A health consultation information push system and method based on points of interest

Abstract

The present invention discloses a system and method for pushing health consultation information based on points of interest. The method includes: Step S100: Classifying each user into different populations to be mined; respectively aggregating the points of interest obtained from the users within different populations to be mined into different point-of-interest sets; Step S200: Setting the status of the points of interest of the users within different populations to be mined or calculating the penetration rate Q of the points of interest of the users within different populations to be mined; Step S300: Conducting discriminant analysis and processing on the implicit and explicit points of interest of the users within different populations to be mined; Step S400: Based on Steps S100 - S300, obtaining the explicit and implicit points of interest of each user; calculating the similarity values between the respective implicit points of interest of the users with the same explicit point of interest, and establishing an interest association relationship among the users; and mutually pushing the health consultation information included in the respective implicit points of interest of the users with an interest association relationship.

Description

Technical Field

[0001] The present invention relates to the technical field of health consultation information processing, and in particular to a health consultation information push system and method based on points of interest. Background Art

[0002] In recent years, as people pay more and more attention to health issues, health consultation management services are also needed by more and more people; using big data to push relevant health consultation information to specific groups of people can largely solve user needs. Because in the process of information push, if relevant information can be pushed based on the user's interests, it can take into account both effectiveness and accuracy, and at the same time improve the efficiency of information push. Therefore, it is particularly important to fully explore the user's interests and how to accurately push information to the user based on the mined interests. Summary of the invention

[0003] The purpose of the present invention is to provide a health consultation information push system and method based on points of interest to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for pushing health consultation information based on points of interest, the pushing method comprising:

[0005] Step S100: Based on the determination and analysis of the age, gender and work information of each user, each user is classified into different groups to be mined; the groups to be mined include a first group to be mined and a second group to be mined; the second group to be mined is a group to be mined obtained by further dividing the first group to be mined; the points of interest obtained from users in different groups to be mined are respectively aggregated into different interest point sets, and the interest point sets include a first interest point set and a second interest point set;

[0006] Step S200: setting the status of the interest points of users in different to-be-mined groups or calculating the popularity rate Q of the interest points of users in different to-be-mined groups based on the complete overlap relationship or partial overlap relationship of the interest point sets between different to-be-mined groups, and obtaining a third interest point set based on the calculated popularity rate Q of the interest points; one type of health problem corresponds to one interest point;

[0007] Step S300: using the interest points in the third interest point set and the interest points in the first interest point set and the second interest point set that partially overlap with each other to discriminate and analyze the implicit interest points and explicit interest points of users in different groups to be mined;

[0008] Step S400: Based on Steps S100 - S300, obtain the explicit interest points and implicit interest points of each user; calculate the similarity values between the implicit interest points of users with the same explicit interest point, and when the similarity value is greater than the similarity value threshold, establish an interest association relationship between the users; push the health consultation information included in the implicit interest points of users with an interest association relationship to each other; one category of health problems corresponds to one implicit interest point, and one implicit interest point contains health consultation information associated with this interest point.

[0009] Further, Step S100 includes:

[0010] Step S101: Obtain the age and gender of the user; set several users with the same age, the same gender, or an age difference less than the age difference threshold and the same gender as the first group of users to be mined; respectively accumulate the consultation quantities of various categories of health problems for the first group of users to be mined; sort the consultation quantities corresponding to various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity rankings as the first interest point set of the users included in the first group of users to be mined; one category of health problems corresponds to one implicit interest point;

[0011] Step S102: Obtain the work information of each user in the first group of users to be mined, and obtain and determine the daily working hours, office form, and office mode in the user's work information; the office form includes, but is not limited to, sedentary office work and mobile office work; if the proportion of the time the user sits at work is greater than the proportion threshold, determine that the user's office form is sedentary office work; where, t 1 represents the time the user is in a sedentary state within the total daily working hours, and T represents the total daily working hours of the user; if the proportion of the time the user moves around at work is greater than the proportion threshold, determine that the user's office form is mobile office work; where, t 2 represents the time the user is in a moving state within the total daily working hours, and T represents the total daily working hours of the user; among them, take the office content with the highest repetition rate within the total daily working hours of the user as the user's office mode; set several users with the same daily working hours, office form, and office mode in the first group of users to be mined as the second group of users to be mined, and divide the first group of users to be mined into several second groups of users to be mined; respectively accumulate the consultation quantities of various categories of health problems for several second groups of users to be mined, sort the consultation quantities of various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity rankings as the second interest point set of the users included in the second group of users to be mined;

[0012] In the above steps, the age and gender of the user are taken as the major common characteristics of a group of people, and the number of consultations on various categories of health problems by people with these major common characteristics is obtained. In this application, the number of consultations is used as the data reflecting the interest of the group of people in a certain category of health problems; the work information of the people in the group with major common characteristics is analyzed and determined. In order to use the work information as the characteristics of each user in the group with major common characteristics, it is equivalent to mining the individual characteristics existing in the group of people with major common characteristics. Taking the work information of each user as the individual characteristics is conducive to enhancing the connection with the identity characteristics of each user when analyzing the interest points of different groups of people. Because in practical applications, health problems are often accompanied by people's actual work, starting from people's work information can make the conclusions obtained when analyzing users' interest points in health more reference-worthy and accurate at the same time.

