CN115204450A - Purchasing behavior prediction method and device, electronic equipment and storage medium - Google Patents
Purchasing behavior prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115204450A CN115204450A CN202210583291.6A CN202210583291A CN115204450A CN 115204450 A CN115204450 A CN 115204450A CN 202210583291 A CN202210583291 A CN 202210583291A CN 115204450 A CN115204450 A CN 115204450A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user
- purchasing behavior
- historical
- generate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Accounting & Taxation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present disclosure proposes a purchasing behavior prediction method, apparatus, electronic device, and storage medium, the method comprising: acquiring user purchase data, and acquiring associated commodity order information from the user purchase data according to the commodity embedded expression set; acquiring historical purchasing behavior data of a user in a target time period; analyzing and processing historical purchasing behavior data of the user to generate a suspected target personnel set; and carrying out purchasing behavior prediction according to the associated commodity order information and the suspected target personnel set so as to generate a prediction result. Therefore, the purchasing behavior of the enterprise user can be predicted, and the distribution mode of the commodities can be adjusted to reduce the distribution cost.
Description
Technical Field
The present disclosure relates to the field of data processing technologies, and in particular, to a purchasing behavior prediction method and apparatus, an electronic device, and a storage medium.
Background
With the explosion of e-commerce and logistics, more and more individuals and businesses choose to purchase goods online. In a logistics scene, an individual user generally makes an order at a main station (C (Consumer) end) of an e-commerce website and uses a C network for distribution, so that the requirement on timeliness is high, and the cost is high; the enterprise user generally makes an order at an enterprise purchasing platform (B (Business) end) and uses a B network for distribution, so that the time effect requirement is loose and the cost is lower.
However, sometimes enterprise users will place orders at the C terminal for discount benefits, and the main performance behavior is to split a large order into a plurality of small orders, and although the orders do not have the demand of timely delivery, the orders placed at the C terminal will be delivered by the C network in time, so that the logistics transportation cost is increased, and the normal delivery of other orders in the C network is influenced because large batches of orders break through inventory.
For example, a company may want to issue annual employee benefits, purchase multiple mobile phones, multiple tablet computers, and multiple cameras, and place an order at end C whenever the host site has a promotional offer. However, because the promotion preferential activities of the master station are regularly limited, the same account number cannot be used for purchasing, and therefore the enterprise arranges a plurality of employees to place orders according to a plurality of orders. After the order placing is successful, the E-business platform arranges the C network to deliver the orders, however, the orders should be placed at the B terminal and delivered by the B network, and the orders are placed at the C terminal and delivered by the C network logistics, so that the great transportation cost waste is caused.
Disclosure of Invention
The embodiment of the first aspect of the disclosure provides a purchasing behavior prediction method, which can predict a purchasing behavior of an enterprise user, so as to adjust a commodity distribution mode to reduce distribution cost.
The embodiment of the second aspect of the present disclosure provides a purchasing behavior prediction device.
An embodiment of a third aspect of the present disclosure provides an electronic device.
A fourth aspect of the present disclosure provides a computer-readable storage medium.
An embodiment of a first aspect of the present disclosure provides a purchasing behavior prediction method, including: acquiring user purchase data, and acquiring associated commodity order information from the user purchase data according to a commodity embedded expression set; acquiring historical purchasing behavior data of the user in the target time period; analyzing and processing the historical purchasing behavior data of the user to generate a suspected target personnel set; and carrying out purchasing behavior prediction according to the associated commodity order information and the suspected target personnel set so as to generate a prediction result.
According to the purchasing behavior prediction method disclosed by the embodiment of the disclosure, firstly, user purchasing data is obtained, associated commodity order information is obtained from the user purchasing data according to a commodity embedded expression set, then, user historical purchasing behavior data in a target time period is obtained, the user historical purchasing behavior data is analyzed and processed to generate a suspected target personnel set, and purchasing behavior prediction is carried out according to the associated commodity order information and the suspected target personnel set to generate a prediction result. Therefore, the purchasing behavior of the enterprise user can be predicted, and the distribution mode of the commodities can be adjusted to reduce the distribution cost.
In addition, the purchasing behavior prediction method according to the above embodiment of the present disclosure may further have the following additional technical features:
in one embodiment of the present disclosure, the commodity embedded expression set is obtained by: acquiring historical purchase data of enterprises in a target time period; analyzing and processing the historical enterprise purchase data to generate an embedded expression corresponding to each commodity in the historical enterprise purchase data; and generating the commodity embedded expression set according to the embedded expression corresponding to each commodity.
In an embodiment of the present disclosure, the obtaining historical purchase data of the enterprise within the target time period includes: and acquiring historical purchase data of the enterprise in the target time period from a database of the first client.
In an embodiment of the present disclosure, the analyzing the historical purchase data of the enterprise to generate an embedded expression corresponding to each commodity in the historical purchase data of the enterprise includes: acquiring each commodity in the historical enterprise purchase data and an enterprise user corresponding to each commodity; drawing according to each commodity and the enterprise user corresponding to each commodity to generate an undirected graph; and processing the undirected graph according to an embedded expression generation model to generate an embedded expression corresponding to each commodity.
In an embodiment of the present disclosure, the obtaining user purchase data and obtaining associated commodity order information from the user purchase data according to a commodity embedded expression set includes: acquiring unfinished order information from a database of a second client; acquiring commodity order information corresponding to the same receiving address and the same distribution time period from the order information, and taking the commodity order information as the user purchase data; acquiring a to-be-associated commodity set from the user purchase data according to the commodity embedded expression set, and acquiring an embedded expression corresponding to each to-be-associated commodity in the to-be-associated commodity set; and associating the commodities to be associated in the commodity set to be associated according to the association strategy and the embedded expression corresponding to each commodity to be associated so as to generate the order information of the associated commodities.
In an embodiment of the present disclosure, the obtaining of the historical purchasing behavior data of the user in the target time period includes: and acquiring the historical purchasing behavior data of the user in the target time period from the database of the second client.
In an embodiment of the present disclosure, the analyzing the historical purchasing behavior data of the user to generate a suspected target person set includes: obtaining an analysis model; and analyzing and processing the historical purchasing behavior data of the user according to the analysis model to generate the suspected target personnel set.
