CN114741595A

CN114741595A - Information pushing method and device

Info

Publication number: CN114741595A
Application number: CN202210378006.7A
Authority: CN
Inventors: 盛琨; 胡安民; 余晴; 郑宇�; 张钧波
Original assignee: Jingdong City Beijing Digital Technology Co Ltd
Current assignee: Jingdong City Beijing Digital Technology Co Ltd
Priority date: 2022-04-12
Filing date: 2022-04-12
Publication date: 2022-07-12

Abstract

The invention discloses a method and a device for pushing information, and relates to the technical field of computers. One embodiment of the method comprises: acquiring a user identifier set to be aligned, and performing identifier alignment processing on the user identifier set to be aligned and a business district user identifier set to obtain an aligned user identifier; obtaining moving track data of the aligned user identification according to service data, and determining a dense source of a visiting user based on the moving track data; respectively calculating the guest group portraits of the visiting users in each intensive source area according to the feature data of the visiting users; and comparing the similarity of the customer group image with the historical user images of the business circles to determine a target user, and pushing information of the target user. The embodiment can make up the defect of single characteristic dimension of user data, can better, more timely and more accurately depict the user customer group images, more accurately determines potential customers of a business circle, and is convenient for improving the success rate of information push.

Description

Information pushing method and device

Technical Field

The invention relates to the technical field of computers, in particular to an information pushing method and device.

Background

In the marketing scenario (e.g., advertisement placement, merchandise recommendation, etc.) of the current business segment, marketing is mainly concerned with both the aspect of existing customers to buy again and the aspect of activating potential new customers. At present, when potential new customers are mined, people are mostly explored from local long-term and short-term user behavior data of a business circle.

In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art:

the business district operator only conducts crowd exploration according to local long-term and short-term user behavior data of the business district, and detailed knowledge of the crowd condition of the outside of the business district and the consumption condition of shops in the business district is lacked; and due to the reasons of commercial interest protection and personal data privacy security, the business circle cannot effectively use the data of other external organizations, so that the data dimension is single, and the marketing cannot be effectively developed. The consumption characteristics of users in the guest group images are influenced by market public opinion, and the characteristic values at different times are different greatly. The existing business district insights do not consider the real-time characteristics of user data, so that the obtained customer images are inaccurate and cannot meet the real-time dynamic query in business analysis. In the prior art, when a target new guest is matched based on an existing visiting user, the method is performed only by expanding similar crowds according to a single user characteristic attribute, and the obtained target new guest is often inaccurate, so that the marketing effect is influenced.

In summary, when potential new customers are mined in a business circle at present, the conditions that the data dimension is single, the customer group images do not have real-time performance, and the target new customers are not accurately matched exist, so that the marketing effect of the business circle is reduced.

Disclosure of Invention

In view of this, embodiments of the present invention provide an information pushing method and apparatus, which can aggregate multi-party data to perform user portrait description and information pushing, make up for the defect of single characteristic dimension of user data in a business district operation scene, better, more real-time, and more accurately perform user guest group portrait description, more accurately determine potential customers of the business district, and facilitate improvement of information pushing success rate.

To achieve the above object, according to an aspect of the embodiments of the present invention, an information pushing method is provided.

A method for pushing information comprises the following steps:

acquiring a user identifier set to be aligned, and performing identifier alignment processing on the user identifier set to be aligned and a business district user identifier set to obtain an aligned user identifier;

obtaining moving track data of the aligned user identification according to service data, and determining a dense source of a visiting user based on the moving track data;

respectively calculating the guest group portraits of the visiting users in each intensive source area according to the feature data of the visiting users;

and comparing the similarity of the customer group image with the historical user images of the business circles to determine a target user, and pushing information of the target user.

Optionally, the identifier alignment processing includes:

taking intersection processing, taking union processing, left outer connection processing or right outer connection processing.

Optionally, performing identifier alignment processing on the to-be-aligned user identifier set and the business district user identifier set includes:

and sending the user identifier set to be aligned to a third party arbitration mechanism so that the third party arbitration mechanism carries out identifier alignment processing according to the user identifier set to be aligned and the business district user identifier set.

Optionally, before performing identifier alignment processing on the to-be-aligned user identifier set and the business district user identifier set, the method further includes:

acquiring a processing rule of a user identifier agreed with a business circle in advance, wherein the processing rule comprises a data format and an encryption mode of the user identifier;

and processing the user identifier set to be aligned by using the processing rule, and taking the processed user identifier set to be aligned as the user identifier set to be aligned.

Optionally, determining a dense source of visiting users based on the movement trajectory data comprises:

determining a place of residence and a place of work for each visiting user based on the movement trajectory data;

clustering the residential areas and the workplaces by adopting a space clustering algorithm to obtain a plurality of clustering areas;

and selecting a specified number of clustering areas with the most densely distributed visiting users in the space from the plurality of clustering areas as dense source areas of the visiting users.

