TWI782242B - Intelligent marketing method and system thereof - Google Patents

Abstract

An intelligent marketing system includes a communication module, a storage module, and a processing module, the storage module stores a marketing website, and a plurality of transactions related to multiple customers connected to the marketing website by using their terminals The operation record of a series of operation actions, wherein, the different operation actions connected to the marketing website by any client are defined as a first state, a second state, a third state, and a fourth state, each operation The record is an action sequence composed of the first state, the second state, the third state, and the fourth state. The processing module calculates a target conversion probability and a target state according to the operation records. When When the current operation performed by a client is in the target state, the processing module sends a promotional material to the client.

Description

Intelligent Marketing Method and System

The present invention relates to a kind of marketing of consumer profile analysis, in particular to a marketing method using an intelligent system to recommend preferential products for customers.

At present, the rapid development of science and technology, especially the rapid development of the Internet has led to the development of various marketing, and the traditional method of product promotion through written and printed materials has gradually been replaced by digital marketing methods through the Internet. The current existing marketing method is mainly implemented by a marketing website. The marketing website includes a portal page, a product page, and a checkout page. The portal page is used to guide the user terminal linked to the portal page to the product page. The product webpage provides product-related information of a marketing product to be marketed, and the checkout webpage is used for the user to submit a transaction request related to the purchase of the marketing product. The existing marketing method includes the following steps: a customer through a marketing The marketing website URL provided by the personnel is connected to the marketing website through its client; when the customer uses its client to link to the product webpage of the marketing website, the product webpage displays the relevant information of the product on the client; and When the user terminal is connected to the checkout web page, the customer confirms and sends a transaction request related to purchasing the marketing product through the user terminal.

However, the existing marketing methods can only simply provide customers with product-related information about the marketing product, and cannot make further recommendations based on the customer's behavior. The marketing website will not perform other actions on the current operation behavior of the customer connecting to the marketing website through the client, so customers may easily give up buying because they do not fully understand the marketing product, which will undoubtedly consume the marketing cost of the marketing website.

Therefore, the object of the present invention is to provide an intelligent marketing method for marketing according to user operation behavior.

Another object of the present invention is to provide an intelligent marketing system for marketing according to user's operation behavior.

Therefore, the intelligent marketing method of the present invention is implemented by a server that connects multiple clients via a communication network and can provide a marketing website. The server records and stores multiple transactions related to multiple clients in a previous time zone. A series of operation records of a series of operations performed when using its client to connect to a marketing website. The marketing website includes a portal page, an activity page, a profile page, and a transaction page. The operation action of connecting to the portal webpage from the terminal is defined as a first state, and the operation action of any client using its client to connect to the active webpage is defined as a second state, any client using its client to connect to the The action of staying on the active webpage and staying on the active webpage for more than a stay threshold, or the action of linking to the profile webpage is defined as a third state, and any customer uses his client to link to the transaction webpage to send The operation action of a transaction request to the server is defined as a fourth state, and each operation record is composed of at least one of the first state, the second state, the third state, and the fourth state behavior sequence, and the intelligent marketing method of the present invention includes a step (A), a step (B), a step (C), a step (D), and a step (E).

The step (A) is that the server calculates the first state, the second state, the third state, and each of the fourth state transitions to at least one transition probability of at least one of the first state, the second state, the third state, and the fourth state.

The step (B) is that the server selects at least one candidate conversion probability from all the conversion probabilities calculated in the step (A), each candidate conversion probability is from the first state, the second state, and A transition probability for one of the third states to transition to the fourth state.

The step (C) is that the server selects a target transition probability with the highest value from the at least one candidate transition probability, and uses the previous state corresponding to the target transition probability to transition to the fourth state as a target state .

In the step (D), when another client connects to the marketing website, the server continuously judges whether the current operation performed by the other client on the marketing website is in the target state.

In the step (E), when the server determines that the current operation performed by the other user on the marketing website is the target state, the server sends a promotional material to the other user.