In an embodiment of the present disclosure, the predicting the purchasing behavior according to the associated commodity order information and the suspected target person set to generate a prediction result includes: respectively taking each suspected target person in the suspected target person set as an index, and inquiring from the associated commodity order information to generate an inquiry result; and carrying out purchasing behavior prediction according to the query result to generate the prediction result.
An embodiment of a second aspect of the present disclosure provides a purchasing behavior prediction apparatus, including: the first acquisition module is used for acquiring user purchase data and acquiring associated commodity order information from the user purchase data according to the commodity embedded expression set; the second acquisition module is used for acquiring historical purchasing behavior data of the user in the target time period; the first analysis processing module is used for analyzing and processing the historical purchasing behavior data of the user to generate a suspected target personnel set; and the prediction module is used for predicting the purchasing behavior according to the associated commodity order information and the suspected target personnel set so as to generate a prediction result.
According to the purchasing behavior prediction device, firstly, user purchasing data are obtained through a first obtaining module, related commodity order information is obtained from the user purchasing data according to a commodity embedded expression set, then, user historical purchasing behavior data in a target time period are obtained through a second obtaining module, the user historical purchasing behavior data are analyzed and processed through a first analyzing module, a suspected target personnel set is generated, and purchasing behavior prediction is conducted through a prediction module according to the related commodity order information and the suspected target personnel set, so that a prediction result is generated. Therefore, the purchasing behavior of the enterprise user can be predicted, and the distribution mode of the commodities can be adjusted to reduce the distribution cost.
In addition, the purchasing behavior prediction device according to the above embodiment of the present disclosure may further have the following additional technical features:
in one embodiment of the present disclosure, the third obtaining module is configured to obtain historical purchase data of the enterprise within a target time period; the second analysis processing module is used for analyzing and processing the enterprise historical purchase data to generate an embedded expression corresponding to each commodity in the enterprise historical purchase data; and the generating module is used for generating the commodity embedded expression set according to the embedded expression corresponding to each commodity.
In an embodiment of the disclosure, the third obtaining module is specifically configured to: and acquiring historical purchase data of the enterprise in the target time period from a database of the first client.
In an embodiment of the disclosure, the second analysis processing module is specifically configured to: acquiring each commodity in the historical enterprise purchase data and an enterprise user corresponding to each commodity; drawing according to each commodity and the enterprise user corresponding to each commodity to generate an undirected graph; and processing the undirected graph according to an embedded expression generation model to generate an embedded expression corresponding to each commodity.
In an embodiment of the disclosure, the first obtaining module is specifically configured to: acquiring unfinished order information from a database of a second client; acquiring commodity order information corresponding to the same receiving address and the same distribution time period from the order information, and taking the commodity order information as the user purchase data; acquiring a to-be-associated commodity set from the user purchase data according to the commodity embedded expression set, and acquiring an embedded expression corresponding to each to-be-associated commodity in the to-be-associated commodity set; and associating the commodities to be associated in the commodity set to be associated according to the association strategy and the embedded expression corresponding to each commodity to be associated so as to generate the associated commodity order information.
In an embodiment of the disclosure, the second obtaining module is specifically configured to: and acquiring the historical purchasing behavior data of the user in the target time period from the database of the second client.
In an embodiment of the disclosure, the second analysis processing module is specifically configured to: obtaining an analysis model; and analyzing and processing the historical purchasing behavior data of the user according to the analysis model to generate the suspected target personnel set.
In an embodiment of the disclosure, the prediction module is specifically configured to: respectively taking each suspected target person in the suspected target person set as an index, and inquiring from the associated commodity order information to generate an inquiry result; and carrying out purchasing behavior prediction according to the query result to generate the prediction result.
An embodiment of a third aspect of the present disclosure provides an electronic device, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method for purchasing behavior prediction as described in the foregoing embodiments of the first aspect when executing the program.
The electronic device of the embodiment of the disclosure can predict the purchasing behavior of the enterprise user by executing the computer program stored in the memory through the processor, thereby adjusting the distribution mode of the commodity to reduce the distribution cost.
A fourth aspect of the present disclosure provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the purchasing behavior prediction method according to the first aspect of the present disclosure.
The computer-readable storage medium of the embodiment of the present disclosure, by storing a computer program and being executed by a processor, can predict a purchasing behavior of an enterprise user, thereby adjusting a delivery manner of a commodity to reduce a delivery cost.
Additional aspects and advantages of the disclosure will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the disclosure.
Drawings
The foregoing and/or additional aspects and advantages of the present disclosure will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic flow chart diagram of a method for purchasing behavior prediction according to one embodiment of the present disclosure;
FIG. 2 is a schematic flow chart diagram of a method for purchasing behavior prediction according to another embodiment of the present disclosure;
FIG. 3 is a schematic flow chart diagram of a purchasing behavior prediction method according to another embodiment of the present disclosure;
FIG. 4 is a schematic flow chart diagram of a purchasing behavior prediction method according to another embodiment of the present disclosure;
FIG. 5 is a schematic flow chart diagram of a purchasing behavior prediction method according to another embodiment of the present disclosure;
FIG. 6 is a schematic flow chart diagram of a purchasing behavior prediction method according to another embodiment of the present disclosure;
FIG. 7 is an undirected graph according to one embodiment of the present disclosure;
figure 8 is a schematic diagram of a node2vec walk strategy according to one embodiment of the present disclosure;
figure 9 is a schematic diagram of a node2vec walk strategy according to another embodiment of the present disclosure;
FIG. 10 is a schematic diagram of a Skip-gram model according to one embodiment of the present disclosure;
FIG. 11 is a flowchart illustrating a specific example of a method for purchasing behavior prediction according to one embodiment of the present disclosure;
FIG. 12 is a block schematic diagram of a purchasing behavior prediction apparatus according to one embodiment of the present disclosure; and
fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
Detailed Description
Reference will now be made in detail to the embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are illustrative and intended to explain the present disclosure, and should not be construed as limiting the present disclosure.
