A data processing method, apparatus and device

Abstract

The application provides a data processing method, device and equipment, the method comprising: obtaining time series data of a target object; forming a target data set according to the time series data, the target data set comprising part of sub-sequences in a sub-sequence set formed by each element in the time series data; forming a directed topological graph according to the target data set; generating a maximal frequent item set of the time series data according to the directed topological graph; and generating target business data according to the maximal frequent item set of the time series data. According to the scheme of the application, a maximal frequency item can be mined from time series data, which provides a better basis for subsequent data analysis, improves the efficiency of data processing, and reduces the cost of data processing.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a data processing method, apparatus, and device. Background Technology

[0002] As a crucial data collection tool, mobile phones have enabled the collection of massive amounts of data from a large number of users for a considerable period. This data allows for the discovery of users' daily behaviors and the use of data mining techniques to model and analyze their daily activity patterns.

[0003] Since the concept of data mining was proposed, scholars and researchers in various related fields worldwide have shown great interest in it. Currently, data mining technology research has many different branches, among which sequence pattern mining, as a branch, has high value in practical applications. In the continuous evolution of sequence pattern mining algorithms, the following categories of algorithms have emerged: basic data mining algorithms, multi-level multidimensional mining algorithms, incremental mining algorithms, and periodic pattern mining algorithms.

[0004] Existing algorithms require searching all the data, which is computationally intensive and has a high overhead. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a data processing method, apparatus and equipment that can improve the efficiency of data mining, shorten the data processing time and improve data processing efficiency.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] An embodiment of the present invention provides a data processing method, the method comprising:

[0008] Obtain the time-series data of the target object;

[0009] Based on the time series data, a target data set is formed, which includes a portion of the subsequences in the subsequence set formed by each element in the time series data.

[0010] Based on the target data set, a directed topological graph is formed;

[0011] Based on the directed topological graph, generate the maximal frequent itemsets of the time series data;

[0012] Target business data is generated based on the maximally frequent itemsets of the time series data.

[0013] Optionally, based on the time series data, a target data set is formed, including:

[0014] Based on the time series data S = [S1, S2, ..., Sn], form a subsequence set L1 = [P1, P2, ..., Pn] for each element in the time series data. n ]; where P i Includes at least one data point in S, where 1 ≤ i ≤ n;

[0015] Based on the subset of subsequences in the subsequence set that satisfy preset conditions, the target data set L2 = [P1, P2, ..., P] is formed. m ], where m and n are both positive integers, and m <n。

[0016] Optionally, the target data set is formed based on a subset of subsequences in the subsequence set that satisfy preset conditions, including:

[0017] From the subsequence set L1 = [P1, P2, ..., P...] n In the sequence P, select a target subsequence P. k The P k When an element in P represents location information, k Each element in the array stays at its position for a time greater than a preset time threshold, where 1 ≤ i ≤ m;

[0018] The target data set L2 = [P1, P2, ..., P] is formed based on the target subsequence. m ].

[0019] Optionally, a directed topological graph is formed based on the target data set, including:

[0020] Form a vertex set V from all the elements in the target subsequences of the target data set;

[0021] The directed connection relationships of elements in each target subsequence in the target data set are used to form an edge set E;

[0022] Based on the vertex set V and the directed edge set E, the directed topological graph G is obtained.<V,E> .

[0023] Optionally, generating a maximal frequent itemset of the time series data based on the directed topological graph includes:

[0024] Based on the directed topology graph, generate a candidate set;

[0025] Based on the candidate itemset, a maximal frequent itemset of the time series data is generated.

[0026] Optionally, based on the directed topology graph, a candidate set is generated, including:

[0027] Based on the target data set, obtain a frequent itemset M1 = [P1, P2, ..., P...] with a support greater than a preset support threshold and a length of one. j The support is defined as the presence of any item P in the set of directed edges E. j The ratio of the directed edges to all directed edges, where j is a positive integer, and j <m;

[0028] Based on the set of directed edges E of the directed topological graph and each item in the frequent itemset, candidate frequent k-itemsets of length k are generated in a sequential loop.

[0029] If the last item in the candidate frequent k-items set is recorded in the directed edge of the directed topology graph, then a candidate frequent k+1 itemset of length k+1 is obtained; otherwise, no further candidate frequent itemset is generated.