Furthermore, the intelligent marketing system of the present invention is connected to multiple user terminals via a communication network, and the intelligent marketing system includes a communication module, a storage module, and a processing module.

The communication module is connected to the communication network, and the storage module stores a marketing website and a series of operations related to multiple customers in a previous time period when connecting to the marketing website using their client Operation records, wherein, the marketing website includes a portal page, an activity page, a profile page, and a transaction page, and any operation action of a customer using its client to link to the portal page is defined as a first state, The operation action that any customer uses its client to link to the activity webpage is defined as a second state, and any customer uses its client to link to the activity webpage and stay on the activity webpage for a time exceeding a stay threshold Action, or the operation action of linking to the profile webpage is defined as a third state, any customer uses its user end to link to the transaction webpage to send a transaction request to the operation action of the processing module is defined as a fourth state state, and each operation record is an action sequence composed of at least one of the first state, the second state, the third state, and the fourth state.

The processing module is electrically connected to the communication module and the storage module, and calculates from the first state, At least one transition from each of the second state, the third state, and the fourth state to at least one of the first state, the second state, the third state, and the fourth state probability, and from all the calculated transition probabilities, select at least one candidate transition probability, each candidate transition probability is to transition from one of the first state, the second state, and the third state to the first state Four state conversion probabilities, the processing module then selects a target conversion probability with the highest value from the at least one candidate conversion probability, and uses the previous state corresponding to the target conversion probability to convert to the fourth state as a Target state, when another client connects to the marketing website, the processing module continues to judge whether the current operation performed by the other client on the marketing website is the target state, and when the other client is in the When the current operation performed by the marketing website is in the target state, the processing module sends a promotional material to the other client.

The effect of the present invention is: when the other client is connected to the marketing website, the server or the processing module judges whether the current operation performed by the other client on the marketing website is in the target state, and When the server or the processing module determines that the current operation performed by the other user on the marketing website is the target state, the server sends a promotional material to the other user, thereby, the The server side can more accurately strengthen the sales promotion when the customer has a high probability of using the user terminal to operate to enter the fourth state, so that the customer can learn more about the marketing product through the promotion materials, thereby improving its desire to buy.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same numerals.

Referring to FIG. 1 , an embodiment of the intelligent marketing system of the present invention is implemented by a server 1 , and the server 1 is connected to a plurality of user terminals 2 and a row of marketing terminals 3 through a communication network 100 .

The server 1 includes a server communication module 11 connected to the communication network 100, a server storage module 12, and a server processing module electrically connected to the server communication module 11 and the server storage module 12. Module 13. The server storage module 12 stores a marketing website 121, a plurality of transaction numbers corresponding to different time periods, and a plurality of transactions related to a plurality of customers in a previous time period, which are connected to the marketing website 121 by using the terminal 2 An operation record of a series of operation actions performed at that time, wherein the marketing website 121 includes a portal page, an event page, a profile page, and a transaction page. The operation action of any customer using its terminal 2 to link to the portal webpage is defined as a first state, and the operation action of any customer using its terminal 2 to link to the active webpage is defined as a second state. The operation action that the customer uses his terminal 2 to link to the activity webpage and stay on the activity page exceeds a stay threshold, or the operation action of linking to the introduction webpage is defined as a third state, and any customer uses The user terminal 2 is connected to the transaction webpage to send a transaction request to the server. The operation action is defined as a fourth state, and each operation record is composed of the first state, the second state, the third state, and A sequence of behaviors composed of at least one of the fourth states.

Each user terminal 2 is held by a corresponding client, and includes a user terminal communication module 21 connected to the communication network 100, a user terminal input module 22, and an electrical connection between the user terminal communication module 21 and The user input module 22 is the user processing module 23 .

The marketing terminal 3 is held by a salesperson, and includes a marketing terminal communication module 31 connected to the communication network 100, a marketing terminal input module 32, and an electrical connection between the marketing terminal communication module 31 and the marketing terminal. Marketing terminal processing module 33 of terminal input module 32.

