Hidden demand mining method and system for customized products based on data timeliness update

Abstract

The present invention discloses a method for mining hidden demand of customized products based on data timeliness update, and the steps of the method are as follows: obtaining and standardizing historical order information of customized products; constructing a life tree structure with customized product order data, representing project information with a tree path, retaining time information in the corresponding time leaf node, and storing the largest time node information in the tree at the time root node; retrieving all time net leaf nodes in the life tree and calculating the vitality value of the node through a decay window model, if the node vitality value is less than the set minimum vitality threshold, deleting the tree path corresponding to the net leaf node and adjusting the tree structure; reconstructing the life tree to generate association rules according to the frequency threshold; based on the association rule set, realizing the mining of hidden demand of customized product demand. The present invention reduces the generation of inaccurate rules by analyzing the timeliness of order information, and more reliably mines the customized product demand of customers.

Description

Technical Field

[0001] The present invention relates to a technology for mining hidden demands of customized products, and in particular to a method and system for mining hidden demands of customized products based on data timeliness update. Background Art

[0002] Hidden demand refers to the demand that customers do not express or express unclearly during the process of proposing product requirements. Mining hidden demand can make demand information more complete and help designers design products that satisfy customers. With the introduction of the concept of Industry 4.0, the personalization of customer needs has been further improved, and more and more needs are hidden under explicit fuzzy needs. The importance of mining hidden demand has increased again. The timeliness of historical order data as input for mining hidden demand is getting shorter and shorter. Newer orders should have a greater influence in the process of demand mining, and old and low-frequency orders should be deleted from the database in a timely manner to ensure the accuracy of mining hidden demand.

[0003] Existing implicit demand mining methods lack consideration of data timeliness and assume that all historical order data have equal influence. This leads to reduced accuracy of demand mining, indirectly affecting the product design process and reducing customer satisfaction. In order to alleviate this problem, some methods add new lists or tree structures to store time information, evaluate the timeliness of each data one by one, and then reconstruct the data structure, which leads to a waste of computing resources and time. To address this problem, this method proposes a life tree structure that can compress and store order information and time information. When evaluating the timeliness of orders using the decay window model, time information is retrieved through node links, invalid data information in the tree structure is quickly deleted, and the data structure is adjusted. According to the association rules extracted from the life tree structure, efficient and accurate conversion of implicit demand expression is achieved. Summary of the invention

[0004] The purpose of the present invention is to provide a method for mining hidden demand for customized products based on data timeliness update in view of the shortcomings of the prior art. The method uses a decay window model to evaluate the timeliness of orders, retrieves time information through node links, quickly deletes invalid data information in the tree structure and adjusts the data structure. According to the association rules extracted from the life tree structure, the efficient and accurate conversion of hidden demand expression is achieved.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] A first aspect of an embodiment of the present invention provides a method for mining hidden demand for customized products based on data timeliness update, which specifically includes the following steps:

[0007] (1) Obtain and standardize historical order information of customized products;

[0008] (2) constructing a tree of life structure based on customized product order data;

[0009] (3) Retrieve all time-net leaf nodes in the tree of life and calculate the node vitality value through the decay window model based on the frequency and time information of the node;

[0010] (4) Starting from the first time-clean leaf node, if the node vitality value is less than the set minimum threshold, delete the tree path corresponding to the clean leaf node and adjust the tree structure; if the node vitality value is greater than the set minimum threshold, jump to the next clean leaf node until all clean leaf nodes are traversed;

[0011] (5) Reconstruct the life tree based on the frequency threshold and generate association rules;

[0012] (6) Based on the association rule set obtained in step (5), complete the mining of implicit demand for customized product requirements.

[0013] Furthermore, the life tree structure in step (2) is used to store order data; the order data includes project information, time information and project frequency; the project information is represented by a tree path, the project frequency is stored in the tree node, and the time information is in the leaf node of the tree path; the largest time node information in the life tree is stored in the root node.