[0030] Optionally, based on the candidate itemset, generating a maximally frequent itemset of the time series data includes:

[0031] If P in the generated candidate frequent k+1 itemset k+1 If the support of an item is greater than a preset support threshold, the candidate frequent k+1 itemset is marked as a maximally frequent item.

[0032] Embodiments of the present invention also provide a data processing apparatus, the apparatus comprising:

[0033] The acquisition module is used to acquire time-series data of the target object;

[0034] The processing module is configured to form a target data set based on the time series data, wherein the target data set includes a portion of the subsequences in the subsequence set formed by each element in the time series data;

[0035] Based on the target data set, a directed topology graph is formed; based on the directed topology graph, a maximal frequent itemset of the time series data is generated; based on the maximal frequent itemset of the time series data, target business data is generated.

[0036] Embodiments of the present invention also provide a data processing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the data processing method as described above.

[0037] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the data processing method as described above.

[0038] The above-described solution of the present invention has at least the following beneficial effects:

[0039] The above-mentioned solution of the present invention uses time series data to construct a directed topological graph to generate a candidate set. At the same time, it takes into account the adjacency relationship between data and the time sequence of data in the time series data, thereby reducing the amount of data processing and improving the efficiency of the data processing algorithm. Attached Figure Description

[0040] Figure 1 This is a flowchart of the data processing method provided in an embodiment of the present invention;

[0041] Figure 2 This is a directed path topology graph provided in the embodiments of the present invention;

[0042] Figure 3 This is a schematic diagram of the module block of the data processing device provided in an embodiment of the present invention. Detailed Implementation

[0043] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0044] like Figure 1 As shown, an embodiment of the present invention provides a data processing method, the method comprising:

[0045] Step 11: Obtain the time series data of the target object;

[0046] Step 12: Based on the time series data, form a target data set, the target data set including a portion of the subsequences in the subsequence set formed by each element in the time series data;

[0047] Step 13: Based on the target data set, form a directed topological graph;

[0048] Step 14: Generate the maximal frequent itemset of the time series data based on the directed topology graph;

[0049] Step 15: Generate target business data based on the maximal frequent itemsets of the time series data.

[0050] In this embodiment of the invention, the time series data is data with a chronological order, and can be data generated in different application scenarios, such as user shopping data, user location data, internet browsing records, etc. The time series data is processed to generate a target data set. The target data set is only a portion of the subsequences generated from all subsequence sets of each data element in the time series data. Based on preset filtering conditions, the target subsequence is obtained from all subsequence sets to form the target data set, achieving preliminary filtering of the original time series data, reducing the amount of data processed subsequently, and thus reducing the workload of data processing. The definition of a subsequence in this embodiment is as follows: [The text then describes a sequence...]<a,b,c> ,sequence<a,b> It is a sequence<a,b,c> subsequence, sequence<a,c> Not a sequence<a,b,c> The subsequence, because a and c are in the sequence<a,b,c> The positions in the middle are discontinuous.

[0051] The target data set contains data elements and the relationships between them. Based on the data elements and the relationships between them, a directed topological graph is constructed. Based on the directed topological graph and the generation rules, maximal frequent itemsets are obtained from the target data set. Each subset of the maximal frequent itemsets is a frequent subset. Because the topological graph is a directed topological graph, when generating maximal frequent itemsets, the generation of unnecessary candidate frequent itemsets that do not correspond can be filtered out, reducing the probability of generating useless frequent itemsets and improving the performance of the algorithm.

[0052] The method described in the above embodiments of the present invention obtains maximally frequent itemsets, and target business data can be generated based on the maximally frequent itemsets. Target business data, such as map data, target business data generated by data analysis systems of various applications, etc., is applicable to data mining and analysis in different application scenarios.

[0053] In an optional embodiment of the present invention, step 12 may include:

[0054] Step 121: Based on the time series data S = [S1, S2, ..., Sn], form a subsequence set L1 = [P1, P2, ..., Pn] for each element in the time series data. n ]; where P i Includes at least one data point in S, where 1 ≤ i ≤ n;

[0055] Step 122: Based on the subset of subsequences in the subsequence set that satisfy preset conditions, form the target data set L2 = [P1, P2, ..., P...]. m ], where m and n are both positive integers, and m <n。

[0056] In this embodiment, a subsequence set L1 = [P1, P2, ..., P] is generated based on each data element in the time series data and the relationships between the data elements. n ], where, if the data element represents the user's location information, each subsequence can be represented by P. n =[Tn, LocID n The form of a binary tuple, LocID n Tn represents the dwell time at a location point; the subsequence set includes multiple subsequences and connection paths between location points formed by the connection relationships between the location points of the multiple subsequences; in the subsequence set, target subsequences that meet the conditions are selected according to preset conditions to form a target data set, and the amount of data processed in subsequent processing is reduced by filtering through preset conditions.

[0057] Furthermore, step 122 may include:

[0058] Step 1221, from the subsequence set L1 = [P1, P2, ..., P... n In the sequence P, select a target subsequence P. k The P k When an element in P represents location information, k Each element in the array stays at its position for a time greater than a preset time threshold, where 1 ≤ i ≤ m;

[0059] Step 1222: Form the target data set L2 = [P1, P2, ..., P] based on the target subsequence. m ].

[0060] In this embodiment, in the set of subsequences, if any subsequence P i LocID at current location i If the dwell time Tn is greater than a preset time threshold, then the subsequence P i The target subsequence can be set according to the needs of the application scenario; the target data set is formed based on the target subsequence and the relationship between the target subsequences.

[0061] The target data set includes target subsequences that meet the conditions, and connection paths between the location points of the target subsequences. It should be understood that the connection path is not limited to two location points, but can also be between multiple location points. Since the target subsequences meet the preset conditions, the connection paths generated in the target data set should meet the following: if the dwell time of any target subsequence at its location point is greater than a preset time threshold, then the path should terminate, indicating that the user has reached the destination, the movement path terminates, and the next connection path is generated.

[0062] In an optional embodiment of the present invention, step 13 may include:

[0063] Step 131: Form a vertex set V from the elements of all target subsequences in the target data set;

[0064] Step 132: Form a set of directed edges E by connecting the directed connections of the elements in each target subsequence of the target data set;

[0065] Step 133: Based on the vertex set V and the directed edge set E, obtain the directed topological graph G =<V,E> .

[0066] In this embodiment, the directed connection relationship is a directed connection path between the position points of data elements in each target subsequence, that is, a directed connection path between the position points of data elements in each target subsequence formed according to the chronological order. Specifically, it is a directed connection path generated from the position point of a data element in one target subsequence to the position point of a data element in the next target subsequence in chronological order. The directed edge set E of the directed topology graph is formed based on the directed connection path. Taking the position points of two target subsequences as an example, the directed connection relationship of the position points is shown in the following formula:

[0067]

[0068] Among them, Loc link Indicates a directed connection, (L k L k+1 This indicates that two location points appear in the connecting path in the same order according to their chronological sequence. (L) k The 'k' represents the k-th position.

[0069] The set of edges with a value of 1 constitutes the set of edges in the directed topological graph, E. Clearly, according to the definition of connectivity, if Loc... link (L k L k+1 ) = 1, Loc link (L k+1 L k () is not necessarily 1.

[0070] In an optional embodiment of the present invention, step 14 may include:

[0071] Step 141: Generate a candidate set based on the directed topology graph;

[0072] Step 142: Generate a maximal frequent itemset of the time series data based on the candidate itemset.

[0073] Furthermore, step 141 may include:

[0074] Step 1411: Based on the target data set, obtain a frequent itemset M1 = [P1, P2, ..., P] with a support greater than a preset support threshold and a length of one. j The support is defined as the presence of any item P in the set of directed edges E. j The ratio of the directed edges to all directed edges, where j is a positive integer, and j <m;

[0075] Step 1412: Based on the set of directed edges E of the directed topology graph and each sub-item of the frequent item set, generate candidate frequent k-itemsets of length k in a cyclical manner; if there is a directed edge between the item record in the last sub-item of the candidate frequent k-itemsets and the sub-item record in the frequent item set in the directed topology graph, then obtain a candidate frequent k+1-itemset of length k+1; otherwise, do not generate the next candidate frequent item set.

[0076] In this embodiment, frequent itemsets can be mined from the target data set using a sequence pattern mining algorithm and according to the directed path rules in the directed topology graph. All sub-itemsets of the frequent itemsets should be frequent items. In the target data set, a frequent itemset with a support greater than a preset support threshold and a length of one is constructed. That is, in the frequent itemset, each sub-itemet contains one item, and each item is a frequent item.

[0077] Based on each subset of the frequent itemset and the directed edge rules in the directed topology graph, other candidate frequent itemsets are generated cyclically. The loop termination condition is: if the candidