Enterprise and user portrait crowd circling method and system

Abstract

The invention provides an enterprise and user portrait crowd circling method and system. The method comprises the following steps: acquiring basic information, behavior data and business data of an enterprise and a user; constructing basic information labels, behavior data labels and business data labels of the enterprises and the users based on the basic information, the behavior data and the business data of the enterprises and the users; constructing an enterprise label table based on the basic information label, the behavior data label and the business data label of the enterprise, and adding enterprise label codes corresponding to the basic information label, the behavior data label and the business data label in the enterprise label table; constructing a user tag table based on the basic information tag, the behavior data tag and the business data tag of the user, and adding user tag codes corresponding to the basic information tag, the behavior data tag and the business data tag in the user tag table; and delineating the enterprise and the user based on the enterprise tag code and the user tag code.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a method and system for selecting enterprise and user portrait groups. Background Art

[0002] In today's era of information explosion and big data, data has become an indispensable core resource for corporate decision-making and strategic planning. With the rapid development of technology, companies are facing unprecedented market opportunities and challenges. In order to accurately locate and capture target customer groups in this ocean of data, companies are increasingly relying on the construction and analysis of user portraits and corporate portraits. User portraits, as an in-depth description of consumer behavior, preferences, needs and other characteristics, help companies understand their target markets; while corporate portraits focus on the company's operating model, market competitive position, development potential and other dimensions, providing an important basis for cooperation and competition strategies between companies.

[0003] Traditional methods of selecting users and enterprises, such as intuitive judgment based on experience, simple questionnaires or limited database screening, are often limited by insufficient data scale, processing capabilities and analytical depth, and it is difficult to fully and accurately capture the complex characteristics and dynamic changes of users and enterprises. This extensive selection method is not only inefficient, but also easily ignores the subtle differences in market segments, resulting in a lack of relevance and effectiveness of marketing strategies, and ultimately resulting in a waste of resources and missed market opportunities.

[0004] Therefore, the existing technology lacks a rule calculation method that can efficiently and accurately select target companies and users at the same time, mine valuable information from massive data, and build more detailed and multi-dimensional user and company portraits. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method for selecting enterprise and user portrait groups to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a method for selecting a group of enterprises and user portraits, the method comprising the following steps:

[0007] Obtain basic information, behavioral data, and business data of enterprises and users;

[0008] Build the basic information label, behavior data label and business data label of the enterprise based on the basic information, behavior data and business data of the enterprise;

[0009] Build user basic information tags, behavior data tags and business data tags based on user basic information, behavior data and business data;

[0010] Building an enterprise tag table based on the basic information tag, behavior data tag and business data tag of the enterprise, and adding enterprise tag codes corresponding to the basic information tag, behavior data tag and business data tag in the enterprise tag table;

[0011] Building a user tag table based on the user's basic information tag, behavior data tag, and business data tag, and adding user tag codes corresponding to the basic information tag, behavior data tag, and business data tag to the user tag table;

[0012] Enterprises and users are identified based on the enterprise tag codes and user tag codes.

[0013] The above scheme is adopted. This scheme adds labels to enterprises and users through their basic information, behavioral data and business data, and then constructs the labels into codes to obtain portraits of enterprises and users. Then, the codes of enterprises and users are calculated through preset delineation methods to complete the delineation of enterprises and users. This scheme can identify target markets and conduct precision marketing through user and enterprise portraits through flexible delineation methods.

[0014] In some embodiments of the present invention, in the step of constructing an enterprise's basic information tags, behavior data tags, and business data tags based on the enterprise's basic information, behavior data, and business data, or constructing a user's basic information tags, behavior data tags, and business data tags based on the user's basic information, behavior data, and business data, a preset encoding method is used to encode the basic information, behavior data, and business data, and the encoded data is identified to determine the corresponding basic information tags, behavior data tags, and business data tags.

[0015] In some embodiments of the present invention, in the step of encoding the basic information, behavior data and business data using a preset encoding method, a recurrent neural network encoder, a long short-term memory network encoder or a word embedding encoder is used.

[0016] In some embodiments of the present invention, in the step of identifying the encoded data and determining the corresponding basic information tags, behavior data tags and business data tags, a pre-trained neural network model is used to identify the basic information tags, behavior data tags and business data tags of the encoded data.

[0017] In some embodiments of the present invention, in the step of adding enterprise tag codes corresponding to basic information tags, behavior data tags, and business data tags in the enterprise tag table or adding user tag codes corresponding to basic information tags, behavior data tags, and business data tags in the user tag table, a preset comparison table is used to determine the user tag codes corresponding to the basic information tags, behavior data tags, and business data tags.

[0018] In some embodiments of the present invention, in the step of defining enterprises and users based on the enterprise tag codes and user tag codes: enterprises and users are grouped based on the enterprise tag codes and user tag codes, respectively.

[0019] In some embodiments of the present invention, in the step of grouping enterprises and users based on the enterprise tag codes and user tag codes, a combination of enterprise tag codes or a combination of user tag codes is correspondingly set in each group, and the group to which the user and enterprise belong is determined based on the enterprise tag codes and user tag codes of the user and enterprise.

[0020] In some embodiments of the present invention, the method further comprises:

[0021] There are related relationships between the groups of enterprises and users;

[0022] Filter relevant user groups based on the correlation of the groups to which the enterprise belongs;

[0023] Filter related enterprise groups based on the related relationships of the groups to which the user belongs.

[0024] In some embodiments of the present invention, in the correlation relationships set between the groups of enterprises and users, each correlation relationship corresponds to a different group of enterprises and users.

[0025] By adopting the above scheme, the present scheme can set different correlations, and screen out different combinations of push purposes through different correlations, thereby ensuring the flexibility of the scheme.

[0026] The second aspect of the present invention also provides a system for selecting enterprise and user portrait groups, which includes a computer device, wherein the computer device includes a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method described above.

[0027] The third aspect of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps implemented by the aforementioned method for selecting enterprise and user portrait groups.

[0028] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention can be specifically pointed out and obtained in the specification and the accompanying drawings.

[0029] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention.

[0031] Figure 1 A schematic diagram of an implementation method of the enterprise and user portrait crowd selection method of the present invention;

[0032] Figure 2 Schematic diagram of the processing architecture of the enterprise and user portrait population selection method of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0034] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0035] The existing technologies have the following shortcomings in user and enterprise profile selection:

[0036] Separability: Traditional methods tend to focus only on users or businesses, and fail to consider the characteristics of both.

[0037] Low efficiency: The selection rules are complex to set and the response speed is slow, which affects the real-time marketing effect;

[0038] Lack of accuracy: Existing selection rules often lack flexibility and are difficult to adapt to market changes.

[0039] The present invention can provide a rule calculation method capable of simultaneously selecting enterprises and users, so as to improve the efficiency and accuracy of the selection.

[0040] like Figure 1 and 2 As shown, the present invention proposes a method for selecting enterprise and user portrait groups, the steps of the method comprising:

[0041] Step S100, obtaining basic information, behavior data and business data of the enterprise and the user;

[0042] During the specific implementation process, the basic information of the enterprise and the user includes the user's identity information, which may be an ID number and telephone number, etc., and the enterprise's identity information may be the enterprise's name, registered address and enterprise type, etc.; the enterprise's behavioral data may be production data, sales data and financial data, etc., and the user's behavioral data may be browsing data and interactive data, etc., and the interactive data may be likes and comments, etc.; the enterprise's business data may be inventory data and procurement data, etc., and the user's business data includes transaction data and feedback data, and the feedback data may be user satisfaction evaluation data.

[0043] Step S200, constructing basic information tags, behavior data tags and business data tags of the enterprise based on the basic information, behavior data and business data of the enterprise;

[0044] Step S300, constructing a user's basic information tag, behavior data tag, and business data tag based on the user's basic information, behavior data, and business data;

[0045] Step S400, constructing an enterprise tag table based on the basic information tag, behavior data tag and business data tag of the enterprise, and adding enterprise tag codes corresponding to the basic information tag, behavior data tag and business data tag in the enterprise tag table;

[0046] Step S500, constructing a user tag table based on the user's basic information tag, behavior data tag, and business data tag, and adding user tag codes corresponding to the basic information tag, behavior data tag, and business data tag to the user tag table;

[0047] In the specific implementation process, the enterprise tag table and the user tag table are both provided with data columns for representing enterprise tag codes and user tag codes, and the user tag codes correspond one-to-one to basic information tags, behavior data tags and business data tags.

[0048] Step S600: Circle the enterprise and the user based on the enterprise tag code and the user tag code.

[0049] In some embodiments of the present invention, in the step of circled enterprises and users based on the enterprise tag codes and user tag codes, the correlation is calculated based on the enterprise tag codes and user tag codes, and the related enterprises and users are determined based on the correlation, and the related enterprises and users are circled.

[0050] In a specific implementation process, the method of calculating the relevance based on the enterprise tag code and the user tag code may adopt Euclidean distance, cosine similarity, Manhattan distance or Jaccard similarity coefficient;

[0051] Euclidean distance, also known as Euclidean distance, is one of the most commonly used spatial distance metrics and is used to measure the straight-line distance between two points in multidimensional space;

[0052] Cosine similarity is a method commonly used to measure the directional similarity of two vectors in multidimensional space;

[0053] Manhattan distance, also known as L1 distance or city block distance, is a metric used in mathematics to measure the sum of the absolute distances between two points in a standard coordinate system;

[0054] The Jaccard similarity coefficient is an indicator to measure the similarity between two sets. The larger the value, the higher the similarity.

[0055] In the specific implementation process, in the current era of big data, enterprises need to identify target markets and conduct precision marketing through user and enterprise portraits. Traditional user and enterprise selection methods often cannot meet the needs of both users and enterprises at the same time, resulting in low marketing efficiency. Therefore, it is of great significance to develop a rule calculation method that can simultaneously select enterprises and users.

[0056] In the specific implementation process, this solution can determine the labels of related users and enterprises by calculating similarity.

[0057] The above scheme is adopted. This scheme adds labels to enterprises and users through their basic information, behavioral data and business data, and then constructs the labels into codes to obtain portraits of enterprises and users. Then, the codes of enterprises and users are calculated through preset delineation methods to complete the delineation of enterprises and users. This scheme can identify target markets and conduct precision marketing through user and enterprise portraits through flexible delineation methods.

[0058] In some embodiments of the present invention, in the step of constructing an enterprise's basic information tags, behavior data tags, and business data tags based on the enterprise's basic information, behavior data, and business data, or constructing a user's basic information tags, behavior data tags, and business data tags based on the user's basic information, behavior data, and business data, a preset encoding method is used to encode the basic information, behavior data, and business data, and the encoded data is identified to determine the corresponding basic information tags, behavior data tags, and business data tags.

[0059] In some embodiments of the present invention, in the step of encoding the basic information, behavior data and business data using a preset encoding method, a recurrent neural network encoder, a long short-term memory network encoder or a word embedding encoder is used.

[0060] The recurrent neural network (RNN) encoder is one of the earliest encoders used for NLP tasks. It is able to process sequence data and capture the temporal dependencies in the sequence. The RNN encoder passes each element (usually a word or character) in the input sequence to the network in turn and updates its internal state. Ultimately, the output of the RNN encoder can be regarded as a fixed-length representation of the input sequence;

[0061] The Long Short-Term Memory Network (LSTM) encoder is a variant of RNN. It solves the gradient vanishing and gradient exploding problems during long sequence training by introducing a gating mechanism (input gate, forget gate, and output gate). The LSTM encoder can more effectively capture long-term dependencies in a sequence, and therefore performs well in many NLP tasks.

[0062] In natural language processing, word embedding is a technique for mapping vocabulary to a high-dimensional vector space. Word embedding encoders convert words or phrases into fixed-length vector representations that capture the semantic relationship between words. Common word embedding methods include Word2Vec, GloVe, and BERT. These word embedding encoders play an important role in tasks such as text classification, sentiment analysis, and named entity recognition.

[0063] In some embodiments of the present invention, in the step of identifying the encoded data and determining the corresponding basic information tags, behavior data tags and business data tags, a pre-trained neural network model is used to identify the basic information tags, behavior data tags and business data tags of the encoded data.

[0064] In some embodiments of the present invention, in the step of adding enterprise tag codes corresponding to basic information tags, behavior data tags, and business data tags in the enterprise tag table or adding user tag codes corresponding to basic information tags, behavior data tags, and business data tags in the user tag table, a preset comparison table is used to determine the user tag codes corresponding to the basic information tags, behavior data tags, and business data tags.

[0065] In some embodiments of the present invention, in the step of defining enterprises and users based on the enterprise tag codes and user tag codes: enterprises and users are grouped based on the enterprise tag codes and user tag codes, respectively.

[0066] In some embodiments of the present invention, in the step of grouping enterprises and users based on the enterprise tag codes and user tag codes, a combination of enterprise tag codes or a combination of user tag codes is correspondingly set in each group, and the group to which the user and enterprise belong is determined based on the enterprise tag codes and user tag codes of the user and enterprise.

[0067] In the specific implementation process, a fixed combination of enterprise label codes and user label codes is set in each group, and users and enterprises with the above combination are set in the group.

[0068] In some embodiments of the present invention, the method further comprises:

[0069] There are related relationships between the groups of enterprises and users;

[0070] Filter relevant user groups based on the correlation of the groups to which the enterprise belongs;

[0071] Filter related enterprise groups based on the related relationships of the groups to which the user belongs.

[0072] By adopting the above scheme, the present scheme can set different correlations, and screen out different combinations of push purposes through different correlations, thereby ensuring the flexibility of the scheme.

[0073] In some embodiments of the present invention, in the correlation relationships set between the groups of enterprises and users, each correlation relationship corresponds to a different group of enterprises and users.

[0074] In the specific implementation process, the format conversion of the label table and how the data in the label bitmap result table is generated through rule setting for user and enterprise crowd package information specifically include the following modules:

[0075] Data collection module: collects basic information, behavioral data, and business data of users and enterprises.

[0076] Feature extraction module: extracts user and enterprise attribute information, behavior data and related business data from the collected data, calculates and processes them into the characteristic information of the enterprise and user, and stores them in the label table. Each label is stored in a separate table, called the label detail table. (where id is the unique mapping value of enterprise id and user id, cid is enterprise id, uid is user id, the following is an example, test fake data)

[0077] Format conversion module: converts the format of all tag detail table data and stores them in the tag wide table. (where id is the unique mapping value of enterprise id and user id, cid is enterprise id, uid is user id, the following is an example, test fake data)

[0078] Format conversion module: extract enterprises, users and their labels and label values ​​from the label wide table, group them by label and label value, aggregate and store the unique mapping values ​​of enterprises and users as the Doris database bitmap data type, stored in binary format and invisible. (The following four tables are divided into four tables according to data type, one for text, one for date, one for number, and one for decimal)

[0079] Circle the calculation module:

[0080] By using the set selection rules, users and companies are screened and the target population is finally obtained. The rule calculation module is used to combine the characteristics of users and companies to achieve efficient population screening and selection.

[0081] The calculation order guarantees:

[0082] The stack technology in Java is used to ensure that rule calculations are executed in the expected order, and expressions in parentheses are calculated first to ensure the correct logical calculation order.

[0083] Call the push and pop methods, set the rule to start popping when it encounters ")" and stop popping when it encounters "(", and assemble the order of rule calculation during the push and pop process.

[0084] SQL calculation logic optimization:

[0085] Use the bitmap_union() provided by Doris database to process the crowd package aggregation function of multiple rules, use bitmap_and() to process the intersection operation, and use bitmap_or() to process the union operation. These functions can be used to efficiently execute the calculation SQL.

[0086] By rationally assembling SQL logic, the functions of Doris database can be fully utilized to improve computing efficiency and accuracy.

[0087] During the rule calculation process, the bitmap advanced function explode_bitmap (unique mapping value of enterprise id and user id) is used to extract the id, and then associate the cid and uid to calculate the enterprise population package, user population package and result population package (enterprise and user as unique keys).

[0088] Result storage optimization:

[0089] The calculation results are stored in the user grouping table using a bitmap data storage structure to improve data query efficiency and management convenience.

[0090] Set the table structure to use a clustered table, retain historical partitions for 7 days, and automatically clean up expired data to ensure the timeliness and accuracy of the data.

[0091] Maintain effective management of database performance and storage space by regularly cleaning up expired data.

[0092] The beneficial effects of this program include:

[0093] Comprehensiveness: It can take into account the characteristics of both users and enterprises at the same time to improve the comprehensiveness of the selection.

[0094] Efficiency: By optimizing the selection rules, the speed and responsiveness of the selection are improved.

[0095] Flexibility: Supports dynamic adjustment of selection criteria to adapt to market changes.

[0096] An embodiment of the present invention also provides a system for selecting enterprise and user portrait groups, which includes a computer device, wherein the computer device includes a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method described above.

[0097] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps implemented by the aforementioned enterprise and user profile crowd selection method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0098] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0099] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0100] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or replace features of other embodiments.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for selecting enterprise and user portrait groups, characterized in that: The steps of the method include: Obtain basic information, behavioral data, and business data of enterprises and users; Build the basic information label, behavior data label and business data label of the enterprise based on the basic information, behavior data and business data of the enterprise; Build user basic information tags, behavior data tags and business data tags based on user basic information, behavior data and business data; Building an enterprise tag table based on the basic information tag, behavior data tag and business data tag of the enterprise, and adding enterprise tag codes corresponding to the basic information tag, behavior data tag and business data tag in the enterprise tag table; Building a user tag table based on the user's basic information tag, behavior data tag, and business data tag, and adding user tag codes corresponding to the basic information tag, behavior data tag, and business data tag to the user tag table; Enterprises and users are identified based on the enterprise tag codes and user tag codes.

2. The method for selecting enterprise and user portrait groups according to claim 1 is characterized in that: In the step of constructing the enterprise's basic information tags, behavior data tags and business data tags based on the enterprise's basic information, behavior data and business data, or constructing the user's basic information tags, behavior data tags and business data tags based on the user's basic information, behavior data and business data, a preset encoding method is used to encode the basic information, behavior data and business data, the encoded data is identified, and the corresponding basic information tags, behavior data tags and business data tags are determined.

3. The method for selecting enterprise and user portrait groups according to claim 2 is characterized in that: In the step of encoding the basic information, behavior data and business data using a preset encoding method, a recurrent neural network encoder, a long short-term memory network encoder or a word embedding encoder is used.

4. The method for selecting enterprise and user portrait groups according to claim 2 is characterized in that: In the step of identifying the encoded data and determining the corresponding basic information labels, behavior data labels and business data labels, a pre-trained neural network model is used to identify the basic information labels, behavior data labels and business data labels of the encoded data.

5. The method for selecting enterprise and user portrait groups according to any one of claims 1 to 4, characterized in that: In the step of adding enterprise tag codes corresponding to basic information tags, behavior data tags and business data tags in the enterprise tag table or adding user tag codes corresponding to basic information tags, behavior data tags and business data tags in the user tag table, a preset comparison table is used to determine the user tag codes corresponding to the basic information tags, behavior data tags and business data tags.

6. The method for selecting enterprise and user portrait groups according to claim 5 is characterized in that: In the step of defining enterprises and users based on the enterprise tag codes and user tag codes: enterprises and users are grouped respectively based on the enterprise tag codes and user tag codes.

7. The method for selecting enterprise and user portrait groups according to claim 6, characterized in that: In the step of grouping enterprises and users based on the enterprise label codes and user label codes, a combination of enterprise label codes or a combination of user label codes is correspondingly set in each group, and the group to which the user and enterprise belong is determined based on the enterprise label codes and user label codes of the user and enterprise.

8. The method for selecting enterprise and user portrait groups according to claim 6, characterized in that: The method further comprises the steps of: There are related relationships between the groups of enterprises and users; Filter relevant user groups based on the correlation of the groups to which the enterprise belongs; Filter related enterprise groups based on the related relationships of the groups to which the user belongs.

9. The method for selecting enterprise and user portrait groups according to claim 8, characterized in that: In the correlation relationships set between the groups of enterprises and users, different groups of enterprises and users correspond to each correlation relationship.

10. A system for selecting enterprise and user portrait groups, characterized in that: The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions. The processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method as described in any one of claims 1 to 9.

Images