[0014] Further, the tree path of the life tree is represented by the link of the tree nodes; the tree nodes include a time root node, a time leaf node, a path node and a time net leaf node; the time root node is the starting point of the life tree structure, and is used to store the maximum time of the life tree structure; the time leaf node is the tail node of the life tree path, and is used to store the time information and project information of the order data; the path node is a node linking the root node and the leaf node, and is used for the project information of the order data; the time net leaf node is a leaf node that does not belong to the path node of other paths;

[0015] Nodes with the same project name are connected together by node links.

[0016] Furthermore, the nodes of the tree path are arranged in descending order of project frequency, and the higher the frequency of the project node is, the closer it is to the root node; whenever new data is inserted, the project frequency will be updated synchronously, and the tree structure will adjust the node position according to the new project frequency order; the data update steps of the tree structure are:

[0017] Re-count the item frequencies and sort them in descending order to get a list of items with changed frequencies L change ;

[0018] Reconstruct the tree structure according to the new item frequency and retrieve the item list L changeThe tree nodes of the related items in the tree are swapped from bottom to top until the latest item frequency order is met; if in this process, a time leaf node of a tree path becomes a path node, its time information is saved in the newly generated leaf node; if the frequency of the swapped parent node R is greater than the child node Q, the parent node splits out a brother node R′, the frequency count of the brother node R′ is the difference between the frequency counts of the parent node R and the child node Q, and then the parent node R and the child node Q are swapped;

[0019] Insert the items in the new order into the life tree one by one in descending order of item frequency, and save the time information of each order in the leaf node of the corresponding tree path; when the inserted order time is greater than the root node time, update the root node time information. After insertion, if there are two or more nodes with the same item name at each level in the tree structure, merge the nodes and add up their counts; if the new data time is greater than the root node time, update the root node time information; repeat the above steps until all the new order data are inserted into the life tree structure.

[0020] Furthermore, the step (3) specifically comprises:

[0021] Calculate each transaction t d The vitality value v(t d ), the calculation formula is as follows:

[0022] v(t d )=∑v(i,t d ) / |t d |

[0023] Among them, i is the item in the transaction, v(i,t d ) is the item i in transaction t d The relative importance measurement factor in |t d | is the number of items traded.

[0024] Based on the vitality value v(t d ), calculate the vitality value v(l d ), the calculation formula is as follows:

[0025] v(l d )=∑v(t d )×f T-d

[0026] Where d is for each transaction t d The time information of , T is the time information contained in the time root node, and f is the vitality decay factor.

[0027] Furthermore, the step of deleting the invalid path in step (4) is as follows:

[0028] For a single failed path, the time leaf node of the failed path is deleted first, and the count of the time leaf node is subtracted from the count of other path nodes on the path; if the count of the path node is 0 after the count subtraction, the node is deleted; when all failed paths are deleted, the life tree structure is updated and adjusted.

[0029] Furthermore, the step (5) is specifically as follows:

[0030] Through the life tree structure, rare items and frequent items are mined. The support of rare items Supp(X) satisfies maxRare>Supp(X)≥minRare, and the support of frequent items satisfies Supp(X)≥minFre.

[0031] Among them, minRare is the minimum threshold of rare items, maxRare is the maximum threshold of rare items, and minFre is the minimum threshold of frequent items;

[0032] After obtaining the two types of items, the conditional pattern tree and frequent / rare pattern sets are obtained by eliminating the tree paths that do not meet the rare or frequent thresholds in the life tree structure to generate association rules.

[0033] Furthermore, the user's demand information is classified and checked to see whether it meets the product design requirements to obtain demand classification information; the demand classification information is then converted into standard requirements; based on the standard requirements and the association rule set obtained in step (5), the mining of implicit requirements for customized product needs is completed.

[0034] A second aspect of an embodiment of the present invention provides a system for mining hidden demand for customized products based on data timeliness update, comprising:

[0035] Interactive interface, used to assist users in selecting products and expressing demand information;

[0036] Demand analysis module, used to classify demand information, check whether it meets product design, and output demand classification information;

[0037] Demand standardization module, used to convert demand classification information output by demand analysis module into standard demand;

[0038] The life tree construction module is used to compress and save the historical order information into a life tree structure to save the order data;

[0039] The tree structure optimization module is used to calculate the vitality value of the time leaf nodes of the tree structure and delete invalid paths in the tree structure;

[0040] The pattern extraction module is used to obtain the conditional pattern tree of popular items through the support threshold and the final tree structure, and integrate the statistical popular patterns;

[0041] The rule extraction module is used to calculate and filter reliable association rules through popular pattern sets;

[0042] The demand mining module uses the standard requirements output by the demand standardization module and the implicit requirement association rules output by the rule extraction module to mine implicit requirements and transform them into usable explicit requirements.

[0043] A third aspect of an embodiment of the present invention provides a device for mining hidden demands of customized products based on data timeliness update, comprising one or more processors for the above-mentioned method for mining hidden demands of customized products based on data timeliness update.

[0044] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, is used to implement the above-mentioned method for mining implicit demands of customized products based on data timeliness update.

[0045] The beneficial effects of the present invention are as follows: the present invention uses a data structure that can store time information and order information at the same time, uses a decaying window model to evaluate the timeliness of orders, retrieves time information through node links, quickly deletes invalid data information in the tree structure and adjusts the data structure, thereby improving the mining quality and efficiency of association rules and achieving efficient and accurate conversion of implicit demand expressions. The present invention analyzes the timeliness of order information, identifies and removes invalid orders in the data structure, reduces the generation of inaccurate rules, and more reliably mines customized product needs that are not expressed or not clearly expressed by customers. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flowchart of a method for mining hidden demands of customized products based on data timeliness update;

[0047] Figure 2 A schematic diagram of the tree structure of the method for mining hidden needs of customized products of the present invention;

[0048] Figure 3 This is an architecture diagram of a system for mining hidden demands for customized products based on data timeliness updates;

[0049] Figure 4 Schematic diagram of the device of the present invention. DETAILED DESCRIPTION

[0050] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0051] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0052] It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0053] The specific embodiments of the present invention will be described more completely and clearly below in conjunction with the accompanying drawings. Figure 1 This is a method for mining hidden demand for customized products based on data timeliness update, which includes the following steps:

[0054] (1) Obtain and standardize historical order information of customized products and convert user demand information into standardized demand; specifically:

[0055] Export the historical order information of customized products from the user database, and standardize the historical order information and order time. Classify the user's demand information, and check whether it meets the product design requirements to obtain demand classification information; then convert the demand classification information into standard requirements. The standardization specifically includes that the demand parameters and design parameters in the order can be divided into continuous parameters and discrete parameters, and the continuous parameters are divided into discrete parameters according to a certain numerical interval, and all discrete parameters are numerically encoded. The order time is represented in the form of a timestamp to improve data processing efficiency.

[0056] (2) The order data is converted into a life tree structure, and the order information is represented in the form of a tree path; specifically:

[0057] The life tree structure integrates the life cycle characteristics of real trees and is used to store order information and time information data. The order data includes project information, time information and project frequency; the project information is represented by a tree path, the project frequency is stored in the tree node, and the time information is in the leaf node of the tree path; the largest time node information in the life tree is stored in the root node, and when the inserted order time is greater than the root node time, the root node time information is updated.

[0058] The life tree structure has the following characteristics: the tree path is represented by the links of the tree nodes, and the tree nodes can be divided into time root nodes, path nodes, and time leaf nodes. The time root node is the starting point of the tree structure and saves the maximum time inside the tree structure; the time leaf node is the tail node of the tree path and saves the time information and project information of the order; the path node is the node that links the root node and the leaf node and only saves the project information of the order. Nodes with the same project name are connected together through node links. The nodes of the tree path are arranged in descending order of the frequency of the project, and the higher the frequency of the project node, the closer it is to the root node. Whenever new data is inserted, the project frequency will be updated synchronously, and the tree structure will adjust the node position according to the new project frequency order.

[0059] The data update of the tree structure is specifically as follows: first, re-count the frequency information of the items and sort them in descending order. Secondly, by comparing the frequency order of the items before and after, a list of items with changed frequencies is obtained. change . Reconstruct the tree structure according to the new item frequency and retrieve L change The tree nodes related to the items in the project are swapped from bottom to top until the latest item frequency order is met.

[0060] If in this process, a time leaf node of a tree path becomes a path node, its time information is saved in the newly generated leaf node; if the frequency of the swapped parent node R is greater than the child node Q, the parent node splits out a brother node R′, and the frequency count of R′ is the difference between the frequency counts of the parent node R and the child node Q, that is, R′.count=R.count-Q.count, and then the node R and the node Q are swapped.

[0061] Insert the items in the new order into the life tree one by one in descending order of item frequency, and save the time information of each order in the leaf node of the corresponding tree path. When the time of the inserted order is greater than the root node time, update the root node time information. After insertion, if there are two or more nodes with the same item name at each level in the tree structure, merge the nodes and add up their counts. If the new data time is greater than the root node time, update the root node time information. Repeat the above steps until all the new order data are inserted into the life tree structure.

[0062] (3) Calculate the node vitality value through the decay window model; specifically:

[0063] In the decay window model, the time leaf nodes are traversed through the node links to obtain each transaction t in the node. d Time information and project information, calculate each transaction t d The vitality value v(t d ), the calculation formula is as follows:

[0064] v(td )=∑v(i,t d ) / |t d |

[0065] Among them, i is the item in the transaction, v(i,t d ) is the item i in transaction t d The relative importance measurement factor in |t d | is the number of items traded. According to the vitality value v(t d ), calculate the vitality value v(l d ), the calculation formula is as follows:

[0066] v(l d )=∑v(t d )×f T-d

[0067] Where d is for each transaction t d The time information of , T is the time information contained in the time root node, and f is the vitality decay factor.

[0068] (4) Determine and delete invalid tree paths and adjust the tree structure; specifically:

[0069] Starting from the first time net leaf node, when the vitality of the time net leaf node is lower than the set minimum threshold, its associated tree path is defined as a failed path. For a single failed path, the time leaf node of the failed path is first deleted, and the count of the other path nodes on the path is subtracted from the count of the time leaf node. If the count of the path node is 0 after the count is subtracted, the node is deleted. When all failed paths are deleted, the tree structure is updated and adjusted according to the life tree structure characteristics (3).

[0070] If the node vitality value is greater than the set minimum threshold, it jumps to the next time leaf node until all net leaf nodes are traversed.

[0071] (5) Reconstruct the life tree according to the frequency threshold, and generate association rules through the reconstructed life tree; specifically:

[0072] First, rare items and frequent items are mined through the life tree structure. The support of rare items Supp(X) satisfies the formula maxRare>Supp(X)≥minRare, and the support of frequent items satisfies Supp(X)≥minFre, where minRare is the minimum threshold of rare items, maxRare is the maximum threshold of rare items, and minFre is the minimum threshold of frequent items. After obtaining the two types of items, the conditional pattern tree and frequent / rare pattern sets are obtained by eliminating the tree paths that do not meet the rare or frequent thresholds in the life tree structure, and finally generating association rules.

[0073] (6) Based on the association rule set obtained in step (5) and the standard requirements obtained in step (1), the hidden requirements of customized product requirements are mined through the mapping of association rules.

[0074] Figure 2 The schematic diagram of the life tree structure of the method for mining hidden demand of customized products of the present invention, the left side of the figure is the frequency count table of the project, the right side is the life tree structure, the black node in the structure is the time root node, the white node is the path node, the gray node is the time leaf node, the order is stored in the structure in the form of a tree path, and the time in the bracket is the time of the order. The time root node saves the maximum transaction time in the life tree, and the time leaf node saves the time of the order of a single or multiple same tree paths. The items in a tree path are sorted in descending order of project frequency, and the nodes are linked by node links. The same path parts of different tree paths will be merged, and the corresponding node counts are added. The frequency count table is linked to the project nodes corresponding to the life tree structure, which is convenient for node traversal to improve calculation efficiency. When an order needs to be inserted, the project frequency count table is first updated, the order items are arranged in descending order according to the latest project frequency and the life tree structure is reconstructed, and finally the project nodes are inserted into the tree structure in sequence, and the time information of the order is stored in the last leaf node.

[0075] Figure 3 It is a customized product demand conversion system based on association rule-oriented optimization, including:

[0076] (A) Interactive interface: Users can select products and express their needs through the interactive interface.

[0077] (B) Demand analysis module: classify the acquired demand information and check whether it meets the product design, and output demand classification information. (C) Demand standardization module: convert the demand classification information (various personalized demands) output by the demand analysis module into standard demands.

[0078] (D) Life tree construction module: compresses and saves historical order information into a life tree structure, and saves the transaction information and time information of the order.

[0079] (E) Tree structure optimization module: Calculate the vitality value of the time leaf nodes of the tree structure, delete invalid paths in the tree structure, and reduce the interference of invalid transactions on association rule mining.

[0080] (F) Pattern extraction module: Through the support threshold and the final tree structure, the conditional pattern tree of popular items is obtained and the statistical popular patterns are integrated.

[0081] (G) Rule extraction module: By obtaining the popular pattern set, reliable association rules are calculated and screened.

[0082] (H) Requirements mining module: It uses the standard requirements output by the requirements standardization module and the implicit requirements association rules output by the rule extraction module to mine implicit requirements and transform them into usable explicit requirements.

[0083] Corresponding to the aforementioned embodiment of the method for mining hidden demands of customized products based on data timeliness update, the present invention also provides an embodiment of the device for mining hidden demands of customized products based on data timeliness update.

[0084] See also Figure 4 An embodiment of the present invention provides a device for mining hidden demands of customized products based on data timeliness update, including one or more processors for implementing the method for mining hidden demands of customized products based on data timeliness update in the above embodiment.

[0085] The embodiment of the device for mining hidden demands of customized products based on data timeliness update of the present invention can be applied to any device with data processing capability, and the device with data processing capability can be a device or apparatus such as a computer. The device embodiment can be implemented through software, or through hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of any device with data processing capability in which it is located reading the corresponding computer program instructions in the non-volatile memory into the internal memory for execution. From the hardware level, if Figure 4 As shown, it is a hardware structure diagram of any device with data processing capability where the device for mining hidden demand of customized products based on data timeliness update of the present invention is located, except Figure 4 In addition to the processor, memory, network interface, and non-volatile memory shown, any device with data processing capabilities in which the apparatus in the embodiments is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0086] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0087] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of the present invention. Ordinary technicians in this field can understand and implement it without paying creative work.

[0088] An embodiment of the present invention further provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method for mining hidden demands of customized products based on data timeliness update in the above embodiment is implemented.

[0089] The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be any device with data processing capability, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of any device with data processing capability and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.

[0090] The above embodiments are only used to illustrate the design ideas and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, any equivalent changes or modifications made based on the principles and design ideas disclosed by the present invention are within the protection scope of the present invention.

Claims

1. A method for mining hidden demand for customized products based on data timeliness update, characterized in that: The specific steps include: (1) Obtain and standardize historical order information of customized products; (2) Constructing a tree of life structure based on customized product order data; The life tree structure in step (2) is used to store order data; the order data includes item information, time information and item frequency; the item information is represented by a tree path, the item frequency is stored in the tree node, and the time information is in the leaf node of the tree path; the largest time node information in the life tree is stored in the root node; The tree path of the life tree is represented by the links of the tree nodes; the tree nodes include a time root node, a time leaf node, a path node and a time net leaf node; the time root node is the starting point of the life tree structure, and is used to store the maximum time of the life tree structure; the time leaf node is the tail node of the life tree path, and is used to store the time information and project information of the order data; the path node is a node linking the root node and the leaf node, and is used for the project information of the order data; the time net leaf node is a leaf node that does not belong to the path node of other paths; Nodes with the same project name are connected together via node links; The nodes of the tree path are arranged in descending order of project frequency. The higher the frequency of the project node, the closer it is to the root node. Whenever new data is inserted, the project frequency will be updated synchronously, and the tree structure will adjust the node position according to the new project frequency order. The data update steps of the tree structure are: Re-count the item frequencies and sort them in descending order to get a list of items with changed frequencies ; Reconstruct the tree structure according to the new item frequency and retrieve the item list The tree nodes of the related items in the tree are swapped from bottom to top until the latest item frequency order is met; if in this process, a time leaf node of a tree path becomes a path node, its time information is saved in the newly generated leaf node; if the swapped parent node Frequency greater than child node , then the parent node splits out a sibling node , the sibling node Frequency count for parent node With child nodes The difference in frequency counts of the parent node and child nodes swap; Insert the items in the new order into the life tree one by one in descending order of item frequency, and save the time information of each order in the leaf node of the corresponding tree path; When the inserted order time is greater than the root node time, the root node time information is updated. After the insertion, if there are two or more nodes with the same item name at each level in the tree structure, the nodes are merged and their counts are added; If the new data time is greater than the root node time, update the root node time information; Repeat the above steps until all the newly added order data are inserted into the life tree structure; (3) Retrieve all time-net leaf nodes in the tree of life and calculate the node vitality value through the decay window model based on the frequency and time information of the node; (4) Starting from the first time-clean leaf node, if the node vitality value is less than the set minimum threshold, the tree path corresponding to the clean leaf node is deleted and the tree structure is adjusted; If the node vitality value is greater than the set minimum threshold, it jumps to the next clean leaf node until all clean leaf nodes are traversed; (5) Reconstruct the life tree based on the frequency threshold and generate association rules; (6) Based on the association rule set obtained in step (5), complete the mining of implicit demand for customized product requirements.

2. The method for mining hidden demand for customized products based on data timeliness update according to claim 1 is characterized by: The step (3) specifically comprises: Calculate each transaction Vitality , the calculation formula is as follows: ； in, For the items in the transaction, It is a project In trading The relative importance of the factors in is the number of items traded; Vitality value based on each transaction , calculate the vitality value of each time leaf node , the calculation formula is as follows: ； in, For each transaction Time information, is the time information contained in the time root node, is the vitality attenuation factor.

3. The method for mining hidden demand for customized products based on data timeliness update according to claim 1 is characterized by: The steps of deleting the invalid path in step (4) are as follows: For a single invalid path, the time leaf node of the invalid path is deleted first, and the count of the time leaf node is subtracted from the count of other path nodes on the path; if the count of the path node is 0 after the count subtraction, the node is deleted; when all invalid paths are deleted, the life tree structure is updated and adjusted; The step (5) is specifically as follows: Mining rare items and frequent items through the tree of life structure, rare item support satisfy , the frequent item support satisfies ; in is the minimum threshold of rare items, is the maximum threshold of rare items, is the minimum threshold for frequent items; After obtaining the two types of items, the conditional pattern tree and frequent / rare pattern sets are obtained by eliminating the tree paths that do not meet the rare or frequent thresholds in the life tree structure to generate association rules.

4. The method for mining hidden demand for customized products based on data timeliness update according to claim 1 is characterized by: Classify the user's demand information and check whether it meets the product design requirements to obtain demand classification information; then convert the demand classification information into standard requirements; based on the standard requirements and the association rule set obtained in step (5), complete the mining of implicit requirements for customized product needs.

5. A system for mining hidden demand of customized products based on data timeliness update, used to implement the method for mining hidden demand of customized products based on data timeliness update according to any one of claims 1 to 4, characterized in that: include: Interactive interface, used to assist users in selecting products and expressing demand information; The demand analysis module is used to classify demand information, check whether it meets the product design, and output demand classification information; The requirement standardization module is used to convert the requirement classification information output by the requirement analysis module into standard requirements; The life tree construction module is used to compress and save the historical order information into a life tree structure to save the order data; The tree structure optimization module is used to calculate the vitality value of the time leaf nodes of the tree structure and delete invalid paths in the tree structure; The pattern extraction module is used to obtain the conditional pattern tree of popular items through the support threshold and the final tree structure, and integrate the statistical popular patterns; The rule extraction module is used to calculate and filter reliable association rules through popular pattern sets; The requirement mining module uses the standard requirements output by the requirement standardization module and the implicit requirement association rules output by the rule extraction module to mine implicit requirements and transform them into usable explicit requirements.

6. A device for mining hidden demand of customized products based on data timeliness update, characterized in that: It includes one or more processors for implementing the method for mining hidden demand of customized products based on data timeliness update as described in any one of claims 1-4.

7. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, it is used to implement the method for mining hidden demands of customized products based on data timeliness update as described in any one of claims 1 to 4.

Images