the adjustment coefficient of the user's social credit level based on the feedback data is not specifically limited here. One method may use the integral value. For example, the preset adjustment coefficient is =1, when the number of friends actively added by the user in the current social recommendation cycle is greater than 10, the adjustment coefficient increases by 0.2, when the user's average daily online time meets the preset time, the adjustment coefficient increases by 0.2, and when the user is reported once in the current social recommendation cycle, the adjustment coefficient decreases by 0.2, that is, positive feedback data will cause the adjustment coefficient to become larger, while negative feedback data will cause the adjustment coefficient to become smaller. Finally, the adjustment coefficient in the current social recommendation cycle is obtained, which is used to adjust the social credit level in the next social recommendation cycle.

[0085] In the next social recommendation cycle, first repeat S1-S5 to obtain the social credit level in the next social recommendation cycle. and the new tag set, and then obtain the user's final social credit level in the next social recommendation cycle based on the adjustment coefficient , based on the The new tag set is used to make friend recommendations for the user in the next social recommendation cycle.

[0086] Since the user's feedback data includes positive feedback data and negative feedback data, the adjustment coefficient may be greater than 1 or less than 1. The final social credit level after adjustment may be lower than the calculated social credit level or higher than the calculated social credit level, thereby motivating users to restrain their social behavior and avoid a continuous decline in their social credit level.

[0087] Furthermore, users may be marked according to changes in their social credit ratings over multiple consecutive social recommendation cycles.

[0088] If the user is in a continuous If the social credit level in each of the social recommendation cycles is greater than the social credit level in the previous social recommendation cycle, the user is marked as a potential user and When the social recommendation cycle is reached, the adjusted social credit level of the user is increased by 1 as the final social credit level of the user;

[0089] If the user is in a continuous If, in each of the social recommendation cycles, the social credit rating is lower than the social credit rating in the previous social recommendation cycle, the user is marked as a warning user and During the social recommendation period, no friends will be recommended to the user, and the user cannot actively add friends;

[0090] If the user is in a continuous During the social recommendation period, the social credit level is , the user is marked as a high-quality user and is displayed first on the friend recommendation page and friend list;

[0091] If the user is in a continuous If the social credit level is 1 during each of the social recommendation cycles, the user will be marked as a dishonest user and will be prohibited from using social functions.

[0092] By adopting the above-mentioned marking method, it is possible to mine high-quality users and dishonest users from a large number of users, and highlight high-quality users on the friend recommendation page, which can encourage other users to actively add high-quality users, so that under the influence and influence of high-quality users, they can continuously improve their social credit rating and further optimize the social environment. In addition, prohibiting dishonest users from using social functions can not only prevent dishonest users from engaging in improper behavior in social interactions with other users and causing losses to other users, but also motivate dishonest users to work hard to improve their social credit rating and further optimize the social environment.

[0093] Embodiment 2

[0094] like Figure 2 As shown, this embodiment provides a personalized interpersonal relationship push device based on big data, which can be used to implement the above-mentioned personalized interpersonal relationship push method based on big data.

[0095] Includes the following modules:

[0096] Data collection module: used to obtain the user's historical social data and historical transaction data within the current social recommendation cycle;

[0097] Data processing module: used for preprocessing the historical social data and the historical transaction data;

[0098] Personalization module: including a preset credit scoring model and a classification model, used to input the pre-processed historical social data and the historical transaction data into the preset credit scoring model, calculate the social credit score of the user, and obtain the social credit level of the user based on the social credit score of the user; also used to input the pre-processed historical social data and the historical transaction data into the preset classification model respectively, obtain the social tag and transaction tag of the user, and let the user set the interest tag by himself, and the social tag, the transaction tag and the interest tag constitute a tag set;

[0099] Personalized recommendation module: used to recommend friends to the current user according to the social credit level and the tag set corresponding to the user;

[0100] Feedback module: used to obtain feedback data of the user in the current social recommendation cycle, obtain an adjustment coefficient of the social credit level of the user based on the feedback data, and adjust the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient.

[0101] A user's historical social data refers to data related to the user's social behavior, such as the user's activity on the social platform, number of posts, number of friends, community influence, number of complaints from other users, etc., which can be used to characterize the user's social behavior. A user's historical transaction data refers to data related to the user's transaction behavior, such as the user's historical consumption amount, transaction frequency, credit data, etc., which can be used to characterize the user's transaction behavior. The above-mentioned historical social data and historical transaction data can be obtained from various social platforms and transaction websites using the user ID.

[0102] Data preprocessing is due to the different dimensions and numerical ranges of data from different sources. Data of different dimensions need to be normalized to the same level. Otherwise, some factors will have too much influence, which will affect the evaluation results. The methods for preprocessing users' historical social data and historical transaction data mainly include data normalization and data dimensionality reduction based on principal component analysis.

[0103] Data normalization refers to scaling the data so that it falls into a small specific interval. Since the units of measurement of each feature in the feature vector are different, in order to be able to participate in the evaluation calculation, the indicator needs to be normalized and its value mapped to a certain value interval through function transformation.

[0104] Principal components analysis (PCA) is a technique used to explore the structure of high-dimensional data. When extracting and processing features, problems involving high-dimensional feature vectors are often prone to the curse of dimensionality and the problem of strong feature correlation. As the dimension of the data set increases, the number of samples required for algorithm learning increases exponentially. In some applications, encountering such big data is very disadvantageous, and learning from large data sets requires more memory and processing power. In addition, as the dimension increases, the sparsity of the data increases. It is more difficult to explore the same data set in a high-dimensional vector space than in an equally sparse data set. PCA is often used for the exploration and visualization of high-dimensional data sets. It can also be used for data compression, data preprocessing, etc. PCA can also synthesize high-dimensional variables that may be correlated into linearly independent low-dimensional variables, called principal components. The new low-dimensional data set will retain the variables of the original data as much as possible. PCA projects the data into a low-dimensional subspace to achieve dimensionality reduction. For example, the dimensionality reduction of a two-dimensional data set is to project the points into a line, and each sample of the data set can be represented by one value, not two values. A three-dimensional data set can be reduced to two dimensions, that is, mapping the variables into a plane. In general, high-dimensional correlated data can be reduced to low-dimensional non-correlated data through mapping.

[0105] In addition, since the user's social behavior and transaction behavior are continuous, the user's credit rating may also continue to change. A fixed time interval can be preset as a social recommendation cycle. During the current social recommendation cycle, the user's credit score is calculated based on the acquired historical data. A social recommendation cycle can be one week or one month, which is not specifically limited here.

[0106] The preset credit scoring model can be obtained by training a common neural network model based on the relevant data of multiple sample users and the labeled credit scores of the multiple sample users. The specific implementation method can be similar to that in the related technology and will not be repeated here.

[0107] Furthermore, the user's feedback data in the current social recommendation cycle is used to characterize the user's social behavior in the current social recommendation cycle, and may include positive feedback data, such as the number of friends actively added by the user in the current social recommendation cycle, the number of friends passively added, the average daily online time, the proportion of users with high social credit levels among the new friends, etc., and may also include negative feedback data, such as the number of times the user has been deleted by friends and the number of times the user has been reported by friends in the current social recommendation cycle, etc.

[0108] The method of obtaining the adjustment coefficient of the user's social credit level based on the feedback data is not specifically limited here. One method may use the integral value. For example, the preset adjustment coefficient is =1, when the number of friends actively added by the user in the current social recommendation cycle is greater than 10, the adjustment coefficient increases by 0.2, when the user's average daily online time meets the preset time, the adjustment coefficient increases by 0.2, and when the user is reported once in the current social recommendation cycle, the adjustment coefficient decreases by 0.2, that is, positive feedback data will cause the adjustment coefficient to become larger, while negative feedback data will cause the adjustment coefficient to become smaller. Finally, the adjustment coefficient in the current social recommendation cycle is obtained, which is used to adjust the social credit level in the next social recommendation cycle.

[0109] In the next social recommendation cycle, firstly, the social credit level in the next social recommendation cycle is obtained by using the device and the new tag set, and then obtain the user's final social credit level in the next social recommendation cycle based on the adjustment coefficient , so that the device can be based on the The new tag set is used to make friend recommendations for the user in the next social recommendation cycle.

[0110] Since the user's feedback data includes positive feedback data and negative feedback data, the adjustment coefficient may be greater than 1 or less than 1. The final social credit level after adjustment may be lower than the calculated social credit level or higher than the calculated social credit level, thereby motivating users to restrain their social behavior and avoid a continuous decline in their social credit level.

[0111] Embodiment 3

[0112] This embodiment provides a personalized interpersonal relationship push system based on big data, which is used to implement the above-mentioned personalized interpersonal relationship push method based on big data.

[0113] Embodiment 4

[0114] This embodiment provides a personalized interpersonal relationship push device based on big data, and the device includes: a memory and a processor, and the memory and the processor are coupled; the memory stores program instructions, and when the program instructions are executed by the processor, the device executes the above-mentioned personalized interpersonal relationship push method based on big data.

[0115] Embodiment 5

[0116] This embodiment provides a computer-readable storage medium, including a computer program. When the computer program is executed on an electronic device, the electronic device executes the above-mentioned personalized interpersonal relationship push method based on big data.

[0117] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0118] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0119] The block diagrams of the devices, apparatuses, equipment, and systems involved in this disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including," "comprising," "having," and the like are open words, referring to "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or," and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0120] The method and apparatus of the present disclosure may be implemented in many ways. For example, the method and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0121] It should also be noted that in the apparatus, equipment and method of the present disclosure, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present disclosure. The above description of the disclosed aspects is provided to enable any technician in the field to make or use the present disclosure. Various modifications to these aspects are very obvious to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown here, but to the widest scope consistent with the principles and novel features disclosed herein.

[0122] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A personalized interpersonal relationship push method based on big data, characterized in that: The steps include: Step 1: Preset a fixed time interval as a social recommendation cycle, obtain the user's historical social data and historical transaction data within the current social recommendation cycle, and pre-process the historical social data and the historical transaction data; Step 2: Input the pre-processed historical social data and the historical transaction data into a preset credit scoring model to calculate the social credit score of the user; Step 3: Obtaining the social credit level of the user based on the social credit score of the user; Step 4: Input the preprocessed historical social data and the historical transaction data into a preset classification model respectively to obtain the social tag and transaction tag of the user, and let the user set the interest tag by himself, and the social tag, the transaction tag and the interest tag constitute a tag set; Step 5: Recommending friends to the current user according to the social credit level and the tag set corresponding to the user; Step 6: Acquire feedback data of the user in the current social recommendation cycle, obtain an adjustment coefficient of the social credit level of the user based on the feedback data, and adjust the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient; In step 5, making friend recommendations for the current user according to the social credit level and the tag set corresponding to the user specifically includes: Assume that the user's social credit level is M, , like , recommending the social credit level to the user and X users whose tag set is exactly the same as that of the user; like , recommending the social credit level to the user and Y users whose tag set is exactly the same as that of the user; X and Y are both preset natural numbers greater than 1, and Y>X; In step 6, obtaining an adjustment coefficient of the social credit level of the user based on the feedback data, and adjusting the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient, specifically includes: The feedback data is quantified to obtain an adjustment coefficient ; The feedback data includes positive feedback data and negative feedback data; In the next social recommendation cycle, repeat steps 1 to 4 to obtain the social credit rating in the next social recommendation cycle. and a new set of labels, and then adjust the coefficients according to the Get the user's final social credit rating in the next social recommendation cycle , based on the The new tag set is used to make friend recommendations for the user in the next social recommendation cycle.

2. The method according to claim 1, characterized in that In step 1, preprocessing the historical social data and the historical transaction data specifically includes: The historical social data and the historical transaction data are subjected to data normalization processing and data dimension reduction processing based on principal component analysis.

3. The method according to claim 2, characterized in that In step 2, the preset credit scoring model is a trained neural network model.

4. The method according to claim 3, characterized in that In step 3, obtaining the social credit level of the user based on the social credit score of the user specifically includes: Presets social credit levels, 1, 2, ..., Each social credit level corresponds to a social credit score range, and the social credit level of the user is obtained according to the social credit score range in which the credit score of the user is located. is a positive integer not less than 10.

5. The method according to claim 4, characterized in that The method further comprises, If the social credit rating of the user is greater than the social credit rating in the previous social recommendation cycle in P consecutive social recommendation cycles, the user is marked as a potential user, and in the P+1th social recommendation cycle, the adjusted social credit rating of the user is increased by 1 as the final social credit rating of the user; If the social credit rating of the user is lower than the social credit rating of the previous social recommendation cycle in P consecutive social recommendation cycles, the user is marked as a warning user, and no friends are recommended to the user in the P+1th social recommendation cycle, and the user cannot actively add friends; If the user's social credit level is N for P consecutive social recommendation cycles, the user is marked as a high-quality user and is displayed preferentially on the friend recommendation page and friend list; If the social credit level of the user is 1 for P consecutive social recommendation cycles, the user will be marked as a dishonest user and will be prohibited from using social functions.

6. A personalized interpersonal relationship push device based on big data, characterized in that: Includes the following modules: Data collection module: used to obtain the user's historical social data and historical transaction data within the current social recommendation cycle; Data processing module: used for preprocessing the historical social data and the historical transaction data; Personalization module: including a preset credit scoring model and a classification model, used to input the pre-processed historical social data and the historical transaction data into the preset credit scoring model, calculate the social credit score of the user, and obtain the social credit level of the user based on the social credit score of the user; It is also used to input the pre-processed historical social data and the historical transaction data into a preset classification model respectively, to obtain the social tag and transaction tag of the user, and to allow the user to set the interest tag by himself, and to form a tag set with the social tag, the transaction tag and the interest tag; Personalized recommendation module: used to recommend friends to the current user according to the social credit level and the tag set corresponding to the user. The personalized recommendation module is used to assume that the social credit level of the user is M. ,like , recommending the social credit level to the user and X users whose tag set is exactly the same as that of the user; like , recommending the social credit level to the user and Y users whose tag set is exactly the same as that of the user; X and Y are both preset natural numbers greater than 1, and Y>X; Feedback module: used to obtain feedback data of the user in the current social recommendation cycle, obtain an adjustment coefficient of the social credit level of the user based on the feedback data, and adjust the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient; The obtaining an adjustment coefficient of the social credit level of the user based on the feedback data, and adjusting the social credit level of the user in the next social recommendation cycle based on the adjustment coefficient, specifically includes: The feedback data is quantified to obtain an adjustment coefficient ; The feedback data includes positive feedback data and negative feedback data; In the next social recommendation cycle, according to the social credit level in the next social recommendation cycle and a new set of labels, and then adjust the coefficients according to the Get the user's final social credit rating in the next social recommendation cycle .

7. The device according to claim 6, characterized in that The preset credit scoring model is a trained neural network model; The preprocessing of the historical social data and the historical transaction data specifically includes: The historical social data and the historical transaction data are subjected to data normalization processing and data dimension reduction processing based on principal component analysis.

8. A personalized interpersonal relationship push system based on big data, characterized in that: The system is used to implement the method according to any one of claims 1 to 5.

9. An electronic device, characterized in that: The electronic device comprises: a memory and a processor, wherein the memory and the processor are coupled; the memory stores program instructions, and when the program instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The method comprises a computer program, which, when executed on an electronic device, enables the electronic device to execute the method according to any one of claims 1 to 5.

Images