Optionally, the comparing the similarity of the guest group portraits with historical user portraits of a business district to determine a target user comprises:

respectively carrying out category coding on the guest group images and the structured image data in the historical user images of the business circles, and obtaining image characterization vectors through dimension reduction;

respectively carrying out social network representation extraction on the client group image and unstructured image data in the historical user images of the business circles to obtain network representation vectors;

splicing the portrait characterization vector and the network characterization vector together to respectively obtain the characteristic vectors of the customer portrait and the historical user portrait of the business district;

and based on the similarity between the feature vectors, comparing the similarity of the customer group portrait with the historical user portrait of the business circle, and determining the visiting user corresponding to the customer group portrait with the similarity meeting a preset threshold as the target user.

Optionally, the pushing information to the target user includes:

and acquiring a first dense source area of the target user, and returning the first dense source area to a business district so that the business district pushes information according to the first dense source area.

According to another aspect of the embodiment of the invention, an apparatus for pushing information is provided.

An information pushing device, comprising:

the system comprises a user identifier alignment module, a service area user identifier alignment module and a service area user identifier alignment module, wherein the user identifier alignment module is used for acquiring a user identifier set to be aligned and performing identifier alignment processing on the user identifier set to be aligned and a business area user identifier set to obtain an aligned user identifier;

the user source determining module is used for acquiring the moving track data aligned with the user identifier according to the service data and determining the dense source of the visiting user based on the moving track data;

the guest group image calculation module is used for calculating the guest group image of each visiting user in the intensive source area according to the feature data of the visiting user;

and the target user determining module is used for comparing the similarity of the customer group image and the historical user image of the business circle to determine a target user and pushing information of the target user.

According to another aspect of the embodiment of the invention, an electronic device for pushing information is provided.

An electronic device for information push, comprising: one or more processors; the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the information pushing method provided by the embodiment of the invention.

According to yet another aspect of an embodiment of the present invention, a computer-readable medium is provided.

A computer-readable medium, on which a computer program is stored, which, when executed by a processor, implements the method for information push provided by an embodiment of the present invention.

One embodiment of the above invention has the following advantages or benefits: obtaining a user identifier set to be aligned, and performing identifier alignment processing on the user identifier set to be aligned and a business district user identifier set to obtain an aligned user identifier; obtaining moving track data aligned with the user identification according to the service data, and determining a dense source area of the visiting user based on the moving track data; respectively calculating the guest group portraits of the visiting users in each intensive source area according to the feature data of the visiting users; the similarity comparison is carried out on the customer group portrait and the historical user portrait of the business district to determine a target user, and information push is carried out on the target user, so that the user portrait description and the information push can be carried out by integrating multi-party data, the defect of single characteristic dimension of user data under the business district operation scene is overcome, the source of a user who does not visit the business district is fully mined, and the information push such as advertisements is better carried out under the new customer drainage scene of the business district; the accurate dense source of the user is obtained based on the time-space attribute of the user track information, so that the user customer group image can be better, more real-time and more accurately depicted, potential customers of the business circle can be more accurately determined, and the information push success rate is convenient to improve.

Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.

Drawings

The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:

fig. 1 is a schematic diagram of main steps of a method for pushing information according to an embodiment of the present invention;

FIG. 2 is a flow chart illustrating an implementation of an embodiment of the present invention;

FIG. 3 is a schematic diagram of a process of constructing a guest group image and mining a new guest according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of the main modules of an information pushing apparatus according to an embodiment of the present invention;

FIG. 5 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;

fig. 6 is a schematic block diagram of a computer system suitable for use in implementing a terminal device or server of an embodiment of the invention.

Detailed Description

Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

In the marketing scenario of the current business district, marketing mainly focuses on two aspects of repurchasing existing customers and activating potential new customers. But the local data is not only seriously insufficient in data and feature dimension, but is always limited to the data generated in the range of the current business circle. For example, an operating company in a business district may wish to know the source region of potential customers who have not visited the business district, the people figure, etc., and by analyzing and guiding the target new customer source in a targeted manner, the new customer source is attracted to various business states. In order to fully expand the marketing radiation range of a business district, people characteristic data of passenger flows outside the business district need to be fully mined, data characteristics of various dimensions and travel characteristics are analyzed to be comprehensively compared, a focus point of the existing passenger flows is locked, potential passenger source information is found, and data support is provided for a commercial insight board block for new passenger insight. But the operator of the business district does not master the crowd characteristic data and consumption data of customers outside the business district, and even the specific consumption data of shops in the business secret business district are difficult to obtain, so that great resistance is formed for mining and guiding and corresponding marketing of target new customers outside the business district. If the business district can access the online customer flow data, the target new customers can be analyzed into various business states to attract new customer sources, and therefore income is increased.

The main defects of the current business circle in the process of mining new customers are as follows:

1. the characteristic dimension is rare: the business district operator data only carries out crowd exploration according to local long-term and short-term user behavior data of the business district, and detailed knowledge of the crowd condition of the passenger flow outside the business district and the consumption condition of shops inside the business district is lacked; and due to the reasons of commercial interest protection and personal data privacy safety, the business circle cannot effectively use the data of other external organizations, so that the data dimension is single, and the marketing cannot be effectively developed.