A purchasing behavior prediction method, apparatus, electronic device, and storage medium according to an embodiment of the present disclosure are described below with reference to the drawings.
The purchasing behavior prediction method provided by the embodiment of the present disclosure may be executed by an electronic device, which may be a PC (Personal Computer), a tablet Computer, a server, or the like, and is not limited herein.
In the disclosed embodiment, the electronic device may be provided with a processing component, a storage component and a driving component. Alternatively, the driving component and the processing component may be integrated, the storage component may store an operating system, an application program, or other program modules, and the processing component implements the purchasing behavior prediction method provided by the embodiment of the disclosure by executing the application program stored in the storage component.
Fig. 1 is a flow chart of a purchasing behavior prediction method according to one embodiment of the present disclosure.
The purchasing behavior prediction method of the embodiment can be further executed by the purchasing behavior prediction device provided by the embodiment of the disclosure, and the device can be configured in electronic equipment to obtain user purchasing data, obtain associated commodity order information from the user purchasing data according to a commodity embedded expression set, then obtain user historical purchasing behavior data in a target time period, analyze and process the user historical purchasing behavior data to generate a suspected target personnel set, and perform purchasing behavior prediction according to the associated commodity order information and the suspected target personnel set to generate a prediction result, so that purchasing behaviors of enterprise users can be predicted, and further, a distribution mode of commodities is adjusted to reduce distribution cost.
As a possible situation, the purchasing behavior prediction method according to the embodiment of the present disclosure may also be executed at a server, where the server may be a cloud server, and the purchasing behavior prediction method may be executed at a cloud end.
As shown in fig. 1, the method for predicting purchasing behavior may include:
In the embodiment of the disclosure, each commodity has complex semantic information, and for a commodity having a special association relationship, the association relationship of such a commodity cannot be determined through simple text similarity matching or category attributes, for example, a mobile phone and an earphone which are often purchased simultaneously have a special association relationship which is often purchased simultaneously, but the mobile phone and the earphone do not have any similarity in text, and do not belong to the same category in the commodity classification, and the mobile phone and the earphone cannot be associated through the text similarity or the category attributes, that is, the special association relationship which is often purchased simultaneously between the mobile phone and the earphone cannot be determined through the text similarity or the category attributes, and at this time, a high-dimensional representation is required to determine the special association relationship which is often purchased simultaneously between the commodities.
The embodiment of the disclosure can reflect whether the special association relationship frequently purchased at the same time exists between the commodities through the embedded expression of each commodity, i.e. whether the special association relationship frequently purchased at the same time exists between the commodities can be determined through the embedded expression of each commodity. For example, the mobile phone and the headset which have the special association relationship frequently purchased at the same time can be determined to have the special association relationship frequently purchased at the same time through the embedded expression of the mobile phone and the headset.
The commodity embedded expression set can comprise embedded expressions corresponding to each commodity, wherein the embedded expressions corresponding to the commodities can reflect special association relations among the commodities. It should be noted that the commodity embedded expression set described in the embodiments of the present disclosure may be pre-constructed.
Specifically, the server can acquire user purchase data from the user purchase platform, acquire a pre-constructed commodity embedded expression set, and acquire associated commodity order information from the user purchase data according to the commodity embedded expression set.
To clearly illustrate the above embodiment, in an embodiment of the present disclosure, as shown in fig. 2, acquiring user purchase data, and acquiring associated item order information from the user purchase data according to an item embedded expression set, may include:
The second client can be a purchasing platform of an individual user, the user generally places an order to purchase commodities through the second client when purchasing in small batches, and the order in the second client can be distributed through the second distribution network, so that the time efficiency is high, but the distribution cost is high.
It should be noted that the incomplete orders described in this embodiment may be orders of incomplete deliveries in the database of the second client.
In the embodiment of the present disclosure, when a user places an order to purchase a commodity at a second client, order information of the purchase is generated, and the order information is stored in a database of the second client, so as to be called and used as needed.
The order information may include user information, commodity information, a receiving address, a delivery time period, and the like.
Specifically, the server may obtain the incomplete order information from a database of the second client.
Specifically, after acquiring the incomplete order information, the server may filter the incomplete order information to acquire the commodity order information corresponding to the same receiving address and the same delivery time period from the incomplete order information, and use the commodity order information as the user purchase data.
In the disclosed embodiment, if a commodity is often purchased simultaneously with other commodities, the commodity has a corresponding pair of embedded expressions; if an item is not frequently purchased simultaneously with other items, the item does not have a corresponding pair of embedded expressions.
The commodities in the user purchase data comprise two commodities: one item is an item with an embedded expression, and the other item is an item without an embedded expression.
Specifically, the server may search all the commodities with embedded expressions in the user purchase data as the commodities to be associated from the commodity embedded expression set, thereby obtaining a commodity set to be associated, and obtaining an embedded expression corresponding to each commodity to be associated in the commodity set to be associated.
It should be noted that the to-be-associated product set described in this embodiment may further include an order number corresponding to each product.
And 204, associating the commodities to be associated in the commodity set to be associated according to the association strategy and the embedded expression corresponding to each commodity to be associated so as to generate associated commodity order information. The association policy may be calibrated according to actual conditions and requirements, and is not limited herein.
In the embodiment of the present disclosure, after acquiring the to-be-associated commodity set and the embedded expression corresponding to each to-be-associated commodity, the server may associate the to-be-associated commodities in the to-be-associated commodity set according to the association policy and the embedded expression corresponding to each to-be-associated commodity, so as to generate associated commodity order information.
In particular, the commodity in the second client may consist of a 32-dimensional vector, denoted as H = H 1 ,h 2 ,...,h m In the embodiment of the disclosure, the commodities to be associated in the commodity set to be associated can be associated through a hierarchical clustering algorithm, wherein the hierarchical clustering algorithm is a bottom-up merging algorithm, two points which are most similar in all data points are combined by calculating the similarity between the two types of data points, and the process is iterated repeatedly. For two commodities a = a 1 ,a 2 ,...,a 32 ，B＝b 1 ,b 2 ,...,b 32 The present embodiment calculates the distance (similarity) between two commodities by the following formula (1) based on the euclidean distance.
Where dis (A, B) may represent the distance (similarity), x, of item A and item B i Can be a 1 ,a 2 ,...,a 32 ，y i Can be b 1 ,b 2 ,...,b 32 And i may represent a natural number.
Specifically, the distance between two to-be-associated commodities can be calculated through the formula (1), and if the distance between the two to-be-associated commodities is smaller than a distance threshold, the two to-be-associated commodities can be associated together (grouped into one type) to generate associated commodity order information.
It should be noted that the distance threshold described in this embodiment may be calibrated according to actual situations and requirements, and is not limited herein. Alternatively, the pitch threshold may be 0.5.
For clarity of the above embodiment, in one embodiment of the present disclosure, obtaining the historical purchasing behavior data of the user in the target time period may include obtaining the historical purchasing behavior data of the user in the target time period from a database of the second client.
In the embodiment of the disclosure, when a user places an order to purchase a commodity through the second client, the data of the current purchasing behavior of the user can be generated, and the purchasing data is stored in the database of the second client, so that the user can call the commodity for use as required.
The purchasing behavior data of the user may include user information, commodity information, purchasing time, and distribution time period of the commodity.
Specifically, the server may retrieve (acquire) the historical purchasing behavior data of the user in the target time period from the database of the second client.
And 103, analyzing and processing the historical purchasing behavior data of the user to generate a suspected target person set.
To clearly illustrate the above embodiment, in an embodiment of the present disclosure, as shown in fig. 3, the analyzing the historical purchasing behavior data of the user to generate the suspected target person set may include:
In the embodiment of the present disclosure, an enterprise may sometimes arrange several employees (purchasing personnel) to purchase a product by ordering at the second client using several different account numbers for discount benefit, and generally, the personnel arranging the purchasing by the enterprise is relatively fixed, and the purchasing personnel can be regarded as suspected target personnel.
Specifically, after the server obtains the historical purchasing behavior data of the user, an analysis model can be called from a storage space of the server, and the historical purchasing behavior data of the user is input into the analysis model, so that the historical purchasing behavior data of the user is processed through the analysis model, and a suspected target person set output by the analysis generation model is obtained.
As a possible scenario, a correlation mining method may be used to identify suspected target persons from the user of the second client.
Specifically, the suspected target person may be identified from the user of the second client using an association rule mining (Apriori) algorithm, which is mainly determined by the support degree (Sup (X)), and the support degree Sup (X) may be calculated by the following formula (2).
Here, the support Sup (X) may be expressed as a ratio between the number of times a certain user combination appears and the total number of times, sum (X) may be expressed as the number of times a certain user combination appears, and N may be expressed as the total number of times. For example, if a total of 10 orders have purchased a computer and the combination of user X and user Y appears 6 times, the support is 6/10=0.6. By searching for a frequent item set in a mode of iterating K times of mining, namely the item set with the support degree being greater than or equal to the minimum support degree threshold value min _ support, users who most often purchase goods at the second client side at the same time, namely suspected target persons, can be found. Based on experience, in the disclosed embodiment, a minimum support threshold min _ support =0.4, k =3 may be defined.
As another possible scenario, an anomaly detection method may be used to identify a suspected target person from the user of the second client.
Specifically, the number and frequency of the commodities purchased by the suspected target person may be different from those of the ordinary users, and the number and frequency are generally larger and higher. The number and frequency of purchasing the same commodity by the user in a certain time T can be counted, and abnormal users can be found in a sorting mode and can be identified as suspected target persons.
It should be noted that the certain time T described in this embodiment can be expressed according to actual situations and requirements, and is not limited herein. Alternatively, T =365 days.
And 104, predicting the purchasing behavior according to the associated commodity order information and the suspected target personnel set to generate a prediction result.
To clearly illustrate the above embodiment, in an embodiment of the present disclosure, as shown in fig. 4, the performing the purchasing behavior prediction according to the associated item order information and the suspected target person set to generate the prediction result may include:
And step 402, carrying out purchasing behavior prediction according to the query result to generate a prediction result.
The purchasing behavior may include an enterprise purchasing behavior and a personal purchasing behavior, where the enterprise purchasing behavior may be a behavior of an enterprise purchasing goods at a first client, the personal purchasing behavior may be a purchasing behavior of a (personal) user at a second client, and if the user is a purchasing person of the enterprise, a behavior of the user purchasing goods at the second client may also be referred to as an enterprise purchasing behavior.
Specifically, after acquiring the suspected target person set and the associated commodity order information, the server may respectively query, using each suspected target person in the suspected target person set as an index, an associated commodity order in which a purchaser is a suspected target person from the associated commodity order information, and may use a behavior of the purchaser (suspected target person) corresponding to the queried associated commodity order as a prediction basis to perform related prediction to obtain a prediction result (for example, prediction of a purchasing behavior of the first client in the second client). For example, whether the commodity order of the suspected target person is the related commodity order is judged, if yes, the purchasing behavior of the suspected target person is considered as the enterprise purchasing behavior; if not, the purchasing behavior of the suspected target person is considered as the personal purchasing behavior.
For example, assuming that the suspected target person set includes a user a, a user B, a user C, and a user D, the orders of the user a, the user B, the user C, and the user D at the second client are an order a, an order B, an order C, and an order D, respectively, if the orders of the order B and the order C are associated commodity orders, the user B and the user C may be considered as enterprise procurement persons, that is, the procurement behaviors of the user B and the user C may be considered as enterprise procurement behaviors, and the order B and the order C may be delivered through the delivery network of the first client, thereby reducing the delivery cost.
According to the purchasing behavior prediction method, firstly, user purchasing data is obtained, associated commodity order information is obtained from the user purchasing data according to a commodity embedded expression set, then, user historical purchasing behavior data in a target time period is obtained, the user historical purchasing behavior data is analyzed and processed to generate a suspected target personnel set, and finally, purchasing behavior prediction is carried out according to the associated commodity order information and the suspected target personnel set to generate a prediction result. Therefore, the purchasing behavior of the enterprise user can be predicted, and the distribution mode of the commodities can be adjusted to reduce the distribution cost.
In one embodiment of the present disclosure, as shown in fig. 5, the commodity embedded expression set is obtained by:
To clearly illustrate the above embodiment, in one embodiment of the present disclosure, obtaining historical purchase data for the business over the target time period may include obtaining historical purchase data for the business over the target time period from a database of the first client. It should be noted that the target time period described in this embodiment may be two weeks, one month, one year, or the like, and is not limited thereto.
The first client can be an enterprise purchasing platform, an enterprise generally places an order to purchase goods through the first client when purchasing in a large batch, and the order in the first client can be distributed through the first distribution network, so that the distribution time is loose, and the cost is low.
In the embodiment of the disclosure, when an enterprise places an order to purchase a commodity through a first client, the current purchase data of the enterprise can be generated, and the purchase data is stored in the database of the first client, so that the purchase data can be called and used as required.
The purchase data of the enterprise may include enterprise information, commodity information, purchase time, delivery time, and the like.
Specifically, the server may retrieve (acquire) the historical purchase data of the enterprise in the target time period from the database of the first client.
To clearly illustrate the above embodiment, in an embodiment of the present disclosure, as shown in fig. 6, the analyzing the historical purchase data of the enterprise to generate the embedded expression corresponding to each item in the historical purchase data of the enterprise may include:
Specifically, after obtaining the historical purchase data of the enterprise, the server may analyze the historical purchase data of the enterprise to obtain each commodity in the historical purchase data of the enterprise and an enterprise corresponding to each commodity.
In the embodiment of the present disclosure, after obtaining each commodity and the enterprise corresponding to each commodity, the server may perform mapping according to each commodity and the enterprise corresponding to each commodity based on a preset mapping algorithm, so as to generate an undirected graph. The preset mapping algorithm can be calibrated according to actual conditions.
Specifically, referring to fig. 7, after the server obtains each commodity and the enterprise corresponding to each commodity, each commodity may be used as a node, and v is used for each node 1 ,v 2 ,...,v n Each node is represented, if two commodities are purchased by the same enterprise at a first client, an edge can be established between the nodes corresponding to the two commodities, and therefore multiple edges can be established between the nodes corresponding to multiple commodities to generate an undirected graph.
Each edge of the undirected graph has a certain weight and can be called a weighted edge. If the number of times that two commodities are purchased at the same time is more, the weight of the weighted edge between the nodes corresponding to the two commodities is higher; the smaller the number of times two products are purchased simultaneously, the lower the weight of the weighted edge between the nodes corresponding to the two products. The weight of the weighted edge can be calculated by the following formula (3).
Wherein, w i Can represent a weighted edge e i Weight of (1), n i May represent the number of times two products having the authority to be connected are simultaneously purchased, max (n) is the maximum value of the number of times two products having the authority to be connected are simultaneously purchased, and i may be a natural number.
As a possible situation, after acquiring the each commodity and the enterprise corresponding to the each commodity, the server may perform mapping according to the enterprise corresponding to the each commodity and the each commodity through the mapping model to generate an undirected graph. It should be noted that the mapping model described in this embodiment may be trained in advance and pre-stored in the storage space of the server to facilitate retrieval of the application.
The training and the generation of the mapping model can be executed by a related training server, the training server can be a cloud server or a host of a computer, and a communication connection is established between the training server and the server capable of executing the purchasing behavior prediction method provided by the embodiment of the disclosure, wherein the communication connection can be at least one of a wireless network connection and a wired network connection. The training server can send the trained mapping model to the server so that the server can call the mapping model when needed, and therefore the computing pressure of the server is greatly reduced.
Specifically, after acquiring each commodity and the enterprise corresponding to each commodity, the server may call out a mapping model from its own storage space, and input each commodity and the enterprise corresponding to each commodity into the mapping model, so that the enterprise corresponding to each commodity and each commodity is processed through the mapping model to obtain an undirected graph output by the mapping model.
Specifically, after the undirected graph is acquired, the server may call an embedded expression generation model from its own storage space, and input the undirected graph to the embedded expression generation model, so that the undirected graph is processed by the embedded expression generation model to obtain an embedded expression corresponding to each commodity output by the embedded expression generation model.
As a possible scenario, the embedded expression generation model may be obtained by training based on a Node2vec (network extensible function learning) model, wherein the processing procedure of the undirected graph by the Node2vec model may be divided into the following two steps: generating an adjacent sequence of nodes by using a walking strategy; and step two, learning the embedded expression by using a skip-gram (such as a skip-gram model).
Step one, a migration generation sequence: see FIG. 8, which may be based on breadthA degree-First-Search (BFS) and Depth-First-Search (DFS) strategy generates a contiguous sequence of nodes of the undirected graph from the undirected graph. In order to control the proportion of two strategies, two hyperparameters p and q are introduced into a Node2vec model, and the transition probability alpha from one Node to different neighbor nodes of the Node can be calculated through the following formula (4) pq (t,x)。
Wherein, see diagram 9,t may represent the last node, v may represent the current node, x (x) 1 ,x 2 ,x 3 ) The next node may be represented. Walk from current node v, when the next walk is back to previous node t (from node v to node t), i.e. when d tx When the pressure is not higher than 0, the pressure is lower than 0,
when the next step of the traveling jumps to another neighbor node x of the node t 1 (from node v to node x) 1 ) When d is tx When =1, α pq (t, x) =1; when the next step walks to node x which is farther away 2 Or x 3 (from node v to node x) 2 Or from node v to node x 3 ) When d is tx When the ratio is not less than =2,
meanwhile, the node2vec model also considers the influence of the weighted edge weight, and the final bias coefficient can be calculated by the following formula (5).
π vx ＝α pq (t,x)×w vx (5)
Wherein, see FIG. 9, π vx May be the final bias coefficient (the transition probability of the non-regularized node v to node x), t may represent the last node, v may represent the current node, x (x) 1 ,x 2 ,x 3 ) May represent the next node, α pq (t, x) may be the transfer of a node to its different neighbor nodesProbability, w vx May be node v and node x (e.g., node x in FIG. 9) 1 ,x 2 ,x 3 ) With weights of the weighted edges in between.
Further, assume the start node is c 0 = v, the final walk policy, i.e., the formula for randomly walking to select the next node may be the following formula (6).
Wherein, P (c) i ＝x|c i-1 = v) may represent the probability of selecting the next node x from the current node v, E may represent the undirected graph described above, (v, x) may represent a weighted edge, π vx May be the final bias coefficient and Z may be a regularization constant. From the above equation (6), if there is a weighted edge (v, x) in the undirected graph E, the probability is used
And selecting the next node x, wherein the probability of selecting the next node x from the current node v is 0.
In the embodiment of the present disclosure, setting p =1,q =2 makes the random walk time more inclined to the node close to the current node (i.e., breadth search takes precedence), so as to reflect the situation that the commodities are purchased at the same time better, and thus, when learning the embedded expression, the associated commodities (the commodities often purchased at the same time) can be put together better.
And step two, learning and embedding the expression by using a skip-gram model in natural language processing. A schematic diagram of the skip-gram model can be seen in fig. 10, and the model can be composed of an input layer, a hidden layer and an output layer, where the hidden layer can be an embedded expression h. The learning of the model is optimized using the softmax loss function, which can be as the following equation (7).
Where t may be the last node, w (t) may be a prediction function, C may be the number of all goods, y may be the true value, i is the true node to which the current goods node is connected, and i may be a natural number.
In the embodiment of the present disclosure, the parameter h =32 of the embedded expression of the commodity may be set, and then the embedded expression of the commodity is learned through the skip-gram model to obtain the embedded expression of each commodity. In which, the commodities with similar semantics are similar in a high-dimensional space, so that the similar relationship (special association relationship often purchased at the same time) between the commodities can be reflected by the embedded expression of the commodities.
Specifically, after obtaining the embedded expression corresponding to each commodity, the server generates a commodity embedded expression set according to the embedded expression corresponding to each commodity.
It should be noted that the server in the embodiment of the present disclosure may be a server of an enterprise purchase platform, or a server of a user purchase platform, or another third-party server, which is not limited herein.
In the embodiment of the disclosure, the historical purchase data of the enterprise in the target time period is firstly acquired, then the historical purchase data of the enterprise is analyzed to generate the embedded expression corresponding to each commodity in the historical purchase data of the enterprise, and finally the embedded expression set of the commodity is generated according to the embedded expression corresponding to each commodity. Therefore, the embedded expression of each commodity in the historical enterprise purchase data can be determined according to the purchase relation among the commodities in the historical enterprise purchase data, and a commodity embedded expression set can be obtained.
In order to make the present disclosure more clearly understood for those skilled in the art, fig. 11 is a flowchart illustrating a specific example of a purchasing behavior prediction method according to an embodiment of the present disclosure.
As shown in fig. 11, the purchasing behavior prediction method of the embodiment of the present disclosure mainly includes three steps: step one, commodity embedded expression learning: drawing (undirected graph) is built according to each commodity and the enterprise corresponding to each commodity, and the embedded expression of the commodity is learned according to the undirected graph; step two, generating associated commodities: the method comprises the steps of obtaining commodity order information corresponding to the same receiving address and the same distribution time period, and associating commodities through hierarchical clustering according to the embedded expression of the commodities to obtain an associated commodity combination; step three, identifying the purchasing personnel: and finally, combining the commodities and the users according to the associated commodity combination and the suspected target personnel to generate an order combination, and considering the order combination as an order combination of purchasing personnel of the enterprise at a second client. Therefore, the purchasing behavior of the enterprise user at the second client can be predicted, and the enterprise user can be distributed through the second distribution network, so that the distribution cost is reduced.
Fig. 13 is a block diagram of a purchasing behavior prediction apparatus according to one embodiment of the present disclosure.
The purchasing behavior prediction device disclosed by the embodiment of the disclosure can be configured in electronic equipment to obtain user purchasing data, obtain associated commodity order information from the user purchasing data according to a commodity embedded expression set, then obtain user historical purchasing behavior data in a target time period, analyze and process the user historical purchasing behavior data to generate a suspected target personnel set, and then predict the purchasing behavior according to the associated commodity order information and the suspected target personnel set to generate a prediction result, so that the purchasing behavior of enterprise users can be predicted, and further, the distribution mode of commodities is adjusted to reduce the distribution cost.
As shown in fig. 12, the purchasing behavior prediction apparatus 1200 may include: a first acquisition module 1210, a second acquisition module 1220, a first analysis processing module 1230, and a prediction module 1240.
The first obtaining module 1210 is configured to obtain user purchase data, and obtain associated commodity order information from the user purchase data according to the commodity embedded expression set;
a second obtaining module 1220, configured to obtain historical purchasing behavior data of the user in the target time period;
the first analysis processing module 1230 is configured to analyze and process the historical purchasing behavior data of the user to generate a suspected target person set;
and the prediction module 1240 is used for predicting the purchasing behavior according to the associated commodity order information and the suspected target personnel set so as to generate a prediction result.
In an embodiment of the present disclosure, the purchasing behavior prediction apparatus 1200 may further include: a third acquisition module 1250, a second analysis processing module 1260, and a generation module 1270.
The third obtaining module 1250 is configured to obtain historical purchase data of the enterprise in the target time period;
the second analysis processing module 1260 is used for analyzing and processing the historical purchase data of the enterprise to generate an embedded expression corresponding to each commodity in the historical purchase data of the enterprise;
the generating module 1270 is configured to generate a commodity embedded expression set according to the embedded expression corresponding to each commodity.
In an embodiment of the present disclosure, the third obtaining module 1250 is specifically configured to: and acquiring historical purchase data of the enterprise in the target time period from a database of the first client.
In an embodiment of the present disclosure, the second analysis processing module 1260 is specifically configured to: obtaining each commodity in the historical purchase data of the enterprise and an enterprise user corresponding to each commodity; drawing is carried out according to each commodity and the enterprise user corresponding to each commodity so as to generate an undirected graph; and processing the undirected graph according to the embedded expression generation model to generate the embedded expression corresponding to each commodity.
In an embodiment of the present disclosure, the first obtaining module 1210 is specifically configured to: acquiring unfinished order information from a database of a second client; acquiring commodity order information corresponding to the same receiving address and the same distribution time period from the order information, and taking the commodity order information as user purchase data; according to the commodity embedded expression set, acquiring a commodity set to be associated from user purchase data, and acquiring an embedded expression corresponding to each commodity to be associated in the commodity set to be associated; and associating the commodities to be associated in the commodity set to be associated according to the association strategy and the embedded expression corresponding to each commodity to be associated so as to generate associated commodity order information.
In an embodiment of the disclosure, the second obtaining module 1220 is specifically configured to: and obtaining the historical purchasing behavior data of the user in the target time period from the database of the second client.
In an embodiment of the present disclosure, the second analysis processing module 1260 is specifically configured to: obtaining an analysis model; and analyzing and processing the historical purchasing behavior data of the user according to the analysis model to generate a suspected target person set.
In an embodiment of the present disclosure, the prediction module 1240 is specifically configured to: respectively taking each suspected target person in the suspected target person set as an index, and inquiring from the associated commodity order information to generate an inquiry result; and carrying out purchasing behavior prediction according to the query result to generate a prediction result.
In summary, in the purchasing behavior prediction apparatus according to the embodiment of the disclosure, first, the first obtaining module obtains the user purchasing data, and obtains the associated commodity order information from the user purchasing data according to the commodity embedded expression set, then, the second obtaining module obtains the user historical purchasing behavior data in the target time period, and the first analyzing and processing module analyzes and processes the user historical purchasing behavior data to generate the suspected target person set, and finally, the prediction module predicts the purchasing behavior according to the associated commodity order information and the suspected target person set to generate the prediction result. Therefore, the purchasing behavior of the enterprise user can be predicted, and the distribution mode of the commodities can be adjusted to reduce the distribution cost.
In order to implement the foregoing embodiments, as shown in fig. 13, the present invention further provides an electronic device 1300, which includes a memory 1310, a processor 1320, and a computer program stored in the memory 1310 and executable on the processor 1320, wherein the processor 1320 executes the computer program to implement the purchasing behavior prediction method according to the foregoing embodiments of the present disclosure.
The electronic device of the embodiment of the disclosure can predict the purchasing behavior of the enterprise user by executing the computer program stored in the memory through the processor, thereby adjusting the distribution mode of the commodity to reduce the distribution cost.
In order to implement the foregoing embodiments, the present invention also proposes a non-transitory computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the purchasing behavior prediction method proposed by the foregoing embodiments of the present disclosure.
The computer-readable storage medium of the embodiments of the present disclosure, by storing a computer program and being executed by a processor, can predict a purchasing behavior of an enterprise user, thereby adjusting a distribution manner of a commodity to reduce a distribution cost.
In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision, disclosure and other processing of the personal information of the related enterprises and users all accord with the regulations of related laws and regulations, and do not violate the good customs of the public order.
In the description of the present specification, the terms "first", "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present disclosure, "a plurality" means at least two, e.g., two, three, etc., unless explicitly specifically limited otherwise.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure, and that changes, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present disclosure.
Claims (18)
1. A method for predicting purchasing behavior, comprising:
acquiring user purchase data, and acquiring associated commodity order information from the user purchase data according to a commodity embedded expression set;
acquiring historical purchasing behavior data of the user in the target time period;
analyzing and processing the historical purchasing behavior data of the user to generate a suspected target personnel set;
and carrying out purchasing behavior prediction according to the associated commodity order information and the suspected target personnel set so as to generate a prediction result.
2. The purchasing behavior prediction method of claim 1, wherein the commodity embedded expression set is obtained by:
acquiring historical purchase data of enterprises in a target time period;
analyzing and processing the historical enterprise purchase data to generate an embedded expression corresponding to each commodity in the historical enterprise purchase data;
and generating the commodity embedded expression set according to the embedded expression corresponding to each commodity.
3. The method of claim 2, wherein the obtaining historical purchase data of the enterprise within the target time period comprises:
and acquiring historical purchase data of the enterprise in the target time period from a database of the first client.
4. The method for predicting purchasing behavior according to claim 2, wherein the analyzing the historical purchasing data of the enterprise to generate an embedded expression corresponding to each commodity in the historical purchasing data of the enterprise comprises:
acquiring each commodity in the historical enterprise purchase data and an enterprise user corresponding to each commodity;
drawing according to each commodity and the enterprise user corresponding to each commodity to generate an undirected graph;
and processing the undirected graph according to an embedded expression generation model to generate an embedded expression corresponding to each commodity.
5. The method for predicting purchasing behavior according to claim 1, wherein the obtaining user purchasing data and obtaining associated commodity order information from the user purchasing data according to the commodity embedded expression set includes:
acquiring unfinished order information from a database of a second client;
acquiring commodity order information corresponding to the same receiving address and the same distribution time period from the order information, and taking the commodity order information as the user purchase data;
according to the commodity embedded expression set, acquiring a commodity set to be associated from the user purchase data, and acquiring an embedded expression corresponding to each commodity to be associated in the commodity set to be associated;
and associating the commodities to be associated in the commodity set to be associated according to the association strategy and the embedded expression corresponding to each commodity to be associated so as to generate the associated commodity order information.
6. The method according to claim 5, wherein the obtaining historical purchasing behavior data of the user in the target time period comprises:
and acquiring the historical purchasing behavior data of the user in the target time period from the database of the second client.
7. The method of claim 1, wherein the analyzing the historical purchasing behavior data of the user to generate the suspected target person set comprises:
obtaining an analysis model;
and analyzing and processing the historical purchasing behavior data of the user according to the analysis model to generate the suspected target personnel set.
8. The method of claim 1, wherein the predicting the purchasing behavior according to the associated merchandise order information and the suspected target person set to generate a prediction result comprises:
respectively taking each suspected target person in the suspected target person set as an index, and inquiring from the associated commodity order information to generate an inquiry result;
and carrying out purchasing behavior prediction according to the query result to generate the prediction result.
9. A purchasing behavior prediction apparatus, comprising:
the system comprises a first acquisition module, a second acquisition module and a display module, wherein the first acquisition module is used for acquiring user purchase data and acquiring associated commodity order information from the user purchase data according to a commodity embedded expression set;
the second acquisition module is used for acquiring historical purchasing behavior data of the user in the target time period;
the first analysis processing module is used for analyzing and processing the historical purchasing behavior data of the user to generate a suspected target personnel set;
and the prediction module is used for predicting the purchasing behavior according to the associated commodity order information and the suspected target personnel set so as to generate a prediction result.
10. The purchasing behavior prediction apparatus of claim 9, further comprising:
the third acquisition module is used for acquiring historical purchase data of the enterprise in a target time period;
the second analysis processing module is used for analyzing and processing the enterprise historical purchase data to generate an embedded expression corresponding to each commodity in the enterprise historical purchase data;
and the generating module is used for generating the commodity embedded expression set according to the embedded expression corresponding to each commodity.
11. The purchasing behavior prediction apparatus of claim 10, wherein the third obtaining module is specifically configured to:
and acquiring historical purchase data of the enterprise in the target time period from a database of the first client.
12. The purchasing behavior prediction device of claim 10, wherein the second analysis processing module is specifically configured to:
acquiring each commodity in the historical enterprise purchase data and an enterprise user corresponding to each commodity;
drawing according to each commodity and the enterprise user corresponding to each commodity to generate an undirected graph;
and processing the undirected graph according to an embedded expression generation model to generate an embedded expression corresponding to each commodity.
13. The purchasing behavior prediction apparatus of claim 9, wherein the first obtaining module is specifically configured to:
acquiring unfinished order information from a database of a second client;
acquiring commodity order information corresponding to the same receiving address and the same distribution time period from the order information, and taking the commodity order information as the user purchase data;
acquiring a to-be-associated commodity set from the user purchase data according to the commodity embedded expression set, and acquiring an embedded expression corresponding to each to-be-associated commodity in the to-be-associated commodity set;
and associating the commodities to be associated in the commodity set to be associated according to the association strategy and the embedded expression corresponding to each commodity to be associated so as to generate the associated commodity order information.
14. The purchasing behavior prediction apparatus of claim 13, wherein the second obtaining module is specifically configured to:
and acquiring the historical purchasing behavior data of the user in the target time period from the database of the second client.
15. The purchasing behavior prediction device of claim 9, wherein the second analysis processing module is specifically configured to:
obtaining an analysis model;
and analyzing and processing the historical purchasing behavior data of the user according to the analysis model to generate the suspected target personnel set.
16. The purchasing behavior prediction device of claim 9, wherein the prediction module is specifically configured to:
respectively taking each suspected target person in the suspected target person set as an index, and inquiring from the associated commodity order information to generate an inquiry result;
and carrying out purchasing behavior prediction according to the query result to generate the prediction result.
17. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the method of purchasing behavior prediction as claimed in any one of claims 1-8 when executing the program.
18. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the purchasing behavior prediction method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210583291.6A CN115204450A (en) | 2022-05-26 | 2022-05-26 | Purchasing behavior prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210583291.6A CN115204450A (en) | 2022-05-26 | 2022-05-26 | Purchasing behavior prediction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115204450A true CN115204450A (en) | 2022-10-18 |
Family
ID=83577175
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210583291.6A Pending CN115204450A (en) | 2022-05-26 | 2022-05-26 | Purchasing behavior prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115204450A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115471302A (en) * | 2022-11-14 | 2022-12-13 | 山东智豆数字科技有限公司 | Electronic marketing data processing method based on big data analysis |
-
2022
- 2022-05-26 CN CN202210583291.6A patent/CN115204450A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115471302A (en) * | 2022-11-14 | 2022-12-13 | 山东智豆数字科技有限公司 | Electronic marketing data processing method based on big data analysis |
| CN115471302B (en) * | 2022-11-14 | 2023-02-14 | 山东智豆数字科技有限公司 | Electronic marketing data processing method based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9292545B2 (en) | Entity fingerprints | |
| CN112115363A (en) | Recommendation method, computing device and storage medium | |
| US7664671B2 (en) | Methods and systems for profile-based forecasting with dynamic profile selection | |
| US9269049B2 (en) | Methods, apparatus, and systems for using a reduced attribute vector of panel data to determine an attribute of a user | |
| US9275125B1 (en) | System for organizing data from a plurality of users to create individual user profiles | |
| US20210090105A1 (en) | Technology opportunity mapping | |
| CN110852785B (en) | User grading method, device and computer readable storage medium | |
| US8768746B2 (en) | Target marketing | |
| WO2012115965A1 (en) | Entity fingerprints | |
| CN113327151A (en) | Commodity object recommendation method and device, computer equipment and storage medium | |
| CN109509018A (en) | A kind of benefit data providing method, device and equipment | |
| CN115204971B (en) | Product recommendation method, device, electronic equipment and computer readable storage medium | |
| CN111429161A (en) | Feature extraction method, feature extraction device, storage medium, and electronic apparatus | |
| CN113689233A (en) | Advertisement putting and selecting method and corresponding device, equipment and medium thereof | |
| WO2023225529A2 (en) | Predictive systems and processes for product attribute research and development | |
| CN111861679A (en) | Commodity recommendation method based on artificial intelligence | |
| CN115204450A (en) | Purchasing behavior prediction method and device, electronic equipment and storage medium | |
| CN114862480A (en) | Advertisement putting orientation method and its device, equipment, medium and product | |
| CN114580402A (en) | Enterprise-oriented product information acquisition method and device, server and storage medium | |
| US10586163B1 (en) | Geographic locale mapping system for outcome prediction | |
| CN116485503A (en) | Commodity combination recommendation method, device, equipment and medium thereof | |
| Valliyammai et al. | An intelligent personalized recommendation for travel group planning based on reviews | |
| Wu et al. | [Retracted] Using the Mathematical Model on Precision Marketing with Online Transaction Data Computing | |
| Hung | Using Cloud Services to Develop Marketing Information System Applications | |
| CN111177657B (en) | Demand determining method, system, electronic device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |