Method and device for obtaining matter relevance

Abstract

The present invention discloses a method and device for obtaining the relevance of items, and relates to the technical field of big data analysis and mining. A specific implementation of the method includes: calculating the first support of the items contained in multiple pieces of processing data, and screening and sorting the multiple items according to the first support; counting the number of occurrences of the processing results of each item in the application materials of other items; obtaining the central items from the multiple items according to the number of occurrences to establish a frequent item header table; establishing a frequent pattern tree according to the frequent item header table and the data item set corresponding to the processing data, and mining the unordered frequent item set; determining the superset of the unordered frequent item set, and extracting the ordered frequent item sequence in the superset. This implementation can mine the unordered frequent item set and the ordered frequent item sequence in the processing data at one time, with high mining efficiency, and is not affected by subjective factors, and has strong applicability.

Description

Technical Field

[0001] The present invention relates to the field of big data analysis and mining technology, and in particular to a method and device for obtaining the relevance of matters. Background Art

[0002] "Internet + Government Services" includes multiple items, but these items are not completely isolated. There is a certain relationship between them, and some items have a sequence requirement, while some items have no sequence requirement. For example, birth-related items: [birth medical certificate, vaccination certificate, household registration, medical insurance registration, social security card application], there is a strong logical relationship between these items, and users will handle such items in the corresponding order.

[0003] The traditional method of mining the correlation of items is to use the experience of business experts to manually sort out items from a business perspective and then discover the correlation between items. However, this method can only discover certain ordered frequent items with strong logical correlations, and it is difficult to discover disordered frequent items with non-logical correlations. In addition, this method relies on the understanding of the rules of items by business experts, has a large number of subjective factors, low mining efficiency, low accuracy, and poor applicability. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method, device, electronic device and computer-readable medium for obtaining the relevance of matters. The method first filters and sorts the handling matters based on a first filtering condition, and obtains the central matters from the handling matters, and then combines the central matters and the filtering and sorting results to obtain an unordered frequent matter set from the handling data, and then mines the ordered frequent sequence in the unordered frequent matter set. The unordered frequent matter set and the ordered frequent matter sequence in the handling data can be mined at one time, with high mining efficiency, not affected by subjective factors, and strong applicability.

[0005] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for obtaining matter relevance is provided.

[0006] A method for obtaining the relevance of items in an embodiment of the present invention includes: calculating the first support of the items contained in multiple items data, filtering and sorting the multiple items according to the first support and a set first screening condition to obtain the screening and sorting results; obtaining the application materials and corresponding processing results of the multiple items, and counting the number of occurrences of the processing results of each item in the multiple items in the application materials of other items; obtaining the central item from the multiple items according to the number of occurrences and a set second screening condition, so as to establish a frequent item header table according to the central item and the screening and sorting results; establishing a frequent pattern tree according to the frequent item header table and the data item set corresponding to the item data, and mining an unordered frequent item set from the frequent pattern tree; wherein the data item set includes the filtered items, or includes the filtered items and the central item; determining a superset of the unordered frequent item set from the multiple data item sets, and extracting an ordered frequent item sequence in the superset based on a pattern mining algorithm.

[0007] Optionally, the frequent item header table includes the filtered to-do items and the central item; mining the unordered frequent item set from the frequent pattern tree includes: finding the conditional pattern bases corresponding to the to-do items in the frequent item header table from the bottom items of the frequent item header table upward in sequence; and recursively mining the frequent pattern tree using the conditional pattern bases to obtain the unordered frequent item set.

[0008] Optionally, establishing a frequent pattern tree based on the frequent item header table and the data item set corresponding to the task data includes: reading the data item sets corresponding to the multiple task data, and processing according to the following steps each time one of the data item sets is read: inserting the task items of the read data item set as nodes into the frequent pattern tree in the order in which the task items are sorted, and updating the count of the node; establishing a link relationship between the task items in the frequent item header table and the corresponding node.

[0009] Optionally, the recursively mining the frequent pattern tree using the conditional pattern base to obtain an unordered frequent item set includes: recursively mining the frequent pattern tree using the conditional pattern base to obtain an initial frequent item set; and screening the initial frequent item set based on a set improvement rate condition requirement to obtain an unordered frequent item set.

[0010] Optionally, the pattern mining algorithm is based on extracting the ordered frequent event sequence in the superset, including: searching for prefixes of length 1 and corresponding projections in the superset, and calculating the second support of the prefix of length 1 based on the projection; filtering the superset based on the second support to obtain corresponding filtering results and frequent sequences corresponding to the prefix of length 1; repeating the following steps until no mining can be done: recursively mining the frequent sequence corresponding to the prefix of the current length to obtain the frequent sequence corresponding to the prefix of the next length, and incrementing the current length; wherein the initial value of the current length is 1.

[0011] Optionally, determining a superset of the unordered frequent event set from the multiple data item sets includes: checking whether the unordered frequent event set is a subset of the data item set; and if determining that the unordered frequent event set is a subset of the data item set based on the test result, then treating the data item set as a superset of the unordered frequent event set.

[0012] Optionally, the first filtering condition is used to filter out the work items whose first support is greater than or equal to a set first support threshold from the multiple work items; the second filtering condition is used to filter out the work items whose third support is greater than or equal to a set third support threshold from the multiple work items; wherein the first support is calculated based on the number of times the work item appears in the multiple work item data and the total number of times all the work items appear in the multiple work item data; the third support is calculated based on the number of occurrences and the number of the work items; the establishment of a frequent item header table based on the central item and the filtering and sorting results includes: maintaining the work items in the filtering and sorting results into the initialized frequent item header table in descending order of the first support; determining that the frequent item header table does not contain the central item, and adding the central item to the frequent item header table; the method also includes: counting the number of times the work item appears in the multiple work item data, the number of the work items, and the total number of times all the work items appear in the multiple work item data.

[0013] To achieve the above objective, according to another aspect of an embodiment of the present invention, a device for acquiring matter relevance is provided.

[0014] An acquisition device for item association according to an embodiment of the present invention comprises: a screening and sorting module, which is used to calculate the first support of items contained in multiple items data, and screen and sort the multiple items according to the first support and a set first screening condition to obtain the screening and sorting results; an acquisition statistics module, which is used to obtain the application materials and corresponding processing results of the multiple items, and count the number of occurrences of the processing results of each item in the multiple items in the application materials of other items; a linked list establishment module, which is used to obtain the central item from the multiple items according to the number of occurrences and a set second screening condition, so as to establish a frequent item header table according to the central item and the screening and sorting results; a first mining module, which is used to establish a frequent pattern tree according to the frequent item header table and the data item set corresponding to the item data, and mine an unordered frequent item set from the frequent pattern tree; wherein the data item set includes the screened items, or includes the screened items and the central item; a second mining module, which is used to determine a superset of the unordered frequent item set from the multiple data item sets, and extract an ordered frequent item sequence in the superset based on a pattern mining algorithm.

[0015] Optionally, the frequent item header table includes the filtered to-do items and the central item; the first mining module is also used to find the conditional pattern bases corresponding to the to-do items in the frequent item header table from the bottom items of the frequent item header table upward in sequence; use the conditional pattern bases to recursively mine the frequent pattern tree to obtain an unordered frequent item set.

[0016] Optionally, the first mining module is also used to read the data item sets corresponding to the multiple work-related data, and each time a data item set is read, it is processed according to the following steps: according to the sorted order of the work-related items, the work-related items of the read data item set are inserted as nodes into the frequent pattern tree, and the count of the node is updated; and a link relationship between the work-related items in the frequent item header table and the corresponding nodes is established.

[0017] Optionally, the first mining module is further used to recursively mine the frequent pattern tree using the conditional pattern base to obtain an initial frequent item set; based on a set promotion rate condition requirement, the initial frequent item set is screened to obtain an unordered frequent item set.

[0018] Optionally, the second mining module is also used to find prefixes with a length of 1 and corresponding projections in the superset, and calculate the second support of the prefix with a length of 1 based on the projection; filter the superset according to the second support to obtain corresponding filtering results and frequent sequences corresponding to the prefixes with a length of 1; repeat the following steps until mining is impossible: recursively mine the frequent sequences corresponding to the prefixes of the current length to obtain the frequent sequences corresponding to the prefixes of the next length, and increment the current length; wherein the initial value of the current length is 1.

[0019] Optionally, the second mining module is further used to verify whether the unordered frequent event set is a subset of the data item set; if it is determined according to the verification result that the unordered frequent event set is a subset of the data item set, the data item set is regarded as a superset of the unordered frequent event set.

[0020] Optionally, the first screening condition is used to screen out the work items whose first support is greater than or equal to a set first support threshold from the multiple work items; the second screening condition is used to screen out the work items whose third support is greater than or equal to a set third support threshold from the multiple work items; wherein the first support is calculated based on the number of times the work item appears in the multiple work item data and the total number of times all the work items appear in the multiple work item data; the third support is calculated based on the number of occurrences and the number of the work items; the linked list establishment module is also used to maintain the work items in the screening and sorting results into the initialized frequent item header table in descending order of the first support; and determine that the frequent item header table does not contain the central item, and add the central item to the frequent item header table; the device also includes: a statistical module, which is used to count the number of times the work item appears in the multiple work item data, the number of the work items, and the total number of times all the work items appear in the multiple work item data.

[0021] To achieve the above objective, according to another aspect of an embodiment of the present invention, an electronic device is provided.

[0022] An electronic device according to an embodiment of the present invention comprises: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement a method for obtaining the relevance of matters according to an embodiment of the present invention.

[0023] To achieve the above objective, according to another aspect of the embodiments of the present invention, a computer readable medium is provided.

[0024] A computer-readable medium according to an embodiment of the present invention stores a computer program, which, when executed by a processor, implements a method for obtaining the relevance of matters according to an embodiment of the present invention.

[0025] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer program product is provided.

[0026] A computer program product according to an embodiment of the present invention includes a computer program, and when the program is executed by a processor, a method for obtaining the relevance of matters according to an embodiment of the present invention is implemented.

[0027] An embodiment of the above invention has the following advantages or beneficial effects: by first filtering and sorting the handling items based on the first filtering condition, and obtaining the central items from the handling items, and then combining the central items and the filtering and sorting results, obtaining an unordered frequent item set from the handling data, and then mining the ordered frequent sequence in the unordered frequent item set, the unordered frequent item set and the ordered frequent item sequence in the handling data can be mined at one time, the mining efficiency is high, and it is not affected by subjective factors and has strong applicability.

[0028] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0030] Figure 1 is a schematic diagram of the main steps of a method for obtaining matter relevance according to an embodiment of the present invention;

[0031] Figure 2 is a schematic diagram of the main process of a method for obtaining matter relevance according to another embodiment of the present invention;

[0032] Figure 3 is a schematic diagram of the structure of a frequent pattern tree in a method for acquiring matter relevance according to another embodiment of the present invention;

[0033] Figure 4 is a schematic diagram of the main process of a method for obtaining matter relevance according to another embodiment of the present invention;

[0034] Figure 5 is a schematic diagram of main modules of a device for acquiring matter relevance according to an embodiment of the present invention;

[0035] Figure 6 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0036] Figure 7It is a schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0038] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0039] Government services refer to administrative services such as licensing, confirmation, adjudication, rewards, and penalties provided by governments at all levels, relevant departments, and public institutions to social groups, enterprises, institutions, and individuals in accordance with laws and regulations. Government services may include administrative licensing, administrative payments, administrative confirmations, administrative rewards, administrative adjudications, and other administrative power matters. They may also include granting matters other than administrative power matters, i.e., public service matters. These matters are not completely isolated, and there is a certain connection between them.

[0040] In actual applications, some matters are highly correlated and have a strong order of precedence, such as birth-related matters [birth medical certificate, vaccination certificate, household registration, medical insurance registration, social security card application]. Users will handle such matters in the corresponding order; while some matters are highly correlated, but there is no strong order of precedence between them, such as water, electricity and gas transfer-related matters [tap water name transfer, electricity transfer, natural gas transfer]. When handling such matters, users may handle any other matters first, and there is no strong order of precedence between them.

[0041] In order to mine the associations between items, the current common practice is to use the experience of business experts to manually sort out items from a business perspective and then discover the associations between items. However, this method can only discover certain ordered frequent items with strong logical associations, and it is difficult to discover disordered frequent items with no logical associations. In addition, this method relies on the business experts' understanding of the item rules, has a large number of subjective factors, low mining efficiency, low accuracy, and poor applicability.

[0042] In order to solve the above problems, this embodiment provides a method for obtaining the relevance of items, which can mine the relevance patterns of these ordered or disordered items in one go and efficiently.

[0043] Figure 1FIG. 1 is a schematic diagram of the main steps of a method for obtaining the relevance of items according to an embodiment of the present invention. Figure 1 As shown, the method for obtaining the relevance of matters in the embodiment of the present invention mainly includes the following steps:

[0044] Step S101: Calculate the first support of the items included in the multiple items data, filter and sort the multiple items according to the first support and the set first screening condition, and obtain the screening and sorting result. The item data is the data generated by the person handling a specific service item (such as a government service item), which may include the person information, item, status, etc., wherein the item is the item information of the item handled by the person, such as the item identifier.

[0045] The first support of an item can be obtained based on the number of times the item appears in multiple items. Specifically, the first support of each item can be the number of times the corresponding item appears in multiple items, or it can be the ratio of the number of times the item appears in multiple items to the total number of times all items appear in multiple items. In the embodiment, a first filtering condition needs to be set in advance, and the filtering condition is used to filter out items from multiple items whose first support is greater than or equal to the set first support threshold.

[0046] Among them, the first support threshold can be customized. If the first support of an item is less than the first support threshold, the item is considered to be a non-frequent item and is discarded; if the first support of an item is greater than or equal to the first support threshold, the item is considered to be a frequent item and is retained.

[0047] By screening and sorting all the items contained in the multiple items data in the above manner, the screening and sorting results can be obtained. In the embodiment, all the items can be screened first to obtain the screened items; then the screened items can be sorted in descending order based on the first support to obtain the screening and sorting results. All the items can also be sorted in descending order based on the first support to obtain the sorted items; then the sorted items can be screened based on the first support threshold to obtain the screening and sorting results.

[0048] Step S102: Obtain the application materials and corresponding handling results of multiple items, and count the number of times the handling results of each item in the multiple items appear in the application materials of other items. Application materials refer to the materials submitted by the person handling the items, and handling results refer to the materials produced after the items are completed. In this step, it is necessary to disassemble each item in the handling data, find out the application materials and corresponding handling results corresponding to each item, and then compare whether the handling results of each item exist in the application materials of other items, and count the number of occurrences.

[0049] Step S103: According to the occurrence count and the set second filtering condition, a central item is obtained from the multiple items to establish a frequent item header table according to the central item and the filtering and sorting results. The second filtering condition is used to filter out items whose third support is greater than or equal to a set third support threshold from the multiple items. The third support can be calculated based on the occurrence count and the number of items, and the third support threshold can be customized.

[0050] In the embodiment, the central item refers to an item that is related to many items in the item data. Specifically, if the processing results of some items are the processing conditions of many other items, or other items can only be processed after the processing of some items is completed, that is, these items exist in the item data as prerequisites for the processing of other items and meet the defined third support threshold requirement, then the item is considered to be a central item.

[0051] In the embodiment, the frequent item header table includes the filtered work items, the central items and the corresponding quantities. When establishing the frequent item header table, the work items in the filtered and sorted results can be maintained in the initialized frequent item header table in descending order of the first support; and then if it is determined that the frequent item header table does not contain the central item, the central item is added to the frequent item header table to obtain the final frequent item header table.

[0052] Step S104: According to the frequent item header table and the data item set corresponding to the work item data, a frequent pattern tree is established, and unordered frequent item sets are mined from the frequent pattern tree. A frequent pattern tree (FP-tree) is a special tree structure consisting of a root node (value is null), a frequent item header table and an item prefix tree.

[0053] Each node of the item prefix tree consists of three fields: item_name (item identifier), count (count) and node_link (node ​​link), where item_name records the identifier of the item represented by the node, count records the number of transactions in the subpath to reach the node, and node_link is used to connect to the next node with the same identifier in the tree. If there is no next node with the same identifier, the value is "null".

[0054] After filtering the work items of each work item data, the filtering result, or the filtering result and the central matter are used as the data item set of the work item data, that is, the data item set of a work item data includes the work items obtained after filtering the work items of the work item data, and if the central matter is filtered out during the filtering process, the data item set also includes the central matter. In the embodiment, it is necessary to read the data item sets corresponding to each work item data one by one, and insert the work items of the data set as nodes into the frequent pattern tree in the order of the work items, update the node count, and establish the link relationship between the work items in the frequent item header table and the corresponding nodes.

[0055] After all data item sets are inserted into the frequent pattern tree, the establishment of the frequent pattern tree is completed. After the frequent pattern tree is established, the conditional pattern base can be used to mine unordered frequent item sets from the frequent pattern tree.

[0056] Step S105: Determine a superset of the unordered frequent event set from the plurality of data item sets, and extract the ordered frequent event sequence in the superset based on the pattern mining algorithm. Check whether the unordered frequent event set is a subset of each data item set. If the unordered frequent event set is a subset of a data item set, then the data item set is a superset of the unordered frequent event set.

[0057] The pattern mining algorithm is used to mine ordered frequent event sequences in unordered frequent event sets, and FreeSpan algorithm, PrefixSpan algorithm, etc. can be used. In the embodiment, the superset and the second support threshold are used as inputs of the pattern mining algorithm. After being processed by the pattern mining algorithm, all ordered frequent event sequences that meet the support requirements can be output.

[0058] Among them, the FreeSpan algorithm is a sequence mining algorithm based on pattern projection. The basic idea of ​​the algorithm is: using the current mining frequent sequence set, recursively project the sequence database onto a set of smaller projection databases, and grow subsequences on each projection database. The full name of the PrefixSpan algorithm is Prefix-Projected Pattern Growth, which is prefix projection pattern mining. The basic idea of ​​the algorithm is: when projecting, not considering all possible frequent subsequences, only checking the prefix sequence, and then projecting the corresponding suffix into the projection database.

[0059] Figure 2 FIG. 1 is a schematic diagram of the main process of a method for obtaining the relevance of items according to another embodiment of the present invention. Figure 2 As shown, the method for obtaining the relevance of matters in the embodiment of the present invention mainly includes the following steps:

[0060] Step S201: Obtain a data set of documents, traverse the data of the documents in the data set of documents, and calculate the first support of the documents' items in the data set of documents. The documents data includes the document handler and the document handler's items. In the embodiment, the document dataset may be a data structure as shown in Table 1.

[0061] Table 1

[0062]

[0063]

[0064]

[0065] As shown in Table 1, there are 9 pieces of data in the data set, each piece of data includes multiple items, and there are 15 items in total. The total number of times these 15 items appear in the 9 pieces of data is 65. Traverse the data of the data set and calculate the first support of each item. The calculation formula of the first support can be as follows:

[0066] support i =num i / total

[0067] Formula 1

[0068] In the formula, support i is the first support of the i-th task, num i is the number of times the i-th item appears in multiple pieces of data, and total is the total number of times all items appear in multiple pieces of data.

[0069] Step S202: Filter and sort multiple items according to the first support and the set first screening condition to obtain the screening and sorting results. The first support threshold of the first screening condition can be customized, such as 5%. This step is used to compare the first support of each item with the first support threshold, retain the items with the first support greater than or equal to the first support threshold, and discard the items with the first support less than the first support threshold.

[0070] Taking the first support threshold of 5% as an example, the first support of revoking marriage registration, marriage registration, real estate registration, gas appliance installation qualification license, social insurance registration and household registration is less than 5% and needs to be discarded. The first support of business license, fire safety inspection, tax registration, sewage discharge permit, Internet cafe safety permit, Internet cafe charging permit, food business permit, network culture business permit and ID card processing is greater than 5% and needs to be retained.

[0071] After screening the handling items of the handling data, an updated handling data set can be obtained, as shown in Table 2.

[0072] Table 2

[0073]

[0074]

[0075] Based on the first support, the selected items are sorted in descending order. The sorting results are shown in Table 3:

[0076] Table 3

[0077] Matters to be handled quantity Support Industrial and commercial business license 9 14％ Fire safety inspection 9 14％ Tax registration 8 12％ Pollutant Discharge Permit 5 8％ Internet Cafe Security Permit 5 8％ Internet cafe charging license 5 8％ Food business license 5 8％ Internet culture business license 4 6％ ID card application 4 6％

[0078] Step S203: Obtain application materials and corresponding processing results of multiple items, and count the number of occurrences of the processing results of each item in the multiple items in the application materials of other items. Disassemble each item in the processing data, obtain the application materials and corresponding processing results corresponding to each item, and then compare whether the processing results of each item exist in the application materials of other items, and count the number of occurrences. Among them, other items are items other than the current item in all items.

[0079] In the embodiment, the application materials corresponding to each handling matter and the corresponding handling results are shown in Table 4.

[0080] Table 4

[0081]

[0082] Combined with Table 4, we can count the number of times the handling result of each item appears in the application materials of other items. The specific results are shown in Table 5.

[0083] Table 5

[0084]

[0085]

[0086] Step S204: According to the number of occurrences and the set second filtering condition, the central item is obtained from multiple items, so as to establish a frequent item header table according to the central item and the filtering and sorting results. In this step, the third support is first calculated based on the number of occurrences and the number of items, and then the third support of each item is compared with the third support threshold of the second filtering condition to obtain the central item. The calculation formula of the third support can be as follows:

[0087]

[0088] In the formula, is the third support of the i-th task, is the number of times the result of the i-th item appears in the application materials of other items, total item The number of items to be handled.

[0089] The third support threshold of the second screening condition can be customized, for example, 70%. According to Formula 2, the third support of each item is calculated, and it can be seen that only the third support of the ID card item is 73.3%, which is greater than 70%, and the third support of other items does not reach the third support threshold. Therefore, the central item is the ID card item.

[0090] The frequent item header table is used to maintain the same work items of different work item data, including the work items after screening, the central items and the corresponding quantities. In the embodiment, the work items in the screening and sorting results are first maintained in the initialized frequent item header table in descending order of the first support; then, when it is determined that the frequent item header table does not contain the central item, the central item and the corresponding quantity are added to the frequent item header table. In combination with the above example, if Table 3 does not contain the central item, the frequent item header table includes the first two columns of data in Table 3, as well as the central item and the corresponding quantity; if Table 3 contains the central item, the frequent item header table is the first two columns of data in Table 3.

[0091] Step S205: Establish a frequent pattern tree based on the frequent item header table and the data item set corresponding to the case data. A data item set of a case data includes the case items after the case data is filtered, or includes the case items and the core items after the filter. Specifically, if there is no core item in the case data, or if there is a core item in the case data and it is not filtered out during the filter process, the data item set includes the filtered case items; if there is a core item in the case data and it is deleted during the filter process, the data item set includes the core items in addition to the filtered case items. For example, the case data numbered 01 has a data item set of [business license, fire safety inspection, tax registration, pollution discharge permit, food business license, ID card processing].

[0092] In this step, the data item sets corresponding to multiple pieces of work-related data are read, and each time a data item set is read, it is processed according to the following steps: the work-related items of the read data item set are inserted as nodes into the frequent pattern tree in the order in which the work-related items are sorted, and the node count is updated; a link relationship between the work-related items in the frequent item header table and the corresponding nodes is established.

[0093] Figure 3 FIG. 1 is a schematic diagram of the structure of a frequent pattern tree in a method for obtaining item relevance according to another embodiment of the present invention. Figure 3 As shown in the figure, the frequent pattern tree includes a root node (null), a frequent item header table, and an item prefix tree. At the beginning, the frequent pattern tree has no data. When the frequent pattern tree is established, the items of the data item set are read one by one and inserted into the frequent pattern tree in the order of the items. The nodes with the highest order are ancestor nodes, and the nodes with the lowest order are descendant nodes. If there is a common ancestor node, the count of the corresponding common ancestor node is increased by 1.

[0094] After inserting the frequent pattern tree, if a new node appears, the node corresponding to the frequent item header table will be linked to the new node through the node linked list. The establishment of the frequent pattern tree is completed until all data are inserted into the frequent pattern tree.

[0095] Taking the insertion of the case data numbered 01 and 02 as an example, first read the data item set of the case data numbered 01: [Industrial and Commercial Business License, Fire Safety Inspection, Tax Registration, Pollutant Discharge Permit, Food Business License, ID Card Processing]. At this time, the frequent pattern tree has no nodes. Therefore, each case item is abstracted into a corresponding node, and inserted into the frequent pattern tree in the order in which the case items are sorted. The count of all nodes is 1, and the frequent item header table is linked to the corresponding new node through the node linked list.

[0096] Then read the data item set of the application data numbered 02: [Business license, fire safety inspection, tax registration, pollution discharge permit, food business license]. The above application items and the original frequent pattern tree have common ancestor nodes, so there is no need to add new nodes. Then add 1 to the count of each node to 2.

[0097] Update the data item set numbered 03-09 in the above way to get Figure 3 The frequent pattern tree shown.

[0098] Step S206: mining the unordered frequent item set from the frequent pattern tree. Specifically, the conditional pattern bases corresponding to the items in the frequent item header table are found from the bottom item of the frequent item header table upwards; then the conditional pattern bases are used to recursively mine the frequent pattern tree to obtain the unordered frequent item set. In the embodiment, the bottom item of the frequent item header table is the node of ID card application.

[0099] The conditional pattern base is the prefix path combination of the node to be mined. The conditional pattern base corresponding to the current node can be obtained by finding the path ending with the current node in the frequent pattern tree, setting the count of all ancestor nodes in the path to the count of the current node, and deleting the nodes whose counts are less than the first support threshold.

[0100] Based on the conditional pattern base, the frequent two-item set of the current node can be obtained, and the frequent two-item set can be recursively merged to obtain the frequent three-item set. In this way, the largest frequent item set of the current node can be mined recursively. In practical applications, the largest frequent item set can be used as the unordered frequent item set of the current node.

[0101] In a preferred embodiment, the largest frequent item set can also be used as the initial frequent item set, and then the initial frequent item set is screened based on the set promotion rate condition to obtain the disordered frequent item set. The promotion rate condition requirement is that the probability of the event occurring in the initial frequent item set is greater than or equal to the probability of it not occurring in the initial frequent item set.

[0102] The lift rate condition is used to determine the correlation between the handling events. The lift rate is greater than 1 and the higher it is, the higher the positive correlation is. The lift rate is less than 1 and the lower it is, the higher the negative correlation is. The lift rate = 1 indicates no correlation. This process will exclude the ID card handling event, and the unordered frequent event set is as follows:

[0103] Disordered frequent items set 1: {Business license, fire safety inspection, tax registration, sewage discharge permit, food business license}

[0104] Disordered frequent items set 2: {Business license, fire safety inspection, Internet cafe safety permit, Internet culture business permit, tax registration, Internet cafe charging permit}

[0105] Step S207: Determine a superset of the unordered frequent item set from multiple data item sets. Check whether the unordered frequent item set is a subset of each data item set. If the unordered frequent item set is a subset of a certain data item set, then take the data item set as a superset of the unordered frequent item set.

[0106] Step S208: Based on the PrefixSpan algorithm, an ordered frequent event sequence is mined from the superset. A sequence is a set of ordered items that are not necessarily directly continuous, but still satisfy the order. The goal of the PrefixSpan algorithm is to mine frequent sequences that satisfy the set second support threshold. The possible ordered frequent event sequences in the superset and the set second support threshold are input into the PrefixSpan algorithm, and the ordered frequent event sequence that satisfies the support requirement is calculated. The PrefixSpan algorithm processes the input data as follows:

[0107] Find the prefix of length 1 and the corresponding projection in the superset, and calculate the second support of the prefix of length 1 according to the projection; then filter the superset according to the second support and the second support threshold to obtain the filtering result and the frequent sequence corresponding to the prefix of length 1;

[0108] Repeat the following steps until no more mining is possible: recursively mine the frequent sequences corresponding to the prefix of the current length, obtain the frequent sequences corresponding to the prefix of the next length, and increment the current length. The initial value of the current length is 1, and the implementation process of recursive mining is:

[0109] Find the projection corresponding to the prefix of the current length. If the projection is empty, return recursively. If the projection is not empty, continue to count the second support of each item in the corresponding projection. If the second support of all items is less than the second support threshold, return recursively. If there is at least one item whose second support is greater than or equal to the second support threshold, merge the items that meet the second support threshold requirements with the current prefix to obtain several new prefixes.

[0110] It should be noted that since the items are arranged in chronological order, items in different orders are different items. Therefore, in the PrefixSpan algorithm, items in different orders such as {Business license, fire safety inspection, tax registration} and {Fire safety inspection, tax registration, Business license} should calculate the support separately.

[0111] After the above processing, the unordered frequent event set 1 is obtained as an ordered frequent event sequence, while the unordered frequent event set 2 cannot obtain an ordered frequent event sequence. Therefore, the ordered frequent event sequence is:

[0112] Set of disordered and frequent items 1: {1. Industrial and commercial business license 2. Fire safety inspection 3. Tax registration 4. Pollutant discharge permit 5. Food business license}.

[0113] Figure 4 FIG. 1 is a schematic diagram of the main process of a method for obtaining the relevance of items according to another embodiment of the present invention. Figure 4 As shown, the method for obtaining the relevance of matters in the embodiment of the present invention mainly includes the following steps:

[0114] Step S401: Obtain a data set of documents, traverse the documents data of the data set of documents, count the number of times each document item appears in multiple documents data, and the total number of times all documents items appear in multiple documents data. Combined with Table 1, taking the document item of industrial and commercial business license as an example, the number of this document item is 9, and the total number of all document items is 65.

[0115] Step S402: Calculate the first support of each item according to the number of times and the total number of times the item appears in multiple items. Still taking the industrial and commercial business license as an example, according to formula 1, its first support = 9 / 65*100% = 14%.

[0116] Step S403: Screen and sort the items according to the first support and the set first screening condition to obtain the screening and sorting results. The specific implementation of this step is shown in step S202, which will not be described here.

[0117] Step S404: Obtain application materials and corresponding handling results of multiple items, and count the number of occurrences of the handling results of each item in the multiple items in the application materials of other items. The specific implementation of this step is shown in step S203, which will not be repeated here.

[0118] Step S405: According to the number of occurrences and the set second screening condition, the central item is obtained from the multiple items to establish a frequent item header table according to the central item and the screening and sorting results. The specific implementation of this step is shown in step S204, which will not be described here.

[0119] Step S406: According to the codes set for each case event, the filtered case data is converted into codes to obtain a data item set corresponding to the case data. In the embodiment, each case item corresponds to a code, which can be numbers, letters, etc. For example, the codes of the industrial and commercial business license, fire safety inspection, tax registration, sewage discharge permit, Internet cafe safety permit, Internet cafe charging permit, food business permit and network culture business permit are A, C, F, K, N, B, L, N and M respectively.

[0120] Step S407: Establish a frequent pattern tree according to the frequent item header table and the data item set corresponding to the case data. The specific implementation of this step is shown in step S205, which will not be described here.

[0121] Step S408: Based on the conditional pattern base and the set promotion rate condition requirement, unordered frequent item sets are mined from the frequent pattern tree. The specific implementation of this step is shown in step S206, which will not be described here.

[0122] Step S409: Determine the superset of the unordered frequent item set from the updated multiple data item sets. It should be noted that based on the requirement of the promotion rate condition, the ID card application item will be excluded, so the corresponding data item set needs to be updated. At this time, the data item sets corresponding to the application data numbered 01-09 are: {ACFKN}, {ACFKN}, {BAFLMC}, {ACBMFLK}, {ACFKN}, {ALMFBC}, {ACFKN}, {ACBMFL} and {BANLC}.

[0123] The codes of unordered frequent event sets 1 and 2 are {ACFKN} and {ACBMFL} respectively. The superset of unordered frequent event set 1 is shown in Table 6, and the superset of unordered frequent event set 2 is shown in Table 7.

[0124] Table 6

[0125] 01 {ACFKN} 02 {ACFKN} 05 {ACFKN} 07 {ACFKN}

[0126] Table 7

[0127] 03 {BAFLMC} 04 {ACBMFLK} 06 {ALMFBC} 08 {ACBMFL}

[0128] Step S410: Based on the PrefixSpan algorithm, mining ordered frequent event sequences from the superset. In the embodiment, the second support threshold is defined as 5%, then the prefix with a length of 1 in the superset of the unordered frequent event set 1 is 、 <c> 、 <f> 、 <k> 、 <n>, and their corresponding projections are <acfkn> 、 <cfkn> 、 <fkn> 、 <kn> 、 <n> , through calculation, we can know that the second support of A, C, F, K, and N all reach the threshold, and the frequent sequence with a length of 1 is:< / n> < / kn> < / fkn> < / cfkn> < / acfkn> < / n> < / k> < / f> < / c> 、 <c> 、 <f> 、 <k> 、 <n>.

[0129] Then, recursively mine the frequent sequences with a length of 1 to obtain the frequent sequences with a prefix length of 2.< / n> < / k> < / f> < / c> For example, the ordered frequent sequence with a length of 2 is: <ac> 、 <af> 、 <ak> 、 <an>. The ordered frequent sequences are mined in increasing increments until no more can be mined. Finally, five ordered frequent sequences that meet the second support threshold are obtained. <acfkn>. The five ordered frequent sequences are converted by encoding to obtain the ordered frequent sequence: {Business license, fire safety inspection, sewage discharge permit, tax registration, food business license}, which is consistent with the unordered frequent item set 1. Therefore, the unordered frequent item set 1 is an ordered frequent event sequence.

[0130] The prefix of length 1 in the superset of the unordered frequent item set 2 is:< / acfkn> < / an> < / ak> < / af> < / ac> 、 <c> 、 、 <m> 、 <f> 、 <l> 、 <k> , through calculation, we can know that the second support of other elements except A is less than the second support threshold, so the frequent sequence with a length of 1 is:< / k> < / l> < / f> < / m> < / c> . Since the prefix is The second support of all items in the corresponding projection is less than the second support threshold, and mining cannot continue. Therefore, no frequent sequence of length 2 can be obtained, and mining is terminated. Finally, no ordered frequent sequence can be obtained, so the unordered frequent item set 2 is an unordered frequent item set.

[0131] In a preferred embodiment, if the following association pattern is found: [applying for a marriage certificate, purchasing a second-hand house, applying for a housing loan, and withdrawing provident funds], it can be seen that most of the applicants will first apply for a marriage certificate, then purchase a second-hand house, then apply for a housing loan, and finally withdraw provident funds. Therefore, it is possible to consider declaring these related matters at one time and merging the approval and registration to further shorten the processing time limit and improve processing efficiency.

[0132] The method for obtaining the relevance of matters in this embodiment first retrieves an unordered frequent matter set that meets the first support threshold requirement from the case data based on the frequent pattern tree, and then uses the Prefixspan algorithm to mine the ordered frequent sequences in the unordered frequent matter set. This method can mine the associations between ordered or unordered matters in one go and efficiently, thereby reducing the time complexity and space complexity of obtaining the relevance of matters.

[0133] Figure 5 FIG. 1 is a schematic diagram of the main modules of the device for obtaining the relevance of items according to an embodiment of the present invention. Figure 5 As shown, the device 500 for obtaining the item relevance in the embodiment of the present invention mainly includes:

[0134] The screening and sorting module 501 is used to calculate the first support of the items included in the multiple items data, screen and sort the multiple items according to the first support and the set first screening condition, and obtain the screening and sorting results. The item data is the data generated by the person handling a specific service item (such as a government service item), and may include the person information, item, status, etc., wherein the item is the item information of the item handled by the person, such as the item identifier.

[0135] The first support of an item can be obtained based on the number of times the item appears in multiple items. Specifically, the first support of each item can be the number of times the corresponding item appears in multiple items, or it can be the ratio of the number of times the item appears in multiple items to the total number of times all items appear in multiple items. In the embodiment, a first filtering condition needs to be set in advance, and the filtering condition is used to filter out items from multiple items whose first support is greater than or equal to the set first support threshold.

[0136] Among them, the first support threshold can be customized. If the first support of an item is less than the first support threshold, the item is considered to be a non-frequent item and is discarded; if the first support of an item is greater than or equal to the first support threshold, the item is considered to be a frequent item and is retained.

[0137] By screening and sorting all the items contained in the multiple items data in the above manner, the screening and sorting results can be obtained. In the embodiment, all the items can be screened first to obtain the screened items; then the screened items can be sorted in descending order based on the first support to obtain the screening and sorting results. All the items can also be sorted in descending order based on the first support to obtain the sorted items; then the sorted items can be screened based on the first support threshold to obtain the screening and sorting results.

[0138] The acquisition statistics module 502 is used to obtain the application materials and corresponding processing results of the multiple items, and count the number of times the processing results of each item in the multiple items appear in the application materials of other items. The application materials are the materials submitted by the person handling the items, and the processing results are the materials produced after the items are completed. In this step, it is necessary to disassemble each item in the processing data, find out the application materials and corresponding processing results corresponding to each item, and then compare whether the processing results of each item exist in the application materials of other items, and count the number of occurrences.

[0139] The linked list establishment module 503 is used to obtain the central item from the multiple items according to the number of occurrences and the set second screening condition, so as to establish a frequent item header table according to the central item and the screening and sorting results. The second screening condition is used to filter out items whose third support is greater than or equal to the set third support threshold from the multiple items. The third support can be calculated based on the number of occurrences and the number of items, and the third support threshold can be customized.

[0140] In the embodiment, the central item refers to an item that is related to many items in the item data. Specifically, if the processing results of some items are the processing conditions of many other items, or other items can only be processed after the processing of some items is completed, that is, these items exist in the item data as prerequisites for the processing of other items and meet the defined third support threshold requirement, then the item is considered to be a central item.

[0141] In the embodiment, the frequent item header table includes the filtered work items, the central items and the corresponding quantities. When establishing the frequent item header table, the work items in the filtered and sorted results can be maintained in the initialized frequent item header table in descending order of the first support; and then if it is determined that the frequent item header table does not contain the central item, the central item is added to the frequent item header table to obtain the final frequent item header table.

[0142] The first mining module 504 is used to establish a frequent pattern tree according to the frequent item header table and the data item set corresponding to the task data, and mine an unordered frequent item set from the frequent pattern tree; wherein the data item set includes the filtered task items, or includes the filtered task items and the central item.

[0143] After filtering the work items of each work item data, the filtering result, or the filtering result and the central matter are used as the data item set of the work item data, that is, the data item set of a work item data includes the work items obtained after filtering the work items of the work item data, and if the central matter is filtered out during the filtering process, the data item set also includes the central matter. In the embodiment, it is necessary to read the data item sets corresponding to each work item data one by one, and insert the work items of the data set as nodes into the frequent pattern tree in the order of the work items, update the node count, and establish the link relationship between the work items in the frequent item header table and the corresponding nodes.

[0144] After all data item sets are inserted into the frequent pattern tree, the establishment of the frequent pattern tree is completed. After the frequent pattern tree is established, the conditional pattern base can be used to mine unordered frequent item sets from the frequent pattern tree.

[0145] The second mining module 505 is used to determine the superset of the unordered frequent event set from the multiple data item sets, and extract the ordered frequent event sequence in the superset based on the pattern mining algorithm. Check whether the unordered frequent event set is a subset of each data item set. If the unordered frequent event set is a subset of a data item set, then the data item set is a superset of the unordered frequent event set.

[0146] The pattern mining algorithm is used to mine ordered frequent event sequences in unordered frequent event sets, and FreeSpan algorithm, PrefixSpan algorithm, etc. can be used. In the embodiment, the superset and the second support threshold are used as inputs of the pattern mining algorithm. After being processed by the pattern mining algorithm, all ordered frequent event sequences that meet the support requirements can be output.

[0147] In addition, the device 500 for obtaining the item relevance of the embodiment of the present invention may further include: a statistical module ( Figure 5 (not shown in the figure), this module is used to count the number of the items to be handled and the total number of all the items to be handled.

[0148] From the above description, it can be seen that by first obtaining an unordered frequent event set that meets the first support requirement from the case data, and then mining the ordered frequent sequence in the unordered frequent event set, the unordered frequent event set and the ordered frequent event sequence in the case data can be mined at one time. The mining efficiency is high, it is not affected by subjective factors, and it has strong applicability.

[0149] Figure 6 An exemplary system architecture 600 is shown to which the method for acquiring matter relevance or the apparatus for acquiring matter relevance according to the embodiments of the present invention can be applied.

[0150] like Figure 6 As shown, system architecture 600 may include terminal devices 601, 602, 603, network 604 and server 605. Network 604 is used to provide a medium for communication links between terminal devices 601, 602, 603 and server 605. Network 604 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0151] Users can use terminal devices 601, 602, 603 to interact with server 605 through network 604 to receive or send messages, etc. Terminal devices 601, 602, 603 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, etc.

[0152] The server 605 may be a server that provides various services, such as a background management server that processes the case data generated by the user using the terminal devices 601, 602, and 603. The background management server may establish a frequent pattern tree, mine disordered frequent event sets, mine ordered frequent event sequences, etc. based on the case data, and feed back the processing results (such as the mined disordered frequent event sets and ordered frequent event sequences) to the terminal device.

[0153] It should be noted that the method for obtaining the relevance of matters provided in the embodiment of the present invention is generally executed by the server 605 , and accordingly, the device for obtaining the relevance of matters is generally arranged in the server 605 .

[0154] It should be understood that Figure 6 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0155] According to an embodiment of the present invention, the present invention also provides an electronic device, a computer-readable medium and a computer program product.

[0156] The electronic device of the present invention comprises: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement a method for obtaining the relevance of matters according to an embodiment of the present invention.

[0157] The computer-readable medium of the present invention stores a computer program thereon, and when the program is executed by a processor, a method for obtaining the relevance of matters in an embodiment of the present invention is implemented.

[0158] The computer program product of the present invention includes a computer program, and when the program is executed by a processor, a method for obtaining the relevance of matters in an embodiment of the present invention is implemented.

[0159] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 of an electronic device suitable for implementing an embodiment of the present invention. Figure 7 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0160] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0161] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0162] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0163] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0164] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0165] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The modules described may also be set in a processor. For example, they may be described as: a processor including a screening and sorting module, a statistics acquisition module, a linked list establishment module, a first mining module, and a second mining module. The names of these modules do not constitute limitations on the modules themselves in certain cases. For example, the screening and sorting module may also be described as "a module that calculates the first support of the items contained in multiple items data, screens and sorts the multiple items according to the first support and the set first screening condition, and obtains the screening and sorting results".

[0166] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The computer-readable medium carries one or more programs. When the one or more programs are executed by a device, the device includes: calculating the first support of the items contained in the multiple items data, filtering and sorting the multiple items according to the first support and the set first screening condition, and obtaining the screening and sorting results; obtaining the application materials and corresponding processing results of the multiple items, and counting the number of occurrences of the processing results of each item in the multiple items in the application materials of other items; obtaining the central item from the multiple items according to the number of occurrences and the set second screening condition, so as to establish a frequent item header table according to the central item and the screening and sorting results; establishing a frequent pattern tree according to the frequent item header table and the data item set corresponding to the item data, and mining an unordered frequent item set from the frequent pattern tree; wherein the data item set includes the filtered items, or includes the filtered items and the central item; determining a superset of the unordered frequent item set from the multiple data item sets, and extracting an ordered frequent item sequence in the superset based on a pattern mining algorithm.

[0167] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for obtaining matter relevance, characterized in that: include: Calculate the first support of the items included in the multiple items data, filter and sort the multiple items according to the first support and the set first screening condition, and obtain the screening and sorting result; the first screening condition is used to filter out the items whose first support is greater than or equal to the set first support threshold from the multiple items; the first support is calculated based on the number of times the items appear in the multiple items data, and the total number of times all the items appear in the multiple items data; Obtaining application materials and corresponding handling results of multiple handling items, and counting the number of occurrences of the handling result of each of the multiple handling items in the application materials of other handling items; According to the occurrence number and the set second screening condition, a central item is obtained from the plurality of the work items, so as to establish a frequent item header table according to the central item and the screening and sorting results; The second screening condition is used to screen out the items whose third support is greater than or equal to the set third support threshold from the plurality of items; the third support is calculated based on the number of occurrences and the number of items; According to the frequent item header table and the data item set corresponding to the handling data, a frequent pattern tree is established, and an unordered frequent item set is mined from the frequent pattern tree; wherein the data item set includes the filtered handling items, or includes the filtered handling items and the central items; Determine a superset of the unordered frequent event set from multiple data item sets, and extract the ordered frequent event sequence in the superset based on the pattern mining algorithm, including: searching for prefixes of length 1 and corresponding projections in the superset, and calculating the second support of the prefixes of length 1 according to the projections; screening the superset according to the second support to obtain corresponding screening results and frequent sequences corresponding to the prefixes of length 1; repeat the following steps until no mining can be done: recursively mine the frequent sequences corresponding to the prefixes of the current length to obtain the frequent sequences corresponding to the prefixes of the next length, and increment the current length; wherein the initial value of the current length is 1.

2. The method according to claim 1, characterized in that The frequent item header table includes the filtered handling items and the central items; The mining of unordered frequent item sets from the frequent pattern tree comprises: Find the conditional pattern base corresponding to the work items in the frequent item header table from the bottom item of the frequent item header table upward in sequence; The frequent pattern tree is recursively mined using the conditional pattern base to obtain an unordered frequent item set.

3. The method according to claim 1, characterized in that The step of establishing a frequent pattern tree according to the frequent item header table and the data item set corresponding to the case data includes: Read the data item sets corresponding to the plurality of pieces of work data, and process each time a data item set is read according to the following steps: Inserting the read tasks of the data item set as nodes into the frequent pattern tree according to the sorted order of the tasks, and updating the count of the nodes; A link relationship between the items in the frequent item header table and the corresponding nodes is established.

4. The method according to claim 2, characterized in that: The method of recursively mining the frequent pattern tree using the conditional pattern base to obtain an unordered frequent item set includes: Recursively mining the frequent pattern tree using the conditional pattern base to obtain an initial frequent item set; Based on the set promotion rate condition requirement, the initial frequent item set is screened to obtain an unordered frequent item set.

5. The method according to claim 1, characterized in that The step of determining a superset of the unordered frequent item set from the plurality of data item sets comprises: Check whether the unordered frequent item set is a subset of the data item set; If it is determined according to the test result that the unordered frequent event set is a subset of the data item set, the data item set is taken as a superset of the unordered frequent event set.

6. The method according to any one of claims 1 to 5, characterized in that: The step of establishing a frequent item header table according to the central event and the screening and sorting results includes: Maintaining the items in the screening and sorting results into an initialized frequent item header table in descending order of the first support; Determining that the frequent item header table does not include the central item, and adding the central item to the frequent item header table; The method also includes: counting the number of times the handling items appear in the multiple handling data, the number of the handling items, and the total number of times all the handling items appear in the multiple handling data.

7. A device for obtaining item relevance, characterized in that: include: A screening and sorting module is used to calculate the first support of the items included in the multiple items data, and screen and sort the multiple items according to the first support and a set first screening condition to obtain a screening and sorting result; the first screening condition is used to screen out items whose first support is greater than or equal to a set first support threshold from the multiple items; the first support is calculated based on the number of times the item appears in the multiple items data, and the total number of times all the items appear in the multiple items data; An acquisition statistics module is used to acquire application materials and corresponding handling results of a plurality of the handling items, and to count the number of occurrences of the handling result of each of the plurality of the handling items in the application materials of other handling items; A linked list building module, used to obtain a central item from the plurality of items according to the number of occurrences and the set second screening condition, so as to build a frequent item header table according to the central item and the screening and sorting results; The second screening condition is used to screen out the items whose third support is greater than or equal to the set third support threshold from the plurality of items; the third support is calculated based on the number of occurrences and the number of items; A first mining module is used to establish a frequent pattern tree according to the frequent item header table and the data item set corresponding to the work item data, and mine the disordered frequent item set from the frequent pattern tree; wherein the data item set includes the screened work item items, or includes the screened work item items and the central item; The second mining module is used to determine a superset of the unordered frequent event set from multiple data item sets, and extract an ordered frequent event sequence in the superset based on a pattern mining algorithm, including: searching for prefixes of length 1 and corresponding projections in the superset, and calculating the second support of the prefix of length 1 according to the projection; screening the superset according to the second support to obtain corresponding screening results and frequent sequences corresponding to the prefix of length 1; repeating the following steps until no mining can be done: recursively mining the frequent sequence corresponding to the prefix of the current length to obtain the frequent sequence corresponding to the prefix of the next length, and incrementing the current length; wherein the initial value of the current length is 1.

8. The device according to claim 7, characterized in that The frequent item header table includes the filtered handling items and the central items; The first mining module is also used to Find the conditional pattern base corresponding to the work items in the frequent item header table from the bottom item of the frequent item header table upward in sequence; The frequent pattern tree is recursively mined using the conditional pattern base to obtain an unordered frequent item set.

9. The device according to claim 7, characterized in that The first mining module is also used to Read the data item sets corresponding to the plurality of pieces of work data, and process each time a data item set is read according to the following steps: Inserting the read tasks of the data item set as nodes into the frequent pattern tree according to the sorted order of the tasks, and updating the count of the nodes; A link relationship between the items in the frequent item header table and the corresponding nodes is established.

10. The device according to claim 8, characterized in that The first mining module is also used to Recursively mining the frequent pattern tree using the conditional pattern base to obtain an initial frequent item set; Based on the set promotion rate condition requirement, the initial frequent item set is screened to obtain an unordered frequent item set.

11. The device according to claim 7, characterized in that The second mining module is also used to Check whether the unordered frequent item set is a subset of the data item set; If it is determined according to the test result that the unordered frequent event set is a subset of the data item set, the data item set is taken as a superset of the unordered frequent event set.

12. The device according to any one of claims 7 to 11, characterized in that The linked list building module is further used to maintain the items in the screening and sorting results into the initialized frequent item header table in descending order of the first support degree; as well as Determining that the frequent item header table does not include the central item, and adding the central item to the frequent item header table; The device also includes: a statistical module, which is used to count the number of times the handling items appear in the multiple handling data, the number of the handling items, and the total number of times all the handling items appear in the multiple handling data.

13. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

14. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

15. A computer program product comprising a computer program, characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images