2. The guest group portraits do not have real-time: the consumption characteristics of users in the guest group images are influenced by market public opinion, and the characteristic values at different times are different greatly. The existing business district insights do not consider the real-time characteristics of user data, so that the obtained customer group images are inaccurate and cannot meet the real-time dynamic query in business analysis.

3. Inaccurate target new guest matching: in the prior art, when a target new guest is matched based on an existing visiting user, only a TF-IDF method is used for expanding similar crowds according to a single user characteristic attribute, and the obtained target new guest is often inaccurate, so that the marketing effect is influenced.

In order to solve the defects in the prior art, the invention mainly adopts the following technical means:

1. the problem of insufficient data dimension of business district operators is solved, data statistics and query of multiple parties and cross-domain data are safely realized by combining data of one or more external partners under the condition that personal user data privacy is protected and business confidentiality is not leaked, data characteristics of users inside and outside the business district are complemented, targeted new customers are found in a gathering place, and the purpose of assisting business marketing to expand business district radiation area is achieved through targeted drainage modes such as advertisement putting, drainage of regular buses and business state optimization;

2. based on the time-space attribute of the user track information, carrying out spatial clustering through long-term residence time periods of different users to obtain the accurate source of the users; by combining the dynamic characteristics of the multi-party characteristics of the user, constructing a guest group portrait of the visiting user and the target new guest according to short-term data in hours and minutes and long-term data in days and weeks, dynamically reflecting the guest group characteristics in real time, and recommending the target new guest in real time according to the visiting user; the business district and the social platform cooperate, a social Graph network is utilized, and the representation reflecting first-order and second-order similarity of the social Graph of the user is extracted as a feature through a Graph Embedding mode of LINE, so that the matching precision of the target new client is improved.

In describing the embodiments of the present invention, the technical terms and their definitions are as follows.

1. The OPTICS algorithm: the OPTICS algorithm is called Ordering points to identify the clustering structure, and aims to cluster the data in the space according to density distribution. The idea is very similar to that of DBSCAN, but DBSCAN is difficult to select input parameters, i.e., DBSCAN is sensitive to input parameters. The difference between OPTICS and DBSCAN is that the density-based clustering structure can present a special sequence, the clustering structure corresponding to the sequence contains the information of the clusters of each level, and the analysis is convenient, and the clustering result is not affected by slight change of the neighborhood radius. The OPTICS algorithm has many advantages:

1) the number of clusters does not need to be specified in advance, and clusters with any shapes can be found;

2) the method is insensitive to abnormal points, and the abnormal points can be automatically identified in the clustering process;

3) clustering results do not depend on the traversal sequence of the nodes;

4) the method is insensitive to neighborhood radius parameters, and the clustering result is more stable.

2. TF-IDF: TF-IDF (Term Frequency-inverse Document Frequency) is a statistical analysis method for keywords and is used for evaluating the importance degree of a word to a file set or a corpus. The importance of a word is proportional to the number of times it appears in the article and inversely proportional to the number of times it appears in the corpus. The calculation mode can effectively avoid the influence of common words on the keywords, and improves the correlation between the keywords and the articles. Where TF refers to the total number of times a word appears in an article, the index is typically normalized to TF (the number of times a word appears in a document/the total word size of a document), which prevents the bias of the result towards too long documents (the same word will typically have a higher word frequency in long documents than in short documents). The IDF inverse document frequency indicates that the less documents containing a word, the greater the IDF value, which means that the word has a strong ability to distinguish between IDF (total number of documents in the corpus/number of documents containing the word +1), and +1 is because the denominator is avoided to be 0. TF-IDF is TFxIDF, and a larger value of TF-IDF indicates a larger importance of the feature word to the text.

3. Cosine similarity: cosine Similarity (Cosine Similarity) their Similarity is evaluated by calculating the Cosine of the angle between two vectors. And drawing the vectors into a vector space according to the coordinate values to obtain the included angles of the vectors, and obtaining cosine values corresponding to the included angles, wherein the cosine values can be used for representing the similarity of the two vectors. The smaller the angle, the closer the cosine value is to 1, and the more they coincide in direction, the more similar.

4. LINE: LINE (Larg-scale Information Network Embedding) was proposed by Jian Tang equal to 2015, which proposes a node Embedding algorithm that can be applied to large networks of any edge type, and implements Network Embedding by considering first-order constraint (local structure) and second-order constraint (global structure). Compared with the previous Graph/Network Embedding method, LINE has the following advantages: 1. any type of network is applicable, where any type is referred to herein primarily as any weight and direction of edges: directed, undirected, weighted, and unweighted. (LINE does not consider heterogeneous networks under different node types and edge types, and has certain limitations.a later study on heterogenous and homogeneous networks has also been made); LINE proposes an edge-sampling (edge-sampling) algorithm to promote and optimize the objective function, thereby overcoming the limitation of the conventional stochastic gradient descent (stochastic gradient destination).

5. Homomorphic encryption: homomorphic Encryption (HE) is an Encryption method with special natural attributes, the concept is firstly proposed by Rivest et al in the 70 th century, compared with a general Encryption algorithm, Homomorphic Encryption can realize various computing functions among ciphertexts besides realizing basic Encryption operation, namely, firstly computing and then decrypting can be equivalent to firstly decrypting and then computing.

Fig. 1 is a schematic diagram of main steps of a method for pushing information according to an embodiment of the present invention. In this embodiment, the execution subject of the information pushing method is a partner of the business circle. The cooperation party can be a mobile phone APP, an online shopping website, an offline shopping platform, an online or offline tourism product selling platform and the like, and mainly depends on business requirements of the business district party and actual information which can be provided by the cooperation party. As shown in fig. 1, the method for pushing information according to the embodiment of the present invention mainly includes the following steps S101 to S104.

Step S101: acquiring a user identifier set to be aligned, and performing identifier alignment processing on the user identifier set to be aligned and a business district user identifier set to obtain an aligned user identifier;

step S102: obtaining the moving track data aligned with the user identification according to the service data, and determining the dense source of the visiting user based on the moving track data;

step S103: respectively calculating the guest group portraits of the visiting users in each intensive source area according to the feature data of the visiting users;

step S104: and comparing the similarity of the customer group image with the historical user image of the business district to determine a target user, and pushing information of the target user.

According to one embodiment of the present invention, the tag alignment process includes: taking intersection processing, taking union processing, left outer connection processing or right outer connection processing. The way of Identification (ID) alignment is not fixed, and there are also many ways to do this based on security considerations, but the core is to take the intersection. If the actual service needs, the method can also be carried out in a mode of including a union set, or a left external connection, a right external connection and the like.

According to another embodiment of the present invention, the process of performing identifier alignment on the to-be-aligned user identifier set and the business district user identifier set includes: and sending the user identifier set to be aligned to a third-party arbitration mechanism so that the third-party arbitration mechanism performs identifier alignment processing according to the user identifier set to be aligned and the business district user identifier set. Wherein the third party arbitration mechanism is an independent third party neutral security node. However, in the specific implementation process, besides summarizing the user identifiers to the third-party arbitration mechanism for identifier alignment, if the merchant party and the business district party trust each other safely, the partner may also directly send the user identifier set to be aligned to the business district operator, and after the business district party completes identifier alignment, the business district party sends the user identifier set to the partner again, or vice versa, depending on which way meets the requirements of service and safety compliance. Namely: the third-party arbitration mechanism can be a third-party neutral security node different from the merchant party and the business circle party, and can also be any one of the merchant party and the business circle party which meet the requirements of business and security compliance.

According to another embodiment of the present invention, before performing identifier alignment processing on the to-be-aligned user identifier set and the business district user identifier set, the method further includes: acquiring a processing rule of a user identifier agreed with a business circle in advance, wherein the processing rule comprises a data format and an encryption mode of the user identifier; and processing the user identifier set to be aligned by using the processing rule, and taking the processed user identifier set to be aligned as the user identifier set to be aligned. Before the identification alignment processing, the user identification set to be aligned and the business district user identification set need to be processed in a unified manner, specifically, a business district party and a partner specify a unified data format of the user identification so as to align identification IDs, and the IDs are only unique identification. The type of ID aligned may be different according to different marketing scenarios. Further, a uniform ID encryption method is also specified, and it is necessary to unify the encryption method so that the ID cannot be transmitted in plain text and must be ciphertext. The encryption mode can be in various forms, MD5 or Hash encryption, or various encryption modes are combined, so that the encrypted ID can be prevented from being decrypted, and the safety and the privacy of data can be better guaranteed.

After the identification alignment, for the aligned user identification, the partner inquires and aligns the movement track information of the user in the past year according to the own service data (such as mobile phone signaling data). The signaling data is data that a mobile operator acquires the current time and the current position of the mobile terminal at a fixed frequency, and can be acquired after the mobile operator performs security processing and authorization. In addition, assuming that the partner is an e-commerce platform, the movement track information of the user can be queried according to the receiving address and the like in the aligned order data of the user, and the like. For different service scenes, the movement track information of the aligned user can be obtained in different modes.

According to yet another embodiment of the present invention, determining a dense source of visiting users based on the movement trajectory data comprises: determining a place of residence and a place of work for each visiting user based on the movement trajectory data; clustering the residential area and the working area by adopting a space clustering algorithm to obtain a plurality of clustering areas; and selecting a specified number of clustering areas with the most densely distributed visiting users in the space from the plurality of clustering areas as dense source areas of the visiting users. In an embodiment of the present invention, assuming that the mobile track information of the aligned user is obtained through the mobile phone signaling data, first, after the track information of the aligned user is obtained, the partner determines the residence and the work place of the user according to the track information, where the determination logic is: the high-frequency residence point of the user at 9-17 o 'clock of the working day is the working place, and the high-frequency residence point of the user at 10-8 o' clock later is the residence place (residence point: the position where the sample track information stays for a long time). And then, calculating the residence and the working place of all the users, and clustering the residence and the working place by adopting a space clustering algorithm to obtain a clustering region. And finally, screening out the clustering areas with the most densely distributed appointed number (for example, 10) of visiting users in the space from the clustering results, and displaying the clustering areas in a map, namely the top ten dense source places of the visiting users in the business district.

Then, after the dense source places of the visiting users are determined, the guest group images of the visiting users of each dense source place are calculated respectively according to the feature data of the visiting users. The dimensions of the guest group portrait include: the method comprises the steps of determining the sample visit quantity in a fixed time period, the time distribution condition of the visited users (such as counting the number of the visited samples in different time periods), composition analysis, social graph analysis and the like, wherein the composition analysis can comprise structural characteristics such as gender, age, presence or absence of children, consumption capacity, shopping preference and the like; the social Graph analysis mainly utilizes relevant data information of social software, and extracts representations reflecting first-order and second-order similarity from the relations between relatives and friends and coworkers of the visiting user by utilizing a LINE algorithm in Graph Embedding, and the representations are used as a basis for matching the guest group image of the visiting user with the target new guest group image in the later period. Meanwhile, in the processing of the structural features, different processing is adopted according to the time attribute of the user label: for static features, the representations of the static features are stored in a content library to facilitate later use, for dynamic features such as features which change frequently along with time, such as user preference, interest and the like, long-term representation vectors can be calculated by taking days and weeks as units, short-term representation vectors can be calculated by taking hours and minutes as units, weighting processing is carried out on the short-term representation vectors and the long-term representation vectors, the weighting processing is used for calculating the matching degree of the later-stage features and the target new customers, and therefore real-time user recommendation is achieved. The guest images do not determine the specific image or tag of the individual.

According to yet another embodiment of the invention, comparing the guest group portraits with historical user portraits of a business district for similarity to determine target users comprises:

splicing the portrait representation vector and the network representation vector together to respectively obtain the characteristic vectors of the customer group portrait and the historical user portrait of the business circle;

and comparing the similarity of the customer group portrait with the historical user portrait of the business district based on the similarity between the feature vectors, and determining the visiting user corresponding to the customer group portrait with the similarity meeting a preset threshold as the target user.

The historical user image of the business district is characterized by extracting features of the historical users visiting the business district, and is realized by a similar user mining algorithm when the similarity of the customer group image and the historical user image of the business district is compared. The similar user mining algorithm is as follows: firstly, carrying out category coding on structured image data (such as age, gender, consumption capability and the like) (such as a male marker of 0001 and a female marker of 1000) and obtaining a more compact image characterization vector through dimension reduction. Then, for unstructured relationships (such as a social Graph), extracting social network representation by using a Graph Embedding LINE algorithm to obtain a network representation vector, and splicing the network representation vector with the portrait representation vector. And finally, calculating the similarity between the samples according to the characterization vectors of the samples obtained after splicing, wherein the samples exceeding a certain threshold value are the target new guest group. The sample here is a user.

According to still another embodiment of the present invention, the pushing information to the target user comprises: and acquiring a first dense source area of the target user, and returning the first dense source area to a business district so that the business district pushes information according to the first dense source area. After the target user is determined, since the movement track of the target user is not limited to a certain business circle, the dense source of the target user can be calculated according to the method for determining the dense source of the user in the foregoing step S102, so as to better determine the range and the push content of information push performed by the business circle.

In addition, in order to ensure the security and privacy of data, the third party arbitration mechanism may be used to compare the similarity between the customer group image and the historical user image of the business district to determine the target user. And after the first dense source place of the target user is determined, the first dense source place can also be returned to the third-party arbitration mechanism for aggregation, and then the third-party arbitration mechanism sends the aggregated data to the business district operator for subsequent marketing planning. The partner may be one or more, depending on the marketing content needs. When the customer group image is constructed, the user characteristics come from the label characteristics of the trade area, the electronic trade label characteristics, the social graph characteristics of the social platform and the like, and the characteristic vector data of all parties cannot be aggregated to one party to calculate the cosine similarity. Therefore, by adopting a homomorphic encryption technology, the results obtained after different feature vector calculations of the three-party platform are encrypted by a public key and sent to the third-party arbitration mechanism, and the third-party arbitration mechanism decrypts the results by a private key and then sends the cosine similarity results to the business district side.

Fig. 2 is a schematic flow chart of implementation of the embodiment of the present invention. As shown in fig. 2, it is shown how information push is performed based on business circle parties, partners and third party arbitration mechanisms in one embodiment of the present invention. The method mainly comprises the following steps:

1. the two parties (business circle party and partner) agree on the processing rule for the user identification ID: at the beginning of the project, the business partner and the partner define a uniform user data ID format so as to align the IDs, wherein the IDs are only unique identification marks. The type of ID aligned may be different according to different marketing scenarios. In addition, a uniform ID encryption mode is also specified, the ID can not be transmitted by plaintext, the ID must be ciphertext, and the encryption mode is uniform. The encryption mode can be in various forms, MD5 or Hash encryption, or a combination of various encryption modes, so that the encrypted ID can be prevented from being decrypted;

2. both parties (business circle party and partner) prepare ID sets and feature sets: the business district provides the user ID visited in the last year according to the consumption data of own businesses (such as dining, lodging, traveling and the like). Similarly, the partner provides the ID of the access user in the last year according to the self service data;

3. ID alignment preparation: the operator of the business circle party can initiate an ID alignment task at the moment and send the ID alignment task to the partner, so that the partner can choose to accept or reject the ID alignment task;

4. ID alignment processing: the two parties (the business circle party and the partner) encrypt the IDs which need to be aligned and respectively send the IDs to a third-party arbitration mechanism (an independent third-party neutral security node), the arbitration mechanism conducts ID alignment (such as intersection taking) operation, and the IDs are transmitted to the partner by the arbitration mechanism after being aligned. In specific implementation, besides summarizing the ID to the arbitration mechanism, under the condition that the business district party and the partner trust each other, the partner can directly send the ID to the business district party, and after the business district party completes the ID alignment, the business district party sends the ID to the partner again, or vice versa, depending on which way meets the requirements of business and security compliance. The way of ID alignment is not fixed, and there are also many ways to do this based on security considerations, but the core is to take the intersection. If the actual service needs, the method can also be carried out in a mode of including a union set, or a left external connection, a right external connection and the like. The cooperation party can be a mobile phone APP, an online shopping website, an offline shopping platform, an online or offline tourism product selling platform and the like, and mainly depends on business requirements of the business district party and actual information which can be provided by the cooperation party;

5. acquiring track information of the alignment ID: after obtaining the encrypted alignment ID (i.e. the data sample shared by both parties), the partner queries the movement track information of the sample in the past year according to the own service data (such as mobile phone signaling). The signaling data is data of the current time and the current position of a mobile terminal acquired by a mobile operator at a fixed frequency;

6. compute dense sources of visiting users: firstly, after obtaining the aligned sample track information, the partner judges the sample residence place and the working place according to the track information, wherein the high-frequency residence point of the logic sample at 9-17 points on the working day is judged as the working place, and the high-frequency residence point of the sample at 10-8 points later is judged as the residence place (residence point: the position where the sample track information stays for a long time). And then, calculating the residence and the working place of all samples, and clustering the residence and the working place by adopting a space clustering algorithm to obtain a clustering region. Finally, 10 clustering areas with the most dense distribution of visiting users in the space are screened out from the clustering results and are displayed in a map, namely ten dense source places in front of the visiting users in the business district;

7. depicting the visitor group of the visitor with the image: and calculating the group images of the samples of all the source places after obtaining the place where the business circles visit the users and are ten dense source places. The population profile dimensions include: the method comprises the steps of determining the sample visit quantity in a fixed time period, the time distribution condition of the visited users (such as counting the number of the visited samples in different time periods), composition analysis, social graph analysis and the like, wherein the composition analysis can comprise structural characteristics such as gender, age, presence or absence of children, consumption capacity, shopping preference and the like; the social Graph analysis mainly utilizes relevant data information of social software, and extracts representations reflecting first-order and second-order similarity from the relations between relatives and friends and coworkers of the visiting user by utilizing a LINE algorithm in Graph Embedding, and the representations are used as a basis for matching the guest group image of the visiting user with the target new guest group image in the later period. Meanwhile, in the processing of the structural features, different processing is adopted according to the time attribute of the user label: for static features, the representations of the static features are stored in a content library to facilitate later use, for dynamic features such as features which change frequently along with time, such as user preference, interest and the like, long-term representation vectors can be calculated by taking days and weeks as units, short-term representation vectors can be calculated by taking hours and minutes as units, weighting processing is carried out on the short-term representation vectors and the long-term representation vectors, the weighting processing is used for calculating the matching degree of the later-stage features and the target new customers, and therefore real-time user recommendation is achieved. The group image does not determine the specific image or label of the individual;

8. acquiring a target new guest group: and the partner obtains the user ID with higher similarity with the visiting user in the missed visiting user, namely the target new guest ID, from the obtained user portrait of the visiting user of the business circle and the historical user portrait of the business circle by a similar user mining algorithm. The similar user mining algorithm is as follows: firstly, carrying out category coding on structured image data (such as age, gender, consumption capability and the like) (such as a male marker of 0001 and a female marker of 1000) and obtaining a more compact image characterization vector through dimension reduction. Then, extracting the social network representation for the unstructured relationship (such as a social Graph) according to a LINE algorithm of Graph Embedding to obtain a network representation vector, and splicing the network representation vector and the portrait representation vector together. And finally, calculating the similarity between the samples according to the characterization vectors of the samples, wherein the samples exceeding a certain threshold value are the target new passenger population.

9. Acquiring a target new guest group dense source: obtaining a target new passenger group, and calculating the top ten dense source places of the target new passenger group by using the method in the step 6;

10. and returning a result: and returning the information to a third-party arbitration mechanism for gathering, and then sending the information to a business district operator by the arbitration mechanism for subsequent marketing planning. The partner can be one or more, depending on the marketing content;

11. the actual marketing campaign is conducted by the business community.

Fig. 3 is a schematic diagram of the process of constructing the guest group images and mining new guests according to the embodiment of the present invention. As shown in fig. 3, after user feature extraction and dimension reduction are performed respectively by the business community party and the partner party, a group image is obtained, and both user i and user j shown in the figure represent the group image. And then, obtaining an intermediate result by carrying out similarity comparison on the group images, and then carrying out homomorphic encryption on the intermediate result and sending the intermediate result to a third party arbitration mechanism. In the whole process, the third party arbitration mechanism, the business district party or the partner can not directly acquire the original data of other parties, and the original data are the acquired data after encryption processing or the data after privacy processing, so that the safety and the privacy of the data are ensured.

Fig. 4 is a schematic diagram of main blocks of an information pushing apparatus according to an embodiment of the present invention. As shown in fig. 4, the apparatus 400 for pushing information according to the embodiment of the present invention mainly includes a user identifier alignment module 401, a user source location determining module 402, a guest group representation calculating module 403, and a target user determining module 404.

The user identifier alignment module 401 is configured to obtain a user identifier set to be aligned, and perform identifier alignment processing on the user identifier set to be aligned and a business district user identifier set to obtain an aligned user identifier;

a user source determining module 402, configured to obtain the movement trajectory data of the aligned user identifier according to service data, and determine a dense source of the visited user based on the movement trajectory data;

a guest group portrait calculation module 403, configured to calculate a guest group portrait of each visiting user in the dense source area according to the feature data of the visiting user;

and the target user determining module 404 is configured to compare the similarity between the customer group image and the historical user image of the business district to determine a target user, and perform information push on the target user.

According to one embodiment of the present invention, the tag alignment process includes: taking intersection processing, taking union processing, left outer connection processing or right outer connection processing.

According to another embodiment of the present invention, the subscriber identity alignment module 401 may be further configured to: and sending the user identifier set to be aligned to a third party arbitration mechanism so that the third party arbitration mechanism carries out identifier alignment processing according to the user identifier set to be aligned and the business district user identifier set.

According to another embodiment of the present invention, the information pushing apparatus 400 further includes an identification set preprocessing module (not shown in the figure) configured to:

before carrying out identification alignment processing on the user identification set to be aligned and a business district user identification set, acquiring a processing rule of a user identification agreed with a business district in advance, wherein the processing rule comprises a data format and an encryption mode of the user identification;

According to yet another embodiment of the invention, the user-origin-determining module 402 may be further configured to:

According to yet another embodiment of the present invention, the target user determination module 604 may be further configured to:

According to the technical scheme of the embodiment of the invention, the alignment user identification is obtained by obtaining the user identification set to be aligned and carrying out identification alignment treatment on the user identification set to be aligned and the business district user identification set; obtaining moving track data aligned with the user identification according to the service data, and determining a dense source area of the visiting user based on the moving track data; respectively calculating the guest group portraits of the visiting users in each intensive source area according to the feature data of the visiting users; the similarity comparison is carried out on the customer group portrait and the historical user portrait of the business district to determine a target user, and information pushing is carried out on the target user, so that the user portrait portrayal and the information pushing can be carried out by gathering multi-party data, the defect of single characteristic dimension of user data under the business district operation scene is overcome, the source of a user who does not visit the business district is fully mined, and information pushing such as advertisements is better carried out under the new customer drainage scene of the business district; based on the space-time attribute of the user trajectory information, the accurate dense source of the user is obtained, so that the user customer group image can be depicted better, more timely and more accurately, potential customers of the business circle can be determined more accurately, and the information push success rate is improved conveniently.

According to the technical scheme, firstly, the time attribute of the track is utilized to screen out an accurate residence place as a source place of a user; and then, the spatial attributes are utilized, the exploration and excavation of a target new passenger source area are realized through the existing passenger flow analysis by adopting an OPTIC density clustering algorithm and a similar user excavation algorithm, the potential target passenger source information is locked, and the problems of 'facing to, advertising where, how, and when' are solved, and the efficiency of drainage planning and marketing is greatly improved. When the customer group image is constructed, the user characteristics come from the label characteristics of the trade area, the label characteristics of the e-commerce trade mark and the social graph characteristics of the social platform. The feature vector data of each party cannot be aggregated to one party to calculate the cosine similarity. Therefore, by adopting a homomorphic encryption technology, the results obtained after different feature vectors of the three-party platform are calculated are encrypted by a public key and are sent to the third-party arbitration mechanism, and the third-party arbitration mechanism decrypts the results by a private key and then sends the cosine similarity results to the business district.

Fig. 5 shows an exemplary system architecture 500 of an information pushing method or an information pushing apparatus to which an embodiment of the present invention may be applied.

As shown in fig. 5, the system architecture 500 may include

terminal devices

501, 502, 503, a network 504, and a server 505. The network 504 is the medium used to provide communication links between

terminal devices

501, 502, 503 and the server 505. Network 504 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal devices

501, 502, 503 to interact with a server 505 over a network 504 to receive or send messages or the like. The

terminal devices

501, 502, 503 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, advertisement push tools, social platform software, etc. (by way of example only).

The

terminal devices

501, 502, 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 505 may be a server providing various services, such as a background management server (for example only) providing support for shopping websites browsed by users using the

terminal devices

501, 502, 503. The background management server can acquire a to-be-aligned user identifier set from data such as a received information pushing request, and perform identifier alignment processing on the to-be-aligned user identifier set and a business district user identifier set to obtain an aligned user identifier; obtaining moving track data of the aligned user identification according to service data, and determining a dense source of a visiting user based on the moving track data; respectively calculating the guest group portraits of the visiting users in each intensive source area according to the feature data of the visiting users; and comparing the similarity of the customer group image with the historical user images of the business circles to determine a target user, performing information push and other processing on the target user, and feeding back processing results (such as information push results and product information push results, which are only examples) to the terminal equipment.

It should be noted that the method for pushing information provided by the embodiment of the present invention is generally executed by the server 505, and accordingly, a device for pushing information is generally disposed in the server 505.

It should be understood that the number of terminal devices, networks, and servers in fig. 5 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 6, a block diagram of a computer system 600 suitable for use with a terminal device or server implementing an embodiment of the invention is shown. The terminal device or the server shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.

As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU)601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary, so that a computer program read out therefrom is mounted in the storage section 608 as necessary.

In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609, and/or installed from the removable medium 611. The computer program performs the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 601.

It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present invention, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units or modules described in the embodiments of the present invention may be implemented by software, or may be implemented by hardware. The described units or modules may also be provided in a processor, and may be described as: a processor includes a user identification alignment module, a user source determination module, a guest community representation calculation module, and a target user determination module. The names of these units or modules do not form a limitation on the units or modules themselves in some cases, for example, the user identifier alignment module may also be described as "a module for obtaining a set of user identifiers to be aligned and performing identifier alignment processing on the set of user identifiers to be aligned and a set of business district user identifiers to obtain aligned user identifiers".

As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: acquiring a user identifier set to be aligned, and performing identifier alignment processing on the user identifier set to be aligned and a business district user identifier set to obtain an aligned user identifier; obtaining moving track data of the aligned user identification according to service data, and determining a dense source of a visiting user based on the moving track data; respectively calculating the guest group portraits of the visiting users in each intensive source area according to the feature data of the visiting users; and comparing the similarity of the customer group image with the historical user image of the business district to determine a target user, and pushing information of the target user.

According to the technical scheme of the embodiment of the invention, the alignment user identification is obtained by obtaining the user identification set to be aligned and carrying out identification alignment treatment on the user identification set to be aligned and the business district user identification set; obtaining moving track data aligned with the user identification according to the service data, and determining a dense source area of the visiting user based on the moving track data; respectively calculating the guest group portraits of the visiting users of each dense source area according to the feature data of the visiting users; the similarity comparison is carried out on the customer group portrait and the historical user portrait of the business district to determine a target user, and information pushing is carried out on the target user, so that the user portrait portrayal and the information pushing can be carried out by gathering multi-party data, the defect of single characteristic dimension of user data under the business district operation scene is overcome, the source of a user who does not visit the business district is fully mined, and information pushing such as advertisements is better carried out under the new customer drainage scene of the business district; the accurate dense source of the user is obtained based on the time-space attribute of the user track information, so that the user customer group image can be better, more real-time and more accurately depicted, potential customers of the business circle can be more accurately determined, and the information push success rate is convenient to improve.

The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method of information push, comprising:

respectively calculating the guest group portraits of the visiting users of each dense source area according to the feature data of the visiting users;

and comparing the similarity of the customer group image with the historical user image of the business district to determine a target user, and pushing information of the target user.

2. The method of claim 1, wherein the identity alignment process comprises:

3. The method of claim 1, wherein performing identifier alignment processing on the to-be-aligned user identifier set and a business district user identifier set comprises:

4. The method according to claim 1, further comprising, before performing identifier alignment processing on the to-be-aligned user identifier set and the business district user identifier set:

5. The method of claim 1, wherein determining a dense source of visiting users based on the movement trajectory data comprises:

and selecting a specified number of clustering regions with the visiting users distributed most densely in the space from the plurality of clustering regions as dense sources of the visiting users.

6. The method of claim 1, wherein comparing the guest group imagery to a similarity of historical user imagery of a business district to determine a target user comprises:

7. The method of claim 1, wherein pushing information to the target user comprises:

8. An information pushing apparatus, comprising:

the guest group image calculation module is used for calculating the guest group images of the visiting users in each dense source area according to the feature data of the visiting users;

9. An electronic device for pushing information, comprising:

one or more processors;

a storage device to store one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-7.

10. A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.