CN112311612A

CN112311612A - Family portrait construction method and device and storage medium

Info

Publication number: CN112311612A
Application number: CN201910690845.0A
Authority: CN
Inventors: 俄万有
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-07-29
Filing date: 2019-07-29
Publication date: 2021-02-02
Anticipated expiration: 2039-07-29
Also published as: CN112311612B

Abstract

The embodiment of the invention discloses a method and a device for constructing a family portrait and a storage medium. The scheme is that a first user log collected by a plurality of service systems is obtained, and the first user log carries a network equipment identifier, a user identity identifier and position information; identifying a home network device identifier from the network device identifiers collected by the plurality of business systems based on the location information; generating connection state information according to a second user log corresponding to the home network equipment identifier, and screening a plurality of user identity identifiers associated with the home network equipment identifier according to the connection state information to determine a home user identity identifier; and extracting user characteristics from a third user log corresponding to the home user identity identifier, and generating a home portrait corresponding to the home network equipment identifier according to the user characteristics, thereby realizing portrait construction by taking a home as a unit.

Description

Family portrait construction method and device and storage medium

Technical Field

The invention relates to the technical field of family portrait construction, in particular to a family portrait construction method, a device and a storage medium.

Background

With the rapid development of artificial intelligence, mobile internet and big data technology, user portrait data formed based on mining analysis of mobile equipment and user data reported by APP is widely applied to scenes such as content recommendation, advertisement delivery and product refinement operation.

Therefore, the user data is mostly used based on statistics and analysis of data of a single user, and portrait data of the single user is created, so that targeted content recommendation is performed according to characteristics of the user.

However, in the course of research and practice on the prior art, the inventors of the present invention found that the construction and analysis of a home portrait cannot be performed to the maximum extent by simply using a single user as a unit, and that more and more businesses need to be developed in a home unit, and thus a solution for realizing the construction of a portrait in a home unit is urgently needed.

Disclosure of Invention

The embodiment of the invention provides a method, a device and a storage medium for constructing a family portrait, and aims to realize the construction of the family portrait.

The embodiment of the invention provides a family portrait construction method, which comprises the following steps:

acquiring a first user log collected by a plurality of service systems, wherein the first user log carries a network equipment identifier, a user identity identifier and position information;

identifying a home network device identifier from the network device identifiers collected by the plurality of business systems based on the location information;

generating connection state information according to a second user log corresponding to the home network equipment identifier, and screening a plurality of user identity identifiers associated with the home network equipment identifier according to the connection state information to determine a home user identity identifier;

and extracting user characteristics from a third user log corresponding to the home user identity identifier, and generating a home portrait corresponding to the home network equipment identifier according to the user characteristics.

The embodiment of the present invention further provides a family portrait creating device, including:

the system comprises a data acquisition unit, a service management unit and a service management unit, wherein the data acquisition unit is used for acquiring a first user log collected by a plurality of service systems, and the first user log carries a network equipment identifier, a user identity identifier and position information;

a device identification unit, configured to identify a home network device identifier from network device identifiers collected by the plurality of service systems based on the location information;

the identification association unit is used for generating connection state information according to a second user log corresponding to the home network equipment identification, and screening a plurality of user identification identifications associated with the home network equipment identification according to the connection state information to determine the home user identification;

and the portrait construction unit is used for extracting user characteristics from a third user log corresponding to the home user identity identifier and generating the home portrait corresponding to the home network equipment identifier according to the user characteristics.

The embodiment of the invention also provides a storage medium, wherein the storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute any family portrait construction method provided by the embodiment of the invention.

The embodiment of the invention also provides a server, which comprises a processor and a memory, wherein the memory is provided with a computer program, and the processor executes any family portrait construction method provided by the embodiment of the invention by calling the computer program.

According to the family portrait construction scheme provided by the embodiment of the invention, a family network device identifier belonging to a family network device is identified from all collected network device identifiers by obtaining a first user log collected by a plurality of service systems based on position information carried in the user log, connection state information is generated according to a second user log corresponding to the family network device identifier, a plurality of user identity identifiers related to the family network device identifier are screened according to the connection state information to determine the family user identity identifier, and a family portrait corresponding to the family network device identifier is generated according to a third user log corresponding to the family user identity identifier. The scheme of the invention integrates the user logs of a plurality of service systems, uses the characteristics of the geographic positions in the user logs to identify the home network equipment identification belonging to the home network and the home user identity identification corresponding to the home network, uses the home network equipment identification as a unified identification of a family, and analyzes the user logs corresponding to the home user identity identification so as to realize the creation of the family portrait.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1a is a schematic system architecture diagram of a family portrait construction method according to an embodiment of the present invention.

FIG. 1b is a schematic diagram of a first process of a family representation construction method according to an embodiment of the present invention;

FIG. 2 is a second flowchart of a method for constructing a family portrait according to an embodiment of the present invention;

FIG. 3a is a schematic diagram of a first configuration of a family representation constructing apparatus according to an embodiment of the present invention;

FIG. 3b is a schematic diagram of a second configuration of the home image construction apparatus according to the embodiment of the present invention;

FIG. 3c is a schematic diagram of a third configuration of a family representation constructing apparatus according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.

The embodiment of the invention provides a family portrait construction method, and an execution main body of the family portrait construction method can be the family portrait construction device provided by the embodiment of the invention or a server integrated with the family portrait construction device, wherein the family portrait construction device can be realized in a hardware or software mode.

Referring to fig. 1a, fig. 1a is a schematic diagram of a system architecture of a family portrait construction method according to an embodiment of the present invention. The system architecture deployed in the server comprises a home device layer, a data collection layer, a feature extraction layer and a data mining layer.

Various electronic devices in a family acquire internet services by connecting wifi, and the electronic devices include mobile terminals such as smart phones, tablet computers and handheld computers, PCs (personal computers) such as notebook computers and desktop computers, and intelligent household appliances such as smart televisions, smart sound boxes and smart refrigerators. When family members use these electronic devices, the electronic devices are typically connected to the family wifi for a long period of time or for some period of time. The application programs running in the electronic devices can report user logs generated in the using process of the users to the corresponding server side. Applications such as video playing, music, shopping, payment, etc. on a smart phone report a user log generated when the user uses the application to an application server. For another example, the smart television reports a user log generated when the user watches the television program to a server corresponding to the smart television. In this way, each electronic device of the home device layer reports the user log to its respective server. The user log reported by each electronic device at least carries the following information: network device identification, user identity identification, and location information.

And the data collection layer collects user logs reported by each service system. And the characteristic extraction layer extracts the corresponding user characteristics from the user logs collected by the service systems. The data mining layer includes a home tag computation service and an ID mapping computation service.

The ID mapping calculation service identifies the home network equipment identification based on the position information carried in the user log, obtains the incidence relation between the home network equipment identification and the user identification according to the user log, and determines the home user identification corresponding to the home network equipment from the home network equipment identification, so that the mapping relation between the home network equipment identification and the user identification is established.

Meanwhile, the feature extraction layer takes the service system as a unit, and performs feature extraction on the user logs collected by the receipt collection layer to obtain user features corresponding to the identity of each family user. And the family label calculation service of the data mining layer constructs the family portrait corresponding to each family network identification according to the user characteristics corresponding to each family user identification and the mapping relation between the family network equipment identification and the family user identification.

In some embodiments, the system architecture further includes a data application layer. The data application layer can expand corresponding family information for the user corresponding to each user identity in each service system through wifi-mac-based family image data and mapping relations between wifi-mac and the user identities of each service system, and therefore family consumption marketing, family-related advertisement delivery and content recommendation more suitable for family interests of the user are achieved.

Next, a family portrait construction method provided by an embodiment of the present invention is described in detail based on the system architecture. Referring to fig. 1b, fig. 1b is a first flowchart of a family portrait construction method according to an embodiment of the present invention. The specific flow of the family portrait construction method can be as follows:

101. the method comprises the steps of obtaining a first user log collected by a plurality of service systems, wherein the first user log carries a network equipment identifier, a user identity identifier and position information.

Each electronic device reports the user log to the server side of the electronic device through the respective data reporting channel. When the user logs are stored, the user identity is used as a key to be stored in a large data warehouse, such as Hadoop. For example, when the intelligent household electrical appliance reports data, the device ID is usually used as a user uniform identifier for uniquely identifying one device; for another example, when the PC reports data, an account number, such as a mailbox, a QQ number, etc., where the user logs in is usually used as a user unified identifier; for another example, when reporting data, a Mobile device such as a Mobile phone and a tablet computer and an application thereof usually uses a Mobile phone number, an account number for logging in by the application, or an IMEI (International Mobile Equipment Identity) as a user uniform identifier. The embodiment of the invention does not specifically limit the user identity identification, and the user identity identification can be uniquely identified. In this embodiment, the user logs are classified according to the user identities, so that a user log set corresponding to each user identity can be obtained.

In addition, it can be understood that one user in a family may log in different devices with the same account, or one user has multiple accounts of the same application, for example, the same user has two micro-messages, or the same user logs in an application on a smartphone or a tablet computer, for example, a microblog, a micro-message, a mailbox client, and the like, using the same social network account. The user identification and the number of the users in the family are not in one-to-one correspondence, the user identification and the electronic equipment in the family are not in one-to-one correspondence, and the number of the user identification in one family is determined by the number of the account numbers of all service systems used in the family. The business system refers to a system corresponding to a service provided by the electronic equipment, such as an intelligent television system; or a system corresponding to a service provided by an application program on the electronic device, such as a shopping system corresponding to a shopping APP and a payment system corresponding to a payment APP, where a server of each service collects a user log reported by a client, and the like.

Because a plurality of devices used by users belonging to the same family are generally connected to the same network device, the scheme of the invention takes the network device identification as the key to mine the user characteristics taking the family as the unit. Based on this, besides the user identity identifier, the user log uploaded by each device is also required to carry the network device identifier and the location information. The network device identifier is information that uniquely identifies the network device connected to the electronic device, for example, if the electronic devices in a home are all connected to the network through the same wireless router, the wireless network physical address (wifi-mac) of the wireless router is used as the network device identifier. After the electronic device is connected to the wireless router, the wireless network physical address can be acquired from the network state.

The user performs common network operations on the electronic device, and the electronic device interacts with the server through the wireless network. Taking the electronic device as an example of a smart phone, various communication applications such as WeChat and QQ, shopping applications, web browsing applications, social platform software, mailbox clients and the like are installed on the smart phone. The user can generate log data in the process of using the software, and the electronic equipment can report the log data to the server.

The log data includes user behavior data, such as web page content browsed by the user, user shopping records, user payment records, and the like. Taking the user shopping record as an example, the user shopping record may include the name, price, time, article description document, etc. of the article purchased by the user.

102. Based on the location information, a home network device identification is identified from the network device identifications collected by the plurality of business systems.

The family portrait construction device acquires user logs collected by each service system, and each user log carries position information and a family network device identifier, so that the corresponding relation between the family network device identifier and the position information can be obtained. The position information is a geographical position of the electronic equipment when the electronic equipment is connected with the wireless router and can be obtained by positioning the electronic equipment.

The family portrait creating apparatus summarizes the network device identifiers in all the collected user logs, so that a huge amount of network device identifiers may be obtained, and some non-family network device identifiers, such as network device identifiers of places such as shopping malls, office places, and the like, are included in the collected user logs.

In some embodiments, identifying the home network device identification from the network device identifications collected by the plurality of traffic systems based on the location information may include: analyzing the position information to acquire POI information of the position information in a preset range; if the POI information contains preset POI information, deleting the network equipment identification corresponding to the position information from the network equipment identifications collected by the plurality of service systems; and taking the rest network equipment identifications after the deleting operation as the home network equipment identifications.

Here, a POI (point of information) of a geographic location generally refers to information around a certain geographic location, and is a meaningful point on a map that is not in geographic significance, such as a store, a bank, a gas station, and the like. By identifying the periphery of the position information, for example, POI information within a preset range. And if the preset POI information exists in the POI corresponding to the position information, judging that the network equipment identifier corresponding to the position information is a non-home network equipment identifier. The preset POI information is a POI which can obviously identify that the corresponding position is a non-residential area. In this way, the network device identifiers collected by the plurality of service systems are identified one by one, and the network device identifiers in the non-residential area are filtered out, so that the rest network device identifiers are the network device identifiers in the residential area.

In some embodiments, the user log also carries network connection time; the step of using the remaining network device identifier after the deletion operation as the home network device identifier may include: acquiring network connection time corresponding to the residual network equipment identification after the deleting operation; determining the connection time distribution of the rest network equipment identifications in a first preset time interval according to the network connection time; and taking the rest network equipment identifications with the connection time distribution meeting the preset distribution rule as the home network equipment identifications.

And further filtering the network equipment identifications in the residential area on the basis of filtering out the network equipment identifications of the non-residential area. For example, there may be some non-home locations such as malls, offices, etc. in a residential area. In this embodiment, the home wireless network physical address may be identified from the wireless network physical addresses of the residential area by the network device identifying different characteristics in the distribution of connection times. There is a significant difference in the distribution of connection time between home and non-home wireless networks. The connection time of the home wireless network exhibits an aggregation characteristic from a working day 18 to a next day 9. And the connection time of the wireless network physical address of the non-home places such as the shopping mall, the office place and the like is less distributed in the time period from 18 o 'clock of the working day to 9 o' clock of the next day. And the aggregation characteristic can not appear, and the network equipment identifier which is obtained after the deleting operation and belongs to the residential area is obtained. And further filtering out the network equipment identifications belonging to the non-home wireless network in the residential area through the distribution characteristics of the network equipment identifications on the connection time.

The user log carries network connection time and network equipment identifiers, for each network equipment identifier belonging to a residential area, all network connection time corresponding to one network equipment identifier in a first preset time interval is counted, and connection time distribution corresponding to the network equipment identifier can be generated according to the network connection time. And if the connection time distribution meets a preset distribution rule, judging that the network equipment identifier is a home network equipment identifier. The preset distribution rule may be that the network connection time exhibits an aggregation characteristic in a preset time period, and the preset time period may be set as required, for example, the preset time period is 18 o 'clock to 9 o' clock of monday to friday of week, or the preset time period is 20 o 'clock to 9 o' clock of monday to friday of week. The preset time period can be one month, three months, half a year and the like, and can be set as required.

103. And generating connection state information according to a second user log corresponding to the home network equipment identifier, and screening a plurality of user identity identifiers associated with the home network equipment identifier according to the connection state information to determine the home user identity identifier.

The first user log comprises a second user log which is a user log screened from all the first user logs and corresponding to the home network device identifier. That is, while filtering out the non-home network device identifiers, the user log data corresponding to the non-home network device identifiers is also deleted.

And after the home network equipment identification is determined, constructing a mapping relation between the home network equipment identification and the home user identity identification. Because each second user log carries the network device identifier and the user identity identifier, all second user logs corresponding to one home network device identifier can be obtained according to the second user logs corresponding to the home network device identifiers, and according to the mode, the association relationship between the user identity identifier corresponding to the electronic device connected with the wireless network device of a certain home and the home network device identifier can be established as long as the user identity identifier corresponding to the electronic device is connected with the wireless network device of the certain home. However, the electronic devices corresponding to the user id associated with the home network device id do not necessarily all belong to the family members of the family, such as the visiting guest. Based on this, a plurality of user identification marks associated with the home network device identification need to be screened to determine the home user identification mark.

In addition, when a family member uses an electronic device to establish a connection with a wireless network device of the family, a certain rule, such as connection time, connection frequency, etc., is generally presented on the connection state information, so that the family member is different from the electronic device of a non-family member.

In some embodiments, the connection status information is a connection frequency, and the connection frequency is a connection frequency between the user equipment and the wireless network equipment within a second preset time interval; generating connection state information according to a second user log corresponding to the home network device identifier, and performing screening processing on a plurality of user identifiers associated with the home network device identifier according to the connection state information, wherein the step of determining the home user identifier may include: generating the connection frequency of each user identity mark associated with the home network equipment mark according to the network connection time; and taking the user identity identification with the connection frequency not less than the preset connection frequency as the home user identity identification corresponding to the home network equipment identification.

In this embodiment, for a user id, a user log of the user id in the second time interval may be obtained, and the connection frequency of the user id is obtained according to the network connection time in the user log, for example, the number of times that the electronic device or the application corresponding to the user id is connected to the wireless network device of the home in the last continuous month is obtained. And if the connection frequency is not less than the preset connection frequency, judging that the user identity identification and the home network equipment identification have a mapping relation.

For example, if the connection frequency is preset to be 7 times/three months, if the number of times that the electronic device or the application corresponding to a certain user identity is connected to the wireless network device of the home in the last three consecutive months is greater than 7 times, the user identity is used as the home user identity corresponding to the home network device identity. Through the mode, all the home user identification marks corresponding to each home network equipment mark can be screened out.

In addition, after the screening, a situation that one user id corresponds to multiple home network device identifiers may occur, and in order to more accurately establish a mapping relationship between the user id and the home network device identifiers to accurately construct a home portrait, a unique mapping relationship between the user id and the home network device identifiers needs to be established, that is, one user id only corresponds to one home network device identifier.

In some embodiments, the step of using the user id with the connection frequency not less than the preset connection frequency as the home user id corresponding to the home network device identifier may include: taking the user identity identification with the connection frequency not less than the preset connection frequency as a candidate user identity identification; if the candidate user identity identification corresponds to the plurality of home network equipment identifications, counting the connection times between the candidate user identity identification and the plurality of home network equipment identifications respectively; when the connection times between the candidate user identity identification and the home network identification are not the maximum value of the connection times, deleting the candidate identification; and taking the residual candidate user identification subjected to the deleting operation as the home user identification corresponding to the home network equipment identification.

In this embodiment, after determining a plurality of user identities corresponding to the home network device identity, the plurality of user identities are used as candidate user identities.

And aiming at each candidate user identity, if the candidate user identity corresponds to a plurality of home network equipment identities, calculating the connection times between the candidate user identity and the home network equipment identities, and taking the home network equipment identity with the maximum connection times as the unique home network equipment identity of the candidate user identity.

For example, the home network device identifier W1 corresponds to n candidate user identifiers UID1 and UID2 … … UIDn, and for any one of the candidate user identifiers UIDi, if UIDi corresponds to a plurality of different home network device identifiers (including the home network device identifier W1), the number of connections between UIDi and the plurality of home network device identifiers, respectively, is counted, if the number of connections between UIDi and the home network device identifier W1 is not the largest of the plurality of connections, UIDi is deleted from the home network device identifier W1 corresponding to the plurality of candidate user identifiers, and otherwise, if yes, the UIDi is retained. In this way, the home network device identifiers corresponding to the home user identifiers can be determined, wherein each home user identifier corresponds to a unique home network device identifier.

104. And extracting user characteristics from a third user log corresponding to the home user identity identifier, and generating a home portrait corresponding to the home network equipment identifier according to the user characteristics.

And after the mapping relation between the home network equipment identification and the home user identification is determined, constructing the home portrait corresponding to the home network equipment identification according to the third user logs corresponding to the home user identifications. The second user log comprises a third user log which is a user log screened from all the second user logs and corresponding to the home network device identifier.

In some embodiments, the third user logs of the plurality of home user identities corresponding to the home network device identities are integrated to obtain integrated data, and the integrated data is subjected to secondary mining.

For example, the user log also carries the device type of the electronic device, and the device type corresponding to the identity of each home user, such as a smart phone, a tablet computer, a smart home appliance, and the like, can be obtained according to the third user log. The number of the smart phones under the same home network equipment identification is mined, so that the number of the similar population in the home can be estimated.

For another example, according to the third user log, video playing data reported by a corresponding smart television, smart phone, or tablet computer under a home network device identifier is integrated, so that the age interval of the corresponding family member can be analyzed, specifically, video playing features are extracted from the video playing data reported by the smart television, smart phone, or tablet computer, so as to generate feature vectors, the feature vectors are input into a pre-trained deep neural network model or probability map model, and the distribution of the age intervals of the family members is determined according to the tags output by the deep neural network model, so as to further determine whether the old people or children are present in the family according to the distribution of the age intervals of the family members.

For another example, shopping category data reported by a plurality of corresponding smart phones and tablet computers under one home network device identifier is integrated according to a third user log, so that the shopping habits of the family can be analyzed. Specifically, shopping characteristics are extracted from shopping data reported by a plurality of smart phones and tablet computers, characteristic vectors are generated, and the characteristic vectors are input into a pre-trained deep neural network model or probability map model and the like to obtain relevant labels of shopping habits, consumption levels, consumption capabilities and the like of a family.

For another example, document reading class data reported by a plurality of corresponding smart phones and tablet computers under one home network device identifier is integrated according to a third user log, so that the reading habit of a home can be analyzed. Specifically, document reading data is extracted from shopping data reported by a plurality of smart phones and tablet computers, for example, e-books read by family members on reading APP, articles read on WeChat public numbers, news read on news APP, and the like, and document keywords are extracted by using a document theme generation model based on the document reading data to obtain relevant tags of the reading habits of families.

In other embodiments, the data belonging to other various dimensions may be integrated according to the third user log, and tag data of the family in multiple dimensions, such as a basic identity tag, a consumption tag, a video playing habit tag, a shopping habit tag, a reading habit tag, and the like of the family members, may be extracted, and the tags are integrated to form a family portrait of the family.

Alternatively, in some other embodiments, the step of extracting the user characteristics from the third user log corresponding to the home user id, and generating the home profile corresponding to the home network device id according to the user characteristics may include: extracting user characteristics from a third user log corresponding to the home user identity, and generating a user portrait corresponding to the user identity according to the user characteristics; and generating the family portrait corresponding to the household network equipment identifier according to the user portrait corresponding to the household user identity identifier and the third user log.

In this embodiment, a user portrait corresponding to each home user identifier is obtained based on the third user log corresponding to each home user identifier, and then tag data of the family in multiple dimensions, such as a basic identity tag, a consumption tag, a video playing habit tag, a shopping habit tag, a reading habit tag, and the like of the family members, is extracted according to the enumerated manner based on the third user log corresponding to the home user identifier, and the tags and tags in the user portrait corresponding to each home user identifier are synthesized to form the home portrait.

For example, in some embodiments, the step of extracting the user characteristics from the third user log corresponding to the home user id and generating the user representation corresponding to the user id according to the user characteristics may include: and carrying out data cleaning on a third user log corresponding to the home user identity, wherein the data cleaning comprises missing value processing, abnormal value processing, duplicate removal processing and noise data processing. Extracting user characteristics from the user logs subjected to data cleaning; generating a first label corresponding to the home user identity according to the user characteristics and a first classification model corresponding to the home user identity; and generating a user portrait corresponding to the home user identity according to the first label.

The step of generating the home representation corresponding to the home network device identifier according to the user representation corresponding to the home user identifier and the third user log may include: extracting target user characteristics corresponding to the preset label dimension from a third user log corresponding to the home user identity; generating a second label corresponding to the preset label dimension according to the target user characteristic and a second classification model corresponding to the preset label dimension; and generating the family portrait corresponding to the household network equipment identification according to the second labels corresponding to the plurality of preset label dimensions and the first label in the user portrait corresponding to the household user identification.

Wherein, in addition, a plurality of label dimensions, such as video playing, shopping habits, reading habits, and the like, can be preset. There is a corresponding second classification model for each preset label dimension. And acquiring the user characteristics corresponding to each preset label dimension, wherein the user characteristics can be acquired by referring to the listed modes. The first classification model and the second classification model can be a deep neural network model, a decision tree, a probabilistic graph model and the like, extracted user features are input into the models, corresponding labels can be obtained, a family image is formed through the labels, and multi-dimensional description of a family is achieved.

In some embodiments, after the family portrait is obtained, the family portrait may be used as a unit, and the family portrait is extended to each user in the family according to the tag in the family portrait, and in each business system, the family consumption marketing, the family-related advertisement delivery and the content recommendation more suitable for the family interest of the user are realized according to the tag of the family portrait corresponding to the user.

In particular implementation, the present application is not limited by the execution sequence of the described steps, and some steps may be performed in other sequences or simultaneously without conflict.

As described above, the method for constructing a family portrait according to the embodiment of the present invention includes acquiring a first user log collected by a plurality of service systems, identifying a home network device identifier belonging to a home network device from all collected network device identifiers based on location information carried in the user log, generating connection state information according to a second user log corresponding to the home network device identifier, performing a screening process on a plurality of user identifiers associated with the home network device identifier according to the connection state information, determining a home user identifier, and generating a family portrait corresponding to the home network device identifier according to a third user log corresponding to the home user identifier. The scheme of the invention integrates the user logs of a plurality of service systems, uses the characteristics of the geographic positions in the user logs to identify the home network equipment identification belonging to the home network and the home user identity identification corresponding to the home network, uses the home network equipment identification as a unified identification of a family, and analyzes the user logs corresponding to the home user identity identification so as to realize the creation of the family portrait.

The method according to the preceding embodiment is illustrated in further detail below by way of example.

Referring to fig. 2, fig. 2 is a second flow chart of the method for constructing a family portrait according to the embodiment of the present invention. The method comprises the following steps:

201. the method comprises the steps of obtaining a first user log collected by a plurality of service systems, wherein the first user log carries wifi-mac, user identity identification, position information and network connection time.

And storing the first user log collected by each business system into a big data warehouse, such as a Hadoop, by taking the user identity as a key. Therefore, the scheme of the invention takes wifi-mac as a key to mine the user characteristics in the unit of family. Besides the user identity identification, the user log uploaded by each device is also required to carry wifi-mac and location information. The wifi-mac is a wireless network physical address of the wireless router, namely the wifi-mac.

202. And analyzing the position information to acquire POI information of the position information in a preset range.

203. And if the POI information of the geographical position information contains preset POI information, deleting the wifi-mac corresponding to the position information from the wifi-macs collected by the plurality of service systems.

And analyzing the position information on the map to obtain the POI corresponding to the position information. And judging whether the geographic position is a non-residential area or not according to preset POI information which is preset and can obviously identify that the corresponding position is the non-residential area, and if so, deleting wifi-mac corresponding to the position information. By the method, wifi-macs collected by a plurality of service systems are identified one by one, wifi-macs of non-residential areas are filtered, and the rest wifi-macs of the residential areas are remained.

204. And for the wifi-mac remained after the deleting operation, generating the connection time distribution of the wifi-mac in the first preset time interval according to the corresponding second user log.

205. And taking the residual wifi-mac with the connection time distribution meeting the preset distribution rule as the household wifi-mac.

There may be some non-home locations such as malls, offices, etc. in the residential area. And further filtering the wifi-mac of the residential area based on the network connection time in the second user log, and deleting the wifi-mac of non-family places such as shopping malls and office places.

For example, the connection time of a home wireless network exhibits an aggregation feature during the time period from 18 o 'clock on a weekday to 9 o' clock the next day. And the connection time of the wireless network physical address of the non-home places such as the shopping mall, the office place and the like is less distributed in the time period from 18 o 'clock of the working day to 9 o' clock of the next day. And the aggregation characteristic can not appear, and the wifi-mac which is obtained after the deleting operation and belongs to the residential area is obtained. And further filtering wifi-mac belonging to a non-home wireless network in the residential area through the distribution characteristics of the wifi-mac in the connection time.

206. And generating the connection frequency of each user identity associated with the household wifi-mac according to the network connection time in the second user log corresponding to the household wifi-mac.

207. And taking the user identity mark with the connection frequency not less than the preset connection frequency as a home user identity mark corresponding to the home wifi-mac.

For any user identity associated with the family wifi-mac pair, a second user log of the user identity in a second time interval is obtained, and the connection frequency of the user identity is obtained according to the network connection time in the second user log, for example, the number of times that the electronic device or the application program corresponding to the user identity is connected to the wireless network device of the family in the last continuous month is obtained. And if the connection frequency is not less than the preset connection frequency, judging that the user identity identification and the household wifi-mac have a mapping relation. And taking the user identity as a home user identity corresponding to the home wifi-mac.

208. And determining the unique corresponding home wifi-mac of each home user identity according to the connection times between the home network equipment and the user identity so as to establish the mapping relation between the home wifi-mac and the home user identity.

After the screening, one family wifi-mac generally corresponds to a plurality of family wifi-macs, and meanwhile, the situation that one family wifi-mac corresponds to a plurality of family wifi-macs may occur.

For example, if a certain user identity corresponds to a plurality of household wifi-macs, the connection times between the user identity and the plurality of household wifi-macs are respectively calculated, and one user identity is uniquely bound with one household wifi-mac according to the rule that the connection times are the largest.

209. And extracting user characteristics from a third user log corresponding to the home user identity, and generating a user portrait corresponding to the user identity according to the user characteristics.

210. And generating a family portrait corresponding to the family wifi-mac according to the user portrait corresponding to the family user identity and a third user log.

And obtaining a user portrait corresponding to each home user identification based on a third user log corresponding to each home user identification, integrating the third user logs of the plurality of home user identifications corresponding to the home wifi-mac to obtain integrated data, and performing secondary mining on the integrated data. And obtaining a second label corresponding to the family, integrating the first label in the user portrait and the second label corresponding to the family, and selecting a plurality of labels with the highest weight values to construct the family portrait.

By taking the shopping label dimension as an example, the shopping data reported by a plurality of corresponding smart phones and tablet computers under a family wifi-mac is integrated according to a third user log, so that the shopping habit of the family can be analyzed. Specifically, shopping characteristics are extracted from shopping data reported by multiple smart phones and tablet computers, characteristic vectors are generated, the characteristic vectors are input into a pre-trained deep neural network model or probability map model and the like, relevant labels such as shopping habits, consumption levels and consumption capacities of families and probability values corresponding to the labels are obtained, and the probability values are used as weighted values of the labels. In such a way, the second label corresponding to each preset label dimension can be obtained. In addition, each label in a plurality of corresponding user images under the household wifi-mac corresponds to a weight value, and the labels in all the user images are combined. And integrating the second label obtained by secondary mining and the first label in the user portrait according to the corresponding weight value, deleting the label of which the weight value is smaller than a preset threshold value, and forming the family portrait by using the remaining first label and the second label.

From the above, in the family portrait construction method provided by the embodiment of the present invention, the first user logs collected by the multiple service systems are obtained, based on the POI information of the position information carried in the user logs, the family wifi-mac belonging to the family network device is identified from all the collected wifi-macs, the family wifi-mac is further screened according to the connection time distribution of the wifi-mac in the first preset time interval, then, the multiple user ids associated with the family wifi-mac are screened according to the connection frequency of each user id associated with the family wifi-mac, the family user id is determined, and the mapping relationship between the family wifi-mac and the family user id is established. And generating a user portrait and user characteristics according to a third user log corresponding to the home user identity, and integrating the user portrait and the user characteristics to obtain a home portrait corresponding to the home wifi-mac. According to the scheme, the user logs of a plurality of service systems are integrated, the family wifi-mac belonging to the family network is identified by using the POI characteristics of the geographic positions in the user logs, the mapping relation between the family wifi-mac and the family user identity identification is established, the family wifi-mac is used as a unified identification of a family, and the user portrait and the user characteristics are integrated to obtain the family portrait.

In order to implement the method, an embodiment of the present invention further provides a home image construction apparatus, which may be specifically integrated in a terminal device, such as a mobile phone, a tablet computer, and the like.

For example, please refer to fig. 3a, fig. 3a is a first structural diagram of a family portrait construction apparatus according to an embodiment of the present invention. The family representation construction apparatus may include a data acquisition unit 301, a device identification unit 302, an identification association unit 303, and a representation construction unit 304, as follows:

a data obtaining unit 301, configured to obtain a first user log collected by multiple service systems, where the first user log carries a network device identifier, a user identity identifier, and location information;

a device identification unit 302, configured to identify a home network device identifier from network device identifiers collected by the plurality of service systems based on the location information;

an identifier associating unit 303, configured to generate connection state information according to a second user log corresponding to the home network device identifier, and perform screening processing on multiple user identifiers associated with the home network device identifier according to the connection state information to determine a home user identifier;

and the portrait constructing unit 304 is configured to extract a user feature from a third user log corresponding to the home user identifier, and generate a home portrait corresponding to the home network device identifier according to the user feature.

Referring to FIG. 3b, FIG. 3b is a schematic diagram illustrating a second structure of a family portrait creating apparatus according to an embodiment of the present invention. In some embodiments, portrait construction unit 304 includes a first construction subunit 3041 and a second construction subunit 3042, wherein,

the first construction subunit 3041 is configured to: extracting user characteristics from a third user log corresponding to the home user identification, and generating a user portrait corresponding to the user identification according to the user characteristics;

the second building subunit 3042 is used to: and generating the family portrait corresponding to the household network equipment identification according to the user portrait corresponding to the household user identification and the third user log.

In some embodiments, the first building subunit 3041 is further for: carrying out data cleaning on a third user log corresponding to the home user identity, and extracting user characteristics from the user log subjected to data cleaning; generating a first label corresponding to the home user identity according to the user characteristics and a first classification model corresponding to the home user identity; and generating a user portrait corresponding to the home user identity according to the first label.

In some embodiments, the second building subunit 3042 is further for: extracting target user characteristics corresponding to preset label dimensions from a third user log corresponding to the home user identity; generating a second label corresponding to the preset label dimension according to the target user characteristic and a second classification model corresponding to the preset label dimension; and generating the family portrait corresponding to the household network equipment identification according to the plurality of second labels corresponding to the preset label dimensions and the first label in the user portrait corresponding to the household user identification.

Referring to FIG. 3c, FIG. 3c is a third schematic structural diagram of a family portrait building apparatus according to an embodiment of the present invention. In some embodiments, the device identification unit 302 includes a feature acquisition subunit 3021, a first identification subunit 3022, a second identification subunit 3023, wherein:

the feature acquisition subunit 3021 is configured to: analyzing the position information to acquire POI information of the position information in a preset range;

the first identification subunit 3022 is configured to: if the POI information contains preset POI information, deleting the network equipment identification corresponding to the position information from the network equipment identifications collected by the plurality of service systems;

the second identifier unit 3023 is used to: and taking the rest network equipment identifications after the deleting operation as the home network equipment identifications.

In some embodiments, the second identifying subunit 3023 is also to: acquiring network connection time corresponding to the residual network equipment identification after the deleting operation; determining the connection time distribution of the rest network equipment identifications in a first preset time interval according to the network connection time; and taking the rest network equipment identifications with the connection time distribution meeting the preset distribution rule as the home network equipment identifications.

In some embodiments, the identification association unit 303 is further configured to: generating the connection frequency of each user identity mark associated with the home network equipment mark according to the network connection time; and taking the user identity identification with the connection frequency not less than the preset connection frequency as the home user identity identification corresponding to the home network equipment identification.

In some embodiments, the identification association unit 303 is further configured to: taking the user identity identification with the connection frequency not less than the preset connection frequency as a candidate user identity identification;

if the candidate user identity identification corresponds to a plurality of home network equipment identifications, counting the connection times between the candidate user identity identification and the plurality of home network equipment identifications respectively;

deleting the candidate user identity identifier when the connection frequency between the candidate user identity identifier and the home network identifier is not the maximum value of the plurality of connection frequencies;

and taking the residual candidate user identification after the deletion operation as the home user identification corresponding to the home network equipment identification.

In a specific implementation, the above units may be implemented as independent entities, or may be combined arbitrarily to be implemented as the same or several entities, and the specific implementation of the above units may refer to the foregoing method embodiments, which are not described herein again.

It should be noted that the home portrait creation apparatus provided in the embodiment of the present invention and the home portrait creation method in the foregoing embodiment belong to the same concept, and any method provided in the embodiment of the home portrait creation method can be run on the home portrait creation apparatus, and a specific implementation process thereof is described in detail in the embodiment of the home portrait creation method, and is not described herein again.

The home portrait construction device provided by the embodiment of the invention acquires a first user log collected by a plurality of service systems through the data acquisition unit 301, the device identification unit 302 identifies a home network device identifier belonging to a home network device from all collected network device identifiers based on position information carried in the user log, the identifier association unit 303 generates connection state information according to a second user log corresponding to the home network device identifier, screens a plurality of user identity identifiers associated with the home network device identifier according to the connection state information to determine a home user identity identifier, and the portrait construction unit 304 generates a home portrait corresponding to the home network device identifier according to a third user log corresponding to the home user identity identifier. The scheme of the invention integrates the user logs of a plurality of service systems, uses the characteristics of the geographic positions in the user logs to identify the home network equipment identification belonging to the home network and the home user identity identification corresponding to the home network, uses the home network equipment identification as a unified identification of a family, and analyzes the user logs corresponding to the home user identity identification so as to realize the creation of the family portrait.

Fig. 4 shows a schematic structural diagram of a server according to an embodiment of the present invention, where fig. 4 is a schematic structural diagram of the server according to an embodiment of the present invention. Specifically, the method comprises the following steps:

the server may include components such as a processor 401 of one or more processing cores, memory 402 of one or more computer-readable storage media, a power supply 403, and an input unit 404. Those skilled in the art will appreciate that the configuration of the server shown in FIG. 4 does not constitute a limitation of the server and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components. Wherein:

the processor 401 is a control center of the server, connects various parts of the entire server using various interfaces and lines, and performs various functions of the server and processes data by running or executing software programs and/or modules stored in the memory 402 and calling data stored in the memory 402, thereby performing overall monitoring of the server. Optionally, processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor, which mainly handles operating systems, user interfaces, application programs, etc., and a modem processor, which mainly handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 401.

The memory 402 may be used to store software programs and modules, and the processor 401 may execute various functional applications and family representation construction by executing the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data created according to the use of the server, and the like. Further, the memory 402 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 access to the memory 402.

The server further includes a power supply 403 for supplying power to each component, and preferably, the power supply 403 may be logically connected to the processor 401 through a power management system, so as to implement functions of managing charging, discharging, and power consumption through the power management system. The power supply 403 may also include any component of one or more dc or ac power sources, recharging systems, power failure detection circuitry, power converters or inverters, power status indicators, and the like.

The server may also include an input unit 404, the input unit 404 being operable to receive input numeric or character information and to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

Although not shown, the server may further include a display unit and the like, which will not be described in detail herein. Specifically, in this embodiment, the processor 401 in the server loads the executable file corresponding to the process of one or more application programs into the memory 402 according to the following instructions, and the processor 401 runs the application program stored in the memory 402, thereby implementing various functions as follows:

As described above, the server provided in the embodiment of the present invention identifies, by obtaining a first user log collected by a plurality of service systems, a home network device identifier belonging to a home network device from all collected network device identifiers based on location information carried in the user log, generates connection state information according to a second user log corresponding to the home network device identifier, performs screening processing on a plurality of user identifiers associated with the home network device identifier according to the connection state information, determines a home user identifier, and generates a home representation corresponding to the home network device identifier according to a third user log corresponding to the home user identifier. The scheme of the invention integrates the user logs of a plurality of service systems, uses the characteristics of the geographic positions in the user logs to identify the home network equipment identification belonging to the home network and the home user identity identification corresponding to the home network, uses the home network equipment identification as a unified identification of a family, and analyzes the user logs corresponding to the home user identity identification so as to realize the creation of the family portrait.

To this end, embodiments of the present invention provide a storage medium having stored therein a plurality of instructions, which can be loaded by a processor to perform any of the family representation construction methods provided by embodiments of the present invention. For example, the instructions may perform:

The above operations can be implemented in the foregoing embodiments, and are not described in detail herein.

Wherein the storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.

Since the instructions stored in the storage medium can execute any family portrait construction method provided by the embodiment of the present invention, beneficial effects that can be achieved by any family portrait construction method provided by the embodiment of the present invention can be achieved, which are detailed in the foregoing embodiments and will not be described herein again. The method, the apparatus and the storage medium for constructing a family portrait according to the embodiments of the present invention are described in detail above, and the principles and embodiments of the present invention are described herein by applying specific examples, and the description of the embodiments is only used to help understanding the method and the core idea of the present invention; meanwhile, for those skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims

1. A family portrait construction method is characterized by comprising the following steps:

2. The method for building a family portrait according to claim 1, wherein the step of extracting user characteristics from a third user log corresponding to the home user id and generating a family portrait corresponding to the home network device id according to the user characteristics comprises:

extracting user characteristics from a third user log corresponding to the home user identification, and generating a user portrait corresponding to the user identification according to the user characteristics;

and generating the family portrait corresponding to the household network equipment identification according to the user portrait corresponding to the household user identification and the third user log.

3. The method for building a family portrait according to claim 2, wherein the step of extracting user characteristics from a third user log corresponding to the family user id and generating a user portrait corresponding to the user id according to the user characteristics comprises:

carrying out data cleaning on a third user log corresponding to the home user identity, and extracting user characteristics from the user log subjected to data cleaning;

generating a first label corresponding to the home user identity according to the user characteristics and a first classification model corresponding to the home user identity;

and generating a user portrait corresponding to the home user identity according to the first label.

4. The method for building a family portrait of claim 3, wherein the step of generating a family portrait corresponding to the home network device identifier according to the user portrait corresponding to the home user identifier and the third user log comprises:

extracting target user characteristics corresponding to preset label dimensions from a third user log corresponding to the home user identity;

generating a second label corresponding to the preset label dimension according to the target user characteristic and a second classification model corresponding to the preset label dimension;

and generating the family portrait corresponding to the household network equipment identification according to the plurality of second labels corresponding to the preset label dimensions and the first label in the user portrait corresponding to the household user identification.

5. The family representation construction method of claim 1, wherein said step of identifying a home network device identifier from the network device identifiers collected by the plurality of business systems based on the location information comprises:

analyzing the position information to acquire POI information of the position information in a preset range;

if the POI information contains preset POI information, deleting the network equipment identification corresponding to the position information from the network equipment identifications collected by the plurality of service systems;

and taking the rest network equipment identifications after the deleting operation as the home network equipment identifications.

6. A method for constructing a family representation as defined in claim 5, wherein said user log further carries a network connection time; the step of using the remaining network device identifier after the deletion operation as the home network device identifier includes:

acquiring network connection time corresponding to the residual network equipment identification after the deleting operation;

determining the connection time distribution of the rest network equipment identifications in a first preset time interval according to the network connection time;

and taking the rest network equipment identifications with the connection time distribution meeting the preset distribution rule as the home network equipment identifications.

7. The method for constructing a family portrait according to claim 1, wherein the step of generating connection status information according to a second user log corresponding to the home network device identifier, and performing a screening process on a plurality of user identifiers associated with the home network device identifier according to the connection status information to determine the home user identifier comprises:

generating the connection frequency of each user identity mark associated with the home network equipment mark according to the network connection time;

and taking the user identity identification with the connection frequency not less than the preset connection frequency as the home user identity identification corresponding to the home network equipment identification.

8. The method for constructing a family portrait according to claim 7, wherein the step of using the user id with the connection frequency not less than the preset connection frequency as the home user id corresponding to the home network device id comprises:

taking the user identity identification with the connection frequency not less than the preset connection frequency as a candidate user identity identification;

9. A family representation creation apparatus, comprising:

10. A storage medium storing instructions adapted to be loaded by a processor to perform a method of constructing a family representation according to any of claims 1 to 8.