[0013] Further, the process of step S200 for setting the status of the interest points of users in different groups to be mined or calculating the penetration rate Q of the interest points of users in different groups to be mined includes:

[0014] Step S201: When all the several categories of health problems included in the first interest point set and the second interest point set completely overlap, determine the several categories of health problems included in the first interest point set or the second interest point set as the explicit interest points of the users included in the first group to be mined;

[0015] Step S202: When some of the several categories of health problems included in the first interest point set and the second interest point set overlap, respectively extract the overlapping several categories of health problems and the non-overlapping several categories of health problems; trace the user IDs of each consultation record corresponding to the non-overlapping several categories of health problems one by one; several consultation records correspond to several numbers of consultations;

[0016] Step S203: Calculate the penetration rate Q of the non-overlapping several categories of health problems respectively. The formula is:

[0017]

[0018] where P i represents the total number of consultations on a certain category of health problems in the non-overlapping several categories of health problems by the i-th user ID; n represents the total number of user IDs obtained by tracing the user IDs for a certain category of health problems;

[0019] Step S204: Sort the penetration rates Q of the non-overlapping several categories of health problems from high to low, and select the top three categories of health problems with the sorted penetration rates Q to form the third interest point set;

[0020] If the interest sets obtained based on common characteristics and the interest sets obtained based on individual characteristics completely overlap, it means that most of this group of people have the same interests in health issues. If the interest sets obtained based on common characteristics and the interest sets obtained based on individual characteristics do not completely overlap, the overlapping part means that the users have the same interests in health issues; the non-overlapping part means that the users have different interests in health issues. The data concept of penetration rate Q is proposed to see whether the interests corresponding to the non-overlapping part are universal. The higher the penetration rate Q, the wider the popularity of this interest among the people with large common characteristics, that is, this personality trait is generally present in this group of people. In this case, it is possible that the obtained interests have common characteristics and universality in terms of data meaning, and this is also to achieve the status correction of this interest in the subsequent steps.

[0021] Further, step S300 includes:

[0022] When there is one category of health problems that overlap and two categories of health problems that are different among several categories of health problems included in the first interest point set and the second interest point set, and the overlapping category of health problems belongs to the category of health problems included in the third interest point set, determine the several categories of health problems included in the third interest point set as the explicit interest points of the users included in the first population to be mined, and determine the two different categories of health problems as the implicit interest points of the users included in the second population to be mined; when there is one category of health problems that overlap and two categories of health problems that are different among several categories of health problems included in the first interest point set and the second interest point set, and the overlapping category of health problems does not belong to the category of health problems included in the third interest point set, determine the first two categories of health problems in the third interest point set together with the overlapping category of health problems as the explicit interest points of the users included in the first population to be mined, and determine the two different categories of health problems as the implicit interest points of the users included in the second population to be mined; one category of health problems corresponds to one interest point;

[0023] When there are two types of health problems that overlap between the first set of points of interest and the second set of points of interest, one type of health problem is different, and both of the overlapping types of health problems belong to the types of health problems in the third set of points of interest, then determine the several types of health problems included in the third set of points of interest as the explicit interests of the users included in the first population to be mined, and determine the different type of health problem as the implicit interest of the users included in the second population to be mined; when there are two types of health problems that overlap between the first set of points of interest and the second set of points of interest, one type of health problem is different, and both of the overlapping types of health problems do not belong to the types of health problems in the third set of points of interest, then determine the first type of health problem in the third set of points of interest and the two overlapping types of health problems together as the explicit interests of the users included in the first population to be mined, and determine the different type of health problem as the implicit interest of the users included in the second population to be mined;

[0024] When there are two types of health problems that overlap between the first set of points of interest and the second set of points of interest, one type of health problem is different, and only one of the two overlapping types of health problems belongs to the types of health problems in the third set of points of interest, then sort the types of health problems in the third set of points of interest other than the type of health problem that belongs to the third set of points of interest among the two overlapping types of health problems according to the penetration rate Q, and select the type of health problem with the larger penetration rate Q among the two and the two overlapping types of health problems together as the explicit interests of the users included in the first population to be mined, and determine the different type of health problem as the implicit interest of the users included in the second population to be mined;

[0025] The above states of the points of interest, namely explicit and implicit, are obtained by comprehensively considering the overlapping situations of the points of interest in different sets of points of interest and their respective penetration rate values. The above steps are equivalent to steps for correcting the states of the relevant points of interest.

[0026] Furthermore, the process of calculating the similarity values between the respective implicit interests of users with the same explicit interest in step S400 includes:

[0027] Step S401: Obtain the implicit interests of users with the same explicit interest to obtain their respective corresponding implicit interest sets, and traverse the number of hidden interests in the implicit interest sets of each user; one by one, set the users with the same explicit interest as the users whose association relationship is to be mined, and set the other users whose difference from the number of hidden interests in the implicit interest set of the user whose association relationship is to be mined is less than the difference threshold as the users to be matched and associated with the user whose association relationship is to be mined;

[0028] Step S402: Calculate the association value W between the user whose association relationship is to be mined and each associated user to be matched, using the formula:

[0029]

[0030] Among them, A represents the set A of all hidden interest points of the users whose associations are to be mined; B represents the set B of all hidden interest points of the users whose associations are to be mined; crad(A∩B) represents the number of hidden interest points of the intersection of set A and set B; crad(A∪B) represents the number of hidden interest points of the union of set A and set B;

[0031] Step S403: establishing an interest association relationship between two users whose association value W is greater than an association threshold;

[0032] The above steps analyze and determine the association relationship between different users, and based on the association relationship, push additional information to users, push some information that their associated users are interested in, and that the users themselves may also be interested in, thereby improving the intelligence of the information push process.

[0033] In order to better implement the above method, a push system for health consultation information push method based on points of interest is also proposed, and the push system includes: a user classification module, a point of interest processing module, a point of interest state processing module, a calculation module, a point of interest state judgment module, an interest association relationship analysis module, and a health consultation information push module;

[0034] The user classification module is used to classify each user into different groups to be mined according to the age, gender, and work information of each user. The groups to be mined include the first group to be mined and the second group to be mined. The second group to be mined is the group to be mined obtained by further dividing the first group to be mined.

[0035] An interest point processing module, used to aggregate interest points obtained from users in different to-be-mined groups into different interest point sets, wherein the interest point sets include a first interest point set and a second interest point set;

[0036] The interest point state processing module is used to make the interest points of users in different to-be-mined groups with completely overlapping relationships explicit;

[0037] A calculation module, used to calculate the popularity rate Q of the interest points of users in different to-be-mined groups that partially overlap with the first interest point set and the second interest point set;

[0038] An interest point status judgment module, configured to receive the penetration rate Q in the calculation module, and obtain a third interest point set based on the penetration rate Q; use the interest points in the third interest point set and the interest points in the first interest point set and the second interest point set with partial overlap relationships to perform discriminant analysis on the implicit and explicit interest points of users in different populations to be mined;

[0039] An interest association relationship analysis module, configured to calculate the similarity values between the implicit interest points of each user with the same explicit interest point. When the similarity value is greater than the similarity value threshold, an interest association relationship is established between each user;

[0040] A health consultation information push module, configured to receive the data in the interest association relationship analysis module, and push the health consultation information included in the respective implicit interest points of each user with an interest association relationship to each other.

[0041] Further, the user division module includes a first population to be mined division unit, a second population to be mined division unit, a first interest point set processing unit, and a second interest point set processing unit;

[0042] The first population to be mined division unit is configured to obtain the age and gender of the user; set several users with the same age and gender or an age difference less than the age difference threshold and the same gender as the first population to be mined;

[0043] The second population to be mined division unit is configured to set several users with the same daily working hours, office form, and office mode within the first population to be mined as the second population to be mined;

[0044] A first interest point set processing unit is configured to respectively accumulate the consultation quantities of various categories of health problems by the first population to be mined; sort the consultation quantities corresponding to various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity as the first interest point set of the users included in the first population to be mined;

[0045] The second interest point set processing unit is configured to respectively accumulate the consultation quantities of various categories of health problems by several second populations to be mined, sort the consultation quantities of various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity as the second interest point set of the users included in the second population to be mined.

[0046] Further, the interest point status judgment module includes: a third interest point set processing unit, an interest point status judgment unit, and an interest point status setting unit;

[0047] The third interest point set processing unit is configured to receive the penetration rate Q in the calculation module, and obtain a third interest point set based on the penetration rate Q;

[0048] An interest point status judgment unit is used to distinguish and analyze the implicit and explicit interest points of users in different populations to be mined by comparing the interest points in the third interest point set with those in the first and second interest point sets with a partial overlap relationship.

[0049] An interest point status setting unit is used to receive the data from the interest point status judgment unit and set the implicit or explicit status of the interest points of users in different populations to be mined.

[0050] Furthermore, the interest association relationship analysis module includes: a user to be mined for association relationship setting unit, a user to be matched for association relationship setting unit, an association value calculation unit, and an interest association relationship establishment unit;

[0051] The user to be mined for association relationship setting unit is used to obtain users with the same explicit interest points and set them as users to be mined for association relationships;

[0052] The user to be matched for association relationship setting unit is used to obtain other users whose difference in the number of hidden interest points in the implicit interest point set from the users to be mined for association relationships is less than the difference threshold, and set them as users to be matched for association relationships;

[0053] The association value calculation unit is used to receive the information from the user to be mined for association relationship setting unit and the user to be matched for association relationship setting unit and calculate the association value;

[0054] The interest association relationship establishment unit is used to receive the association value result from the association value calculation unit and establish an interest association relationship between two users whose association value is greater than the association threshold.

[0055] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: The method of dividing the population of users in the present invention first makes a primary division based on the common characteristics among different ranges of populations, and then makes a secondary division of the individual characteristics among the sub-ranges of populations obtained after the primary division. In the secondary division process, by extracting the characteristic information from the work information of each user, it can make the relevant information pushed have a high personal association with the users, improving the effectiveness and accuracy of the pushed information; The present application also analyzes and judges the association relationship between different users, and realizes the supplementary push of information to users based on the association relationship, improving the intelligence in the information push process. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The drawings are used to provide a further understanding 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 to the present invention. In the drawings:

[0057] Figure 1It is a schematic flowchart of a method for pushing health consultation information based on points of interest according to the present invention;

[0058] Figure 2 It is a schematic structural diagram of a system for pushing health consultation information based on points of interest according to the present invention. Specific embodiments

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0060] Please refer to Figure 1 - Figure 2 , the present invention provides a technical solution: a method for pushing health consultation information based on points of interest, and the pushing method includes:

[0061] Step S100: Classify each user into different populations to be mined based on the determination and analysis of the age, gender, and work information of each user; the populations to be mined include the first population to be mined and the second population to be mined; the second population to be mined is the population to be mined obtained by further dividing the first population to be mined; the points of interest obtained from the users in different populations to be mined are respectively compiled into different point-of-interest sets, and the point-of-interest sets include the first point-of-interest set and the second point-of-interest set;

[0062] Among them, step S100 includes:

[0063] Step S101: Obtain the age and gender of the user; set several users with the same age, the same gender, or an age difference less than the age difference threshold and the same gender as the first population to be mined; respectively accumulate the consultation quantities of various categories of health problems by the first population to be mined; sort the consultation quantities corresponding to various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity ranking as the first point-of-interest set of the users included in the first population to be mined; one category of health problem corresponds to one implicit point of interest;

[0064] For example, set all women aged 30 or 30 - 35 as the first population to be mined; accumulate which categories of health problems are concerned by all women aged 30 or 30 - 35, and accumulate the consultation quantities of these categories of health problems;

[0065] Step S102: Obtain the work information of each user in the first population to be mined, and obtain and determine the daily working hours, office form, and office mode of the user in the work information; the office form includes, but is not limited to, sedentary office and mobile office; if the proportion of the user's sitting office time Greater than the proportion threshold, it is determined that the user's work form is sedentary work; where t 1 represents the duration of the user sitting still within the total daily working hours, and T represents the total daily working hours of the user; if the proportion of the duration of the user's mobile work is greater than the proportion threshold, it is determined that the user's work form is mobile work; where t 2 represents the duration of the user moving within the total daily working hours, and T represents the total daily working hours of the user; among them, the work content with the highest repetition rate within the total daily working hours of the user is used as the user's work mode; several users with the same daily working hours, work form, and work mode within the first population to be mined are set as the second population to be mined, and the first population to be mined is divided into several second populations to be mined; respectively accumulate the consultation quantities of several second populations to be mined for various categories of health problems, sort the consultation quantities of various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity ranking as the second interest point set of the users included in the second population to be mined;

[0066] For example, gather the population with a daily working hours of 6 hours, a sedentary work form, and a work mode of repeatedly operating the computer among all women aged 30 or 30 - 35 years old, accumulate which categories of health problems are concerned by all women aged 30 or 30 - 35 years old, and accumulate the consultation quantities of these categories of health problems;

[0067] Step S200: Based on the complete overlap relationship or partial overlap relationship between the interest point sets of different populations to be mined, respectively set the status of the interest points of the users within different populations to be mined or calculate the penetration rate Q of the interest points of the users within different populations to be mined, and obtain the third interest point set based on the calculated penetration rate Q of the interest points; one category of health problem corresponds to one interest point;

[0068] Among them, the process of step S200 setting the status of the interest points of the users within different populations to be mined or calculating the penetration rate Q of the interest points of the users within different populations to be mined includes:

[0069] Step S201: When there is a complete overlap between several categories of health problems included in the first interest point set and the second interest point set, determine the several categories of health problems included in the first interest point set or the second interest point set as the explicit interest points of the users included in the first population to be mined;

[0070] For example, if the categories of health problems that all women aged 30 or 30 - 35 are concerned about completely overlap with the categories of health problems that all women aged 30 or 30 - 35, with a daily working hours of 6 hours, a sedentary office form, and a repetitive computer operation office mode are concerned about, then these several categories of health problems are regarded as explicit interest points;

[0071] Step S202: When there is partial overlap between several categories of health problems included in the first interest point set and the second interest point set, extract the overlapping and non - overlapping categories of health problems respectively; trace the user IDs of each consultation record corresponding to the non - overlapping categories of health problems; several consultation records correspond to several consultation quantities;

[0072] Step S203: Calculate the penetration rate Q of the non - overlapping categories of health problems respectively. The formula is:

[0073]

[0074] where P i represents the total number of times the i - th user ID consults a certain category of health problem within the non - overlapping categories of health problems; n represents the total number of user IDs obtained by tracing the user IDs for a certain category of health problem;

[0075] Step S204: Sort the penetration rates Q of the non - overlapping categories of health problems from high to low, and select the top three categories of health problems with the penetration rate Q to form the third interest point set;

[0076] Step S300: Use the interest points in the third interest point set and the interest points in the first interest point set and the second interest point set that satisfy the partial overlap relationship to perform discriminant analysis on the implicit and explicit interest points of users in different populations to be mined;

[0077] Among them, step S300 includes:

[0078] When there is one type of health problem that overlaps between the first set of points of interest and the second set of points of interest, two types of health problems that are different, and the overlapping type of health problem belongs to the types of health problems within the third set of points of interest, determine the several types of health problems included in the third set of points of interest as the explicit interests of the users included in the first population to be mined, and determine the two different types of health problems as the implicit interests of the users included in the second population to be mined; when there is one type of health problem that overlaps between the first set of points of interest and the second set of points of interest, two types of health problems that are different, and the overlapping type of health problem does not belong to the types of health problems within the third set of points of interest, determine the first two types of health problems within the third set of points of interest together with the overlapping type of health problem as the explicit interests of the users included in the first population to be mined, and determine the two different types of health problems as the implicit interests of the users included in the second population to be mined; one type of health problem corresponds to one point of interest;

[0079] When there are two types of health problems that overlap between the first set of points of interest and the second set of points of interest, one type of health problem that is different, and both of the overlapping types of health problems belong to the types of health problems within the third set of points of interest, determine the several types of health problems included in the third set of points of interest as the explicit interests of the users included in the first population to be mined, and determine the different type of health problem as the implicit interests of the users included in the second population to be mined; when there are two types of health problems that overlap between the first set of points of interest and the second set of points of interest, one type of health problem that is different, and both of the overlapping types of health problems do not belong to the types of health problems within the third set of points of interest, determine the first type of health problem within the third set of points of interest together with the overlapping two types of health problems as the explicit interests of the users included in the first population to be mined, and determine the different type of health problem as the implicit interests of the users included in the second population to be mined;

[0080] When there are two types of health problems that overlap between the first set of points of interest and the second set of points of interest, one type of health problem that is different, and only one of the overlapping types of health problems belongs to the types of health problems within the third set of points of interest, sort the types of health problems within the third set of points of interest except for the type of health problem that belongs to the third set of points of interest among the overlapping two types of health problems according to the penetration rate Q, and select the type of health problem with the larger penetration rate Q among the two together with the overlapping two types of health problems as the explicit interests of the users included in the first population to be mined, and determine the different type of health problem as the implicit interests of the users included in the second population to be mined;

[0081] Step S400: Obtain explicit points of interest and implicit points of interest of each user based on steps S100 to S300; calculate similarity values ​​between implicit points of interest of each user having the same explicit point of interest, and when the similarity value is greater than a similarity value threshold, establish an interest association relationship between the users; mutually push health consultation information contained in the implicit points of interest of each user having an interest association relationship; one category of health problems corresponds to one implicit point of interest, and one implicit point of interest contains health consultation information associated with the point of interest;

[0082] The process of calculating the similarity values ​​between the implicit interest points of each user having the same explicit interest point includes:

[0083] Step S401: Obtain the implicit interest points of each user with the same explicit interest point to obtain the corresponding implicit interest point set, and traverse the number of hidden interest points in the implicit interest point set of each user; set the users with the same explicit interest point one by one as the associated relationship to be mined users, and set other users whose difference between the number of hidden interest points in the implicit interest point set of the associated relationship to be mined is less than the difference threshold as the associated users to be matched with the associated relationship to be mined users;

[0084] Step S402: Calculate the association value W between the user whose association relationship is to be mined and each associated user to be matched, using the formula:

[0085]

[0086] Among them, A represents the set A of all hidden interest points of the users whose associations are to be mined; B represents the set B of all hidden interest points of the users whose associations are to be mined; crad(A∩B) represents the number of hidden interest points of the intersection of set A and set B; crad(A∪B) represents the number of hidden interest points of the union of set A and set B;

[0087] Step S403: establishing an interest association relationship between two users whose association value W is greater than an association threshold.

[0088] In order to better implement the above method, a push system for a health consultation information push method based on points of interest is also proposed, characterized in that the push system includes: a user classification module, a point of interest processing module, a point of interest state processing module, a calculation module, a point of interest state judgment module, an interest association relationship analysis module, and a health consultation information push module;

[0089] A user classification module, which is used to perform judgment and analysis based on the age, gender, and work information of each user, and classify each user into different populations to be mined. The populations to be mined include the first population to be mined and the second population to be mined; the second population to be mined is a population to be mined obtained by further classifying the first population to be mined;

[0090] Among them, the user classification module includes a first population to be mined classification unit, a second population to be mined classification unit, a first point of interest set processing unit, and a second point of interest set processing unit;

[0091] The first population to be mined classification unit is used to obtain the age and gender of the user; several users with the same age, the same gender, or an age difference less than the age difference threshold and the same gender are set as the first population to be mined;

[0092] The second population to be mined classification unit is used to set several users with the same daily working hours, office form, and office mode within the first population to be mined as the second population to be mined;

[0093] The first point of interest set processing unit is used to respectively accumulate the consultation quantities of the first population to be mined for various categories of health problems; sort the consultation quantities corresponding to various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity as the first point of interest set of the users included in the first population to be mined;

[0094] The second point of interest set processing unit is used to respectively accumulate the consultation quantities of several second populations to be mined for various categories of health problems, sort the consultation quantities for various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity as the second point of interest set of the users included in the second population to be mined;

[0095] The point of interest processing module is used to respectively merge the points of interest obtained from the users in different populations to be mined into different point of interest sets. The point of interest sets include the first point of interest set and the second point of interest set;

[0096] The point of interest status processing module is used to perform a transitive display process on the points of interest of the users in different populations to be mined with a completely overlapping relationship;

[0097] The calculation module is used to calculate the penetration rate Q of the points of interest of the users in different populations to be mined where the first point of interest set and the second point of interest set satisfy a partial overlapping relationship;

[0098] The point of interest status judgment module is used to receive the penetration rate Q in the calculation module and obtain the third point of interest set based on the penetration rate Q; use the points of interest in the third point of interest set and the points of interest in the first point of interest set and the second point of interest set with a partial overlapping relationship to perform discriminant analysis on the implicit and explicit points of interest of the users in different populations to be mined;

[0099] Among them, the point of interest status judgment module includes: a third point of interest set processing unit, a point of interest status judgment unit, and a point of interest status setting unit;

[0100] The third point of interest set processing unit is used to receive the penetration rate Q in the calculation module and obtain the third point of interest set based on the penetration rate Q;

[0101] The point of interest status judgment unit is used to discriminate and analyze the implicit and explicit interest points of users in different populations to be mined by comparing the points of interest in the third point of interest set with the points of interest in the first point of interest set and the second point of interest set with a partial coincidence relationship;

[0102] The point of interest status setting unit is used to receive the data in the point of interest status judgment unit and set the implicit or explicit status of the user interest points in different populations to be mined;

[0103] The interest association relationship analysis module is used to calculate the similarity value between the implicit interest points of each user with the same explicit interest point. When the similarity value is greater than the similarity value threshold, an interest association relationship is established between the users;

[0104] Among them, the interest association relationship analysis module includes: an association relationship user to be mined setting unit, a to-be-matched associated user setting unit, an association value calculation unit, and an interest association relationship establishment unit;

[0105] The association relationship user to be mined setting unit is used to obtain the users with the same explicit interest point and set them as the users with the association relationship to be mined;

[0106] The to-be-matched associated user setting unit is used to obtain other users whose difference in the number of hidden interest points in the implicit interest point set of the user with the association relationship to be mined is less than the difference threshold and set them as the to-be-matched associated users;

[0107] The association value calculation unit is used to receive the information in the association relationship user to be mined setting unit and the to-be-matched associated user setting unit to calculate the association value;

[0108] The interest association relationship establishment unit is used to receive the association value result in the association value calculation unit and establish an interest association relationship between two users whose association value is greater than the association threshold;

[0109] The health consultation information push module is used to receive the data in the interest association relationship analysis module and push the health consultation information included in the implicit interest points of each user with an interest association relationship to each other.

[0110] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0111] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for pushing health consultation information based on points of interest, characterized in that, the pushing method includes: Step S100: Classify each user into different populations to be mined based on the determination and analysis of the age, gender, and work information of each user; the populations to be mined include the first population to be mined and the second population to be mined; the second population to be mined is a population to be mined obtained by further dividing the first population to be mined; the points of interest obtained from the users within the different populations to be mined are respectively compiled into different point-of-interest sets, and the point-of-interest sets include the first point-of-interest set and the second point-of-interest set; Step S200: Based on the complete overlap relationship or partial overlap relationship between the point-of-interest sets of different populations to be mined, respectively set the status of the points of interest of the users within the different populations to be mined or calculate the penetration rate Q of the points of interest of the users within the different populations to be mined, and obtain a third point-of-interest set based on the calculated penetration rate Q of the points of interest; one category of health problems corresponds to one point of interest; Step S300: Use the points of interest in the third point-of-interest set and the points of interest in the first point-of-interest set and the second point-of-interest set that satisfy the partial overlap relationship to perform discriminant analysis processing on the implicit and explicit points of interest of the users within the different populations to be mined; Step S400: Based on Steps S100 - S300, obtain the explicit and implicit points of interest of each user; calculate the similarity value between the implicit points of interest of each user with the same explicit point of interest. When the similarity value is greater than the similarity value threshold, establish an interest association relationship between the users; push the health consultation information included in the respective implicit points of interest to the users with the interest association relationship; one category of health problems corresponds to one implicit point of interest, and one implicit point of interest contains health consultation information associated with this point of interest; The process of Step S200 for setting the status of the points of interest of the users within the different populations to be mined or calculating the penetration rate Q of the points of interest of the users within the different populations to be mined includes: Step S201: When there is a complete overlap between several categories of health problems included in the first point-of-interest set and the second point-of-interest set, determine the several categories of health problems included in the first point-of-interest set or the second point-of-interest set as the explicit points of interest of the users included in the first population to be mined; Step S202: When there is a partial overlap between several categories of health problems included in the first point-of-interest set and the second point-of-interest set, respectively extract the overlapping several categories of health problems and the non-overlapping several categories of health problems; trace the user IDs of each consultation record for the non-overlapping several categories of health problems; several consultation records correspond to several consultation quantities; Step S203: Calculate the penetration rate Q of the non-overlapping several categories of health problems respectively, and the formula is: Among them, P i represents the total number of a certain type of health problem among several non-overlapping types of health problems consulted by the i-th user ID; n represents the total number of user IDs obtained by tracing the user ID for the certain type of health problem. Step S204: Sort the penetration rates Q of the non-overlapping several categories of health problems from high to low, and select the top three categories of health problems with the sorted penetration rate Q to compile into the third point-of-interest set.

2. A method for pushing health consultation information based on points of interest according to claim 1, wherein, the step S100 includes: Step S101: Obtain the age and gender of the user; Set several users with the same age and gender or an age difference less than the age difference threshold and the same gender as the first group of users to be mined; Accumulate the consultation quantities of various types of health problems for the first group of users to be mined respectively; Sort the consultation quantities corresponding to various types of health problems from high to low, and select the top three types of health problems in the consultation quantity ranking as the first interest point set of the users included in the first group of users to be mined; One type of health problem corresponds to one implicit interest point. Step S102: Obtain the work information of each user within the first population to be mined, and obtain and determine the daily working hours, working form, and working mode of the users from their work information; the working form includes, but is not limited to, sedentary working and mobile working; if the proportion of the time the user spends sitting at work is greater than the proportion threshold, it is determined that the working form of this user is sedentary working; where t 1 represents the time the user spends in a sedentary state within the total daily working hours, and T represents the total daily working hours of the user; if the proportion of the time the user spends working while moving is greater than the proportion threshold, it is determined that the working form of this user is mobile working; where t 2 represents the time the user spends in a moving state within the total daily working hours, and T represents the total daily working hours of the user; among them, the work content with the highest repetition rate within the total daily working hours of the user is used as the working mode of the user; several users with the same daily working hours, working form, and working mode within the first population to be mined are set as the second population to be mined, and the first population to be mined is divided into several second populations to be mined; respectively accumulate the consultation quantities of various categories of health problems for the several second populations to be mined, sort the consultation quantities of various categories of health problems from high to low, and select the top three categories of health problems with the highest consultation quantity rankings as the second interest point set of the users included in the second population to be mined.

3. A method for pushing health consultation information based on points of interest according to claim 1, wherein: the step S300 includes: When there is one type of health problem that overlaps between the several types of health problems included in the first interest point set and the second interest point set, and there are two types of health problems with differences, and the overlapping one type of health problem belongs to the types of health problems included in the third interest point set, determine the several types of health problems included in the third interest point set as the explicit interest points of the users included in the first group of users to be mined, and determine the two types of health problems with differences as the implicit interest points of the users included in the second group of users to be mined; When there is one type of health problem that overlaps between the several types of health problems included in the first interest point set and the second interest point set, and there are two types of health problems with differences, and the overlapping one type of health problem does not belong to the types of health problems included in the third interest point set, determine the first two types of health problems in the third interest point set together with the overlapping one type of health problem as the explicit interest points of the users included in the first group of users to be mined, and determine the two types of health problems with differences as the implicit interest points of the users included in the second group of users to be mined; One type of health problem corresponds to one interest point. When there are two types of health problems that overlap between the several types of health problems included in the first interest point set and the second interest point set, and there is one type of health problem with differences, and the two overlapping types of health problems both belong to the types of health problems included in the third interest point set, determine the several types of health problems included in the third interest point set as the explicit interest points of the users included in the first group of users to be mined, and determine the one type of health problem with differences as the implicit interest points of the users included in the second group of users to be mined; When there are two types of health problems that overlap between the several types of health problems included in the first interest point set and the second interest point set, and there is one type of health problem with differences, and the two overlapping types of health problems both do not belong to the types of health problems included in the third interest point set, determine the first type of health problem in the third interest point set together with the two overlapping types of health problems as the explicit interest points of the users included in the first group of users to be mined, and determine the one type of health problem with differences as the implicit interest points of the users included in the second group of users to be mined; When there are two overlapping categories of health problems among several categories of health problems included in the first set of points of interest and the second set of points of interest, with one category of health problem being different, and only one of the two overlapping categories of health problems belongs to the category of health problems within the third set of points of interest, sort the categories of health problems within the third set of points of interest other than the category of health problems within the third set of points of interest among the two overlapping categories of health problems according to the penetration rate Q, and select the category of health problem with the larger penetration rate Q among the two as the explicit points of interest of the users included in the first population to be mined, together with the two overlapping categories of health problems, and determine the different category of health problem as the implicit points of interest of the users included in the second population to be mined.

4. A method for pushing health consultation information based on points of interest according to claim 1, wherein, the process of calculating the similarity values between the respective implicit points of interest for each user with the same explicit points of interest in step S400 includes: Step S401: Obtain the implicit points of interest of each user with the same explicit points of interest to obtain their respective corresponding sets of implicit points of interest, and traverse the number of hidden points of interest in the sets of implicit points of interest of each user; successively set the users with the same explicit points of interest as the users whose association relationship is to be mined, and set other users whose difference in the number of hidden points of interest in the set of implicit points of interest of the user whose association relationship is to be mined is less than the difference threshold as the users to be matched and associated with the user whose association relationship is to be mined; Step S402: Calculate the association value W between the user whose association relationship is to be mined and each user to be matched and associated, and the formula is: where A represents the set A formed by all the hidden points of interest of the user whose association relationship is to be mined; B represents the set B formed by all the hidden points of interest of the user to be matched and associated with the user whose association relationship is to be mined; crad(A∩B) represents the number of hidden points of interest in the intersection generated by set A and set B; crad(A∪B) represents the number of hidden points of interest in the union generated by set A and set B; Step S403: Establish an interest association relationship between two users whose association value W is greater than the association threshold.

5. A push system for a method for pushing health consultation information based on points of interest according to any one of claims 1-4, wherein, the push system includes: a user division module, a point of interest processing module, a point of interest status processing module, a calculation module, a point of interest status judgment module, an interest association relationship analysis module, and a health consultation information push module; the user division module is used to classify and analyze each user according to the age, gender, and work information of each user, and classify each user into different populations to be mined, and the populations to be mined include the first population to be mined and the second population to be mined; the second population to be mined is the population to be mined obtained by further dividing the first population to be mined; the point of interest processing module is used to respectively form different sets of points of interest from the points of interest obtained by the users within the different populations to be mined, and the sets of points of interest include the first set of points of interest and the second set of points of interest; The POI status processing module is used to perform an explicit conversion process on the POIs of users within different populations to be mined with a completely overlapping relationship; The calculation module is used to calculate the penetration rate Q of the POIs of users within different populations to be mined where the first POI set and the second POI set satisfy a partial overlapping relationship; The POI status judgment module is used to receive the penetration rate Q in the calculation module and obtain a third POI set based on the penetration rate Q; use the POIs in the third POI set and the POIs in the first POI set and the second POI set with a partial overlapping relationship to perform discriminant analysis on the implicit and explicit POIs of users within different populations to be mined; The interest association relationship analysis module is used to calculate the similarity value between the respective implicit POIs of each user with the same explicit POI. When the similarity value is greater than the similarity value threshold, an interest association relationship is established between each user; The health consultation information push module is used to receive the data in the interest association relationship analysis module and push the health consultation information included in the respective implicit POIs of each user with an interest association relationship to each other.

6. A health consultation information push system based on POIs according to claim 5, wherein, The user division module includes a first population to be mined division unit, a second population to be mined division unit, a first POI set processing unit, and a second POI set processing unit; The first population to be mined division unit is used to obtain the age and gender of the user; Several users with the same age and gender or an age difference less than the age difference threshold and the same gender are set as the first population to be mined; The second population to be mined division unit is used to set several users with the same daily working hours, office form, and office mode within the first population to be mined as the second population to be mined; The first POI set processing unit is used to respectively accumulate the consultation quantities of the first population to be mined for various types of health problems; Sort the consultation quantities corresponding to various types of health problems from high to low, and select the top three types of health problems with the highest consultation quantity as the first POI set of the users included in the first population to be mined; The second POI set processing unit is used to respectively accumulate the consultation quantities of several second populations to be mined for various types of health problems, sort the consultation quantities for various types of health problems from high to low, and select the top three types of health problems with the highest consultation quantity as the second POI set of the users included in the second population to be mined.

7. A health consultation information push system based on POIs according to claim 5, wherein, The POI status judgment module includes: a third POI set processing unit, a POI status judgment unit, and a POI status setting unit; The third POI set processing unit is used to receive the penetration rate Q in the calculation module and obtain a third POI set based on the penetration rate Q; The point of interest status judgment unit is configured to discriminate and analyze the implicit and explicit interest points of users in different populations to be mined by using the points of interest in the third point of interest set and the points of interest in the first point of interest set and the second point of interest set with a partial coincidence relationship; The point of interest status setting unit is configured to receive the data in the point of interest status judgment unit and set the implicit or explicit status of the user interest points in different populations to be mined.

8. A health consultation information push system based on points of interest according to claim 5, wherein, The interest association relationship analysis module includes: an association relationship user to be mined setting unit, a to-be-matched association user setting unit, an association value calculation unit, and an interest association relationship establishment unit; The association relationship user to be mined setting unit is configured to obtain users with the same explicit interest points and set them as users with the association relationship to be mined; The to-be-matched association user setting unit is configured to obtain other users whose difference from the number of hidden interest points in the implicit interest point set of the user with the association relationship to be mined is less than a difference threshold, and set them as to-be-matched association users; The association value calculation unit is configured to receive the information in the association relationship user to be mined setting unit and the to-be-matched association user setting unit and calculate the association value; The interest association relationship establishment unit is configured to receive the association value result in the association value calculation unit and establish an interest association relationship between two users whose association value is greater than an association threshold.

Images