The actions among the components of the server terminal 1 , the user terminals 2 and the marketing terminal 3 will be described below in conjunction with the intelligent marketing method of the present invention. Wherein, the intelligent marketing method of the present invention includes a marketing program and an updating program.

Referring to Fig. 1 and Fig. 2, the marketing program is to promote the operation action performed by connecting the user terminal 2 to the marketing website 121, and record the performance data of the marketing terminal 3, and includes a step 41, a step 42, A step 43 , a step 44 , a step 45 , a step 46 , a step 47 , a step 48 , a step 49 , a step 50 , a step 51 , and a step 52 .

In the step 41, the server-end processing module 13 calculates, according to the operation records corresponding to the previous time zone, the first state, the second state . At least one transition probability from each of the third state and the fourth state to at least one of the first state, the second state, the third state, and the fourth state.

For example, when the operation record of the first client among these clients connected to the marketing website 121 through its client 2 is the first state, the third state, and the fourth state in sequence, these clients The operation records of one of the second customers connecting to the marketing website 121 through its terminal 2 are the first state, the second state, the second state, and the fourth state in sequence. One of these clients The operation records of the third customer connecting to the marketing website 121 through its user terminal 2 are the first state, the second state, the third state, and the fourth state, one of these customers is the fourth customer The operation records connected to the marketing website 121 through its user terminal 2 are the first state, the third state, the third state, and the fourth state in sequence. In this embodiment, the server processing module 13 can use the Markov decision process (Markov decision process, MDP) of Markov chain (Markov chain) to perform calculations. The following is a brief description of part of the calculation process of the Markov decision process. First, the server processing module 13 according to the first client, The operation records of the second customer, the third customer, and the fourth customer calculate that the transition from the first state to the second state occurs twice, and the transition from the first state to the third state occurs twice , the transition from the second state to the second state occurs once, the transition from the second state to the third state occurs once, the transition from the second state to the fourth state occurs once, and the transition from the third The state transition to the third state occurs once, and the transition from the third state to the fourth state occurs three times, so the transition probability from the first state to the second state is 2/(2+2+1+ 1+1+1+3)≒0.182, similarly, the transition probability from the first state to the third state is 0.182, and the transition probability from the second state to the second state is 0.091, from the The transition probability from the second state to the third state is 0.091, the transition probability from the second state to the fourth state is 0.091, and the transition probability from the third state to the third state is 0.091, and The transition probability from the third state to the fourth state is 0.273.

In the step 42, the server processing module 13 selects at least one candidate conversion probability from all the conversion probabilities calculated in the step 41, each candidate conversion probability is from the first state, the second state, and the transition probability of one of the third states to the fourth state, for example, according to the operation records of the first client, the second client, the third client, and the fourth client, the The processing module 13 selects a transition probability from the second state to the fourth state and a transition probability from the third state to the fourth state as two candidate transition probabilities.

In the step 43, the server-side processing module 13 selects a target transition probability with the highest value from the at least one candidate transition probability, and transitions corresponding to the target transition probability to the previous state of the fourth state As a target state, for example, the server processing module 13 selects two candidate conversion probabilities according to the operation records of the first client, the second client, the third client, and the fourth client, wherein, The transition probability from the second state to the fourth state is 0.091, and the transition probability from the third state to the fourth state is 0.273, so the server processing module 13 selects the transition from the third state to the fourth state. The transition probability of the fourth state is the target transition probability, and the third state is the target state.

It is worth noting that in this embodiment, the server processing module 13 calculates the target state by applying the Markov decision process of the Markov chain, but the server processing module 13 can also use other algorithms, such as The web page ranking algorithm calculates the target status, which is not limited to this embodiment.

In step 44, the marketing terminal processing module 33 generates correspondingly according to the input signal from the marketing terminal input module 32 and transmits a URL request for the marketing website 121 to the server via the marketing terminal communication module 31. terminal 1, wherein the website request includes an identification information related to the marketing terminal 3 and a product identification information related to a marketing product. For example, the marketing terminal processing module 33 generates a URL request corresponding to the marketing website and sends it to the server 1 through the marketing terminal communication module 31, wherein the URL of the marketing website 121 is FCB.com. tw, the identity identification data related to the marketing terminal 3 is A1234, and the product identification data related to the marketing product is C007.

In step 45, when the server processing module 13 receives the URL request from the marketing terminal 3 for the marketing website 121 through the server communication module 11, the server processing module 13 The URL of the website 121, the identification information of the marketing terminal 3, and the product identification information of the marketing product, generate a pair of identification information corresponding to the marketing website 121 and corresponding to the marketing terminal 3 and product identification information of the marketing product and send it to the marketing terminal 3 through the communication module 11 of the server. For example, when the server-side processing module 13 receives the website request through the server-side communication module 11, the server-side processing module 13 according to the website FCB.com.tw of the marketing website 121, the marketing terminal 3’s identification information A1234, and the product identification information C007 of the marketing product, generate a marketing website FCB.com corresponding to the marketing website 121 and corresponding to the marketing terminal 3’s identification information and the product identification information of the marketing product .tw/CPID=C007&EID=A1234, and send the marketing URL corresponding to the marketing website 121 and corresponding to the identity identification data of the marketing terminal 3 and the product identification data of the marketing product through the server communication module 11 to In the marketing terminal 3, the server processing module 13 can transmit a two-dimensional barcode or a short URL related to the marketing website through the server communication module 11, but not limited thereto.

In the step 46 , the marketing terminal processing module 33 transmits the marketing URL to a target user terminal 2 among the user terminals 2 through the marketing terminal communication module 31 via the communication network 100 .

In step 47, when the target client 2 receives the marketing website, the client processing module 23 of the target client 2 passes the marketing website provided by the marketing station 3 in a current time period and passes The client communication module 21 of the target client 2 is connected to the marketing website 121 .

In the step 48, when the user-end processing module 23 of the target user-end 2 passes through the marketing website provided by the marketing end 3 and passes through the user-side communication module 21 of the target user-end 2 in the current time period When connecting to the marketing website 121, the server processing module 13 records the identity identification data of the marketing terminal 3 and the product identification data of the marketing product.

In the step 49, the server processing module 13 continues to determine whether the current operation of the target client 2 on the marketing website 121 is in the target state. When the server processing module 13 judges that the current operation action is the target state, the marketing program proceeds to step 50. When none of the current operations is the target state, the marketing program proceeds to step 53 .

In the step 50, the server processing module 13 transmits a promotional material to the target user end 2 through the server communication module 11, that is, when the server processing module 13 determines that the target user end 2 Link to the event webpage and stay on the event webpage for more than the stay threshold, or link to the introduction webpage, the server-side processing module 13 sends the promotion through the server-side communication module 11 through the communication network 100 The data is sent to the target user terminal 2. The promotional data is, for example, a promotional advertisement displayed in a pop-up window, but not limited thereto.

In the step 51, the user-end processing module 23 of the target user-end 2 transmits a corresponding marketing product through the user-end communication module 21 of the target user-end 2 through the communication network 100 in the current time period The transaction request to the server 1.

In step 52, when the server processing module 13 receives the transaction request from the target client 2 through the server communication module 11 in the current time period, the server processing module 13 will Add one to the value of the transaction number corresponding to the current time zone, and at the same time, the server-side processing module 13 calculates the value corresponding to the marketing product and related to the product according to the identification data and product identification data recorded in step 48 A performance data of marketing terminal 3.

In the step 53, the server processing module 13 stores the operation record of a series of operations performed by the client 2 connected to the marketing website 121 in the current time zone on the marketing website 121 in the server The storage module 12 stores it as an operation record corresponding to the client 2 in the current time period.

Referring to FIG. 1 and FIG. 3 , the update program is used to compare whether the target state needs to be replaced, and includes a step 61 , a step 62 , a step 63 , and a step 64 .

In the step 61, the server processing module 13 judges whether the transaction volume in the current time zone is smaller than the transaction volume in the previous time zone (that is, a time zone before the current time zone), When the number of deals in the current time zone is not less than the number of deals in the previous time zone, the update procedure ends; when the number of deals in the current time zone is smaller than the number of deals in the previous time zone, The update program proceeds to step 62 .

In the step 62, the server-side processing module 13 calculates from the first state in a series of operations performed by the clients in the current time period based on all the operation records in the current time period. . At least one transition probability from each of the second state and the third state to at least one of the first state, the second state, the third state, and the fourth state.

For example, the operation record of the fifth customer among the customers in the current time period is the first state, the second state, and the third state in sequence, and the sixth customer among the customers The operation records of the client are the first state, the second state, the third state, and the fourth state in sequence, and the operation records of the seventh client among these clients are the first state, the second state state, and the fourth state, and the operation record of the eighth client among the clients is the first state, the third state, the third state, and the fourth state, then the server processes According to the operation records, the module 13 calculates that the transition probability from the first state to the second state in the current time period is 0.3, and the transition probability from the first state to the third state is 0.1 , the transition probability from the second state to the third state is 0.2, the transition probability from the second state to the fourth state is 0.1, and the transition probability from the third state to the third state is 0.1, and the transition probability from the third state to the fourth state is 0.2.

In the step 63, the server processing module 13 selects at least one specific conversion probability from all the conversion probabilities calculated in the step 62, each specific conversion probability is from the first state, the second state , and the transition probability of one of the third states but not the target state transitioning to the fourth state, for example, all transition probabilities in the current time period are transitioning from the first state to the transition probability from the second state, transition probability from the first state to the third state, transition probability from the second state to the third state, transition from the second state to the fourth state probability, a transition probability from the third state to the third state, and a transition probability from the third state to the fourth state, wherein the target state in the previous time period is the third state, Therefore, the server processing module 13 selects the transition probability from the second state to the fourth state as a specific transition probability.

In the step 64, the server processing module 13 selects a specific conversion rate with the highest value from the at least one specific conversion rate, and re-transitions corresponding to the specific conversion rate to the previous state of the fourth state. state as the target state.

It should be noted that, in this embodiment, the server processing module 13 calculates and updates the target state by applying the Markov decision process of the Markov chain, but the server processing module can also use other algorithms to re-update the target state. The target state is not limited to this embodiment.

To sum up, the intelligent marketing method and system of the present invention select the target state through the server processing module 13, and judge whether the current operation action performed by the target user terminal 2 on the marketing website 121 is the target state , when the current operation behavior of the target user end 2 is the target state, the server processing module 13 sends a promotional material to the target user end 2 through the server communication module 11, thereby, when the user is in the When it is more likely to purchase the marketing product, recommend information to encourage the purchase, and then influence the user to purchase the marketing product. On the other hand, the server-side processing module 13 uses the recorded identification information and product identification information , to calculate a performance data corresponding to the marketing product and related to the marketing terminal 3, whereby the relevant credits of the marketing staff through the marketing terminal 3 can also be recorded and not ignored. Furthermore, The server-side processing module 13 compares the number of transactions in different time periods to determine whether the target state needs to be updated, thereby making the target state not static but correspondingly modified with the operation records corresponding to different time periods , and then make the timing point of marketing can always be close to customer needs, so the purpose of the present invention can be really achieved.

But the above-mentioned ones are only embodiments of the present invention, and should not limit the scope of the present invention. All simple equivalent changes and modifications made according to the patent scope of the present invention and the content of the patent specification are still within the scope of the present invention. Within the scope covered by the patent of the present invention.

1: Server side 100: Communication network 11: Server communication module 12:Server storage module 121:Marketing website 13:Server-side processing module 2: Use end 21: Use end communication module 22: Use terminal input module 23: Use end processing module 3: Marketing side 31: Marketing terminal communication module 32: Marketing terminal input module 33: Marketing terminal processing module Step 41~52: Marketing Program Step 61~64: update program

Other features and effects of the present invention will be clearly presented in the implementation manner with reference to the drawings, wherein: Fig. 1 is a block diagram illustrating an embodiment of the intelligent marketing system of the present invention, which is connected to a user end and a marketing end through a communication network; Fig. 2 is a flowchart illustrating a marketing program of an intelligent marketing method implemented in this embodiment; and FIG. 3 is a flow chart illustrating an update procedure of an intelligent marketing method implemented in this embodiment.

1: Server side

100: Communication network

11: Server communication module

12:Server storage module

121:Marketing website

13:Server processing module

2: Use end

21: Use end communication module

22: Use terminal input module

23: Use end processing module

3: Marketing side

31: Marketing terminal communication module

32: Marketing terminal input module

33: Marketing terminal processing module

Claims (8)

An intelligent marketing method is implemented by a server that connects a plurality of users through a communication network and can provide a marketing website. An operation record of a series of operations performed when the user terminal is connected to the marketing website. The marketing website includes an entry page, an activity page, a profile page, and a transaction page, wherein any customer uses his user terminal to link to The operation action of the portal webpage is defined as a first state, and the operation action of any customer using its terminal to link to the active webpage is defined as a second state, any customer uses its terminal to link to the active webpage and The operation action of staying on the activity webpage for more than a stay threshold, or the operation action of linking to the profile webpage is defined as a third state, and any customer uses his client to link to the transaction webpage to send a transaction The operation action requested to the server is defined as a fourth state, and each operation record is a behavior sequence composed of at least one of the first state, the second state, the third state, and the fourth state , the intelligent marketing method includes the following steps: (A) the server calculates the first state, the second at least one transition probability from each of the two states, the third state, and the fourth state to at least one of the first state, the second state, the third state, and the fourth state; (B) The server selects at least one candidate conversion probability from all the conversion probabilities calculated in step (A), and each candidate conversion probability is from The transition probability of one of the first state, the second state, and the third state to the fourth state; (C) the server selects one with the highest value from the at least one candidate transition probability Target conversion probability, and the previous state corresponding to the target conversion probability converted to the fourth state is used as a target state; (L) when the server receives a URL request from a marketing terminal for the marketing website , wherein the URL request includes the marketing terminal's identification information and a product identification information related to a marketing product, the server end according to the marketing website's URL, the marketing terminal's identification information, and the product identification information of the marketing product , generate a marketing website corresponding to the marketing website and corresponding to the marketing terminal’s identification data and the product identification data of the marketing product, and send it to the marketing terminal; (D) when another user terminal connects to the marketing website , the server continues to judge whether the current operation performed by the other client on the marketing website is in the target state. website, the server also records the marketing terminal’s identification information and the product identification information of the marketing product; and (E) when the server determines that the current operation performed by the other user terminal on the marketing website is In the target state, the server sends a promotional material to the other client. The intelligent marketing method as described in Claim 1, wherein the server also stores a transaction number corresponding to the previous time period, and after the (E) step, the following steps are further included: (F) The server saves the operation record of a series of operations performed by the other user terminal when it connects to the marketing website in the current time zone as the operation record corresponding to the other user terminal in the current time zone; and (G) when the server receives the transaction request in a current time zone, the server adds one to the value of a transaction number corresponding to the current time zone. The intelligent marketing method as described in claim 2 further includes the following steps after step (G): (H) the server determines whether the number of transactions in the current time zone is smaller than the number of transactions in the previous time zone; (1) When the server judges that the number of transactions in the current time zone is less than the number of transactions in the previous time zone, the server calculates the current time period based on all operation records in the current time zone Transition from each of the first state, the second state, and the third state to the first state, the second state, and the third state in a series of operations performed by the customers in the segment state, and at least one conversion probability of at least one of the fourth state; (J) the server selects at least one specific conversion probability from all the conversion probabilities calculated in step (1), and each specific conversion Probability is the transition probability from one of the first state, the second state, and the third state but not the target state to the fourth state; and (K) the server transfers from the at least one For a specific transition probability, select a specific transition probability with the highest value, and re-select the previous state that transitions to the fourth state corresponding to the specific transition probability as the target state. The intelligent marketing method as described in claim item 1, wherein, after the step (E), the following steps are further included: (N) when the server receives a transaction request from the other client, the server executes the Based on the recorded identification information and product identification information, a performance information corresponding to the marketing product and related to the marketing terminal is calculated. An intelligent marketing system is used for intelligent marketing, and connects multiple terminals via a communication network, and includes: a communication module connected to the communication network; a storage module storing a marketing website, and multiple records related to An operation record of a series of operations performed by multiple customers in a previous time period when using their client to connect to the marketing website. The marketing website includes a portal page, an activity page, a profile page, and a transaction page, Among them, the operation action of any customer using its user terminal to link to the portal webpage is defined as a first state, and the operation action of any customer using its user terminal to link to the active webpage is defined as a second state. Any customer uses its client to link to the activity webpage and stay on the activity webpage for a time exceeding a stay threshold value, or the operation action of linking to the introduction webpage is defined as a third state. Any customer uses The operation action in which the client connects to the transaction webpage to send a transaction request to the processing module is defined as a fourth state, and each operation record consists of the first state, the second state, and the third state , and a behavior sequence composed of at least one of the fourth states; and a processing module electrically connected to the communication module and the storage module, where The management module calculates the first state, the second state, the third state, and the fourth state from the first state, the second state, the third state, and the fourth Each of the states transitions to at least one transition probability of at least one of the first state, the second state, the third state, and the fourth state, and from all the transition probabilities calculated, select at least a candidate transition probability, each candidate transition probability is a transition probability from one of the first state, the second state, and the third state to the fourth state, and the processing module transitions from the at least A candidate conversion probability, select a target conversion probability with the highest value, and use the previous state corresponding to the target conversion probability to switch to the fourth state as a target state, when another client connects to the marketing website , the processing module continues to judge whether the current operation action performed by the other user terminal on the marketing website is the target state, and when the current operation action performed by the other user terminal on the marketing website is the target state , the processing module sends a promotional data to the other client; wherein, when the processing module receives a URL request from a marketing terminal for the marketing website, the URL request includes the marketing terminal’s identification information and a product identification data related to a marketing product, the processing module generates a corresponding marketing website and corresponding to The identification information of the marketing terminal and the marketing website of the product identification information of the marketing product are sent to the marketing terminal, and when the other user terminal uses the marketing website provided by the marketing terminal to connect to the marketing website, The processing module also records the identity information and Product identification information of the marketed product. The intelligent marketing system as described in claim item 5, wherein, the processing module also stores a transaction number corresponding to the previous time zone, and when the processing module receives the transaction request in a current time zone, the The processing module will add one to the value of a transaction value corresponding to the current time zone, and the processing module will also record a series of operations performed by the other terminal when connecting to the marketing website in the current time zone The operation record is stored as an operation record corresponding to the other client in the current time zone. The intelligent marketing system as described in claim 6, wherein the processing module judges whether the number of transactions in the current time zone is smaller than the number of transactions in the previous time zone, when the processing module determines that the current time zone When the number of transactions in the time period is less than the number of transactions in the previous time period, the processing module calculates a series of transactions related to these customers in the current time period based on all the operation records in the current time period Transition from each of the first state, the second state, and the third state to at least one of the first state, the second state, the third state, and the fourth state during operation at least one transition probability, and select at least one specific transition probability from all the transition probabilities corresponding to the current time zone, each specific transition probability is from the first state, the second state, and the third one of the states but not the target state is converted to the transition probability of the fourth state, and the processing module selects a specific transition probability with the highest value from the at least one specific transition probability, and re-selects the specific transition probability The state preceding the transition to the fourth state corresponding to the transition probability is used as the target state. The intelligent marketing system as described in claim item 5, wherein, when the processing module is connected to When receiving a transaction request from the other user end, the processing module calculates a performance data corresponding to the marketing product and related to the marketing end based on the recorded identification data and product identification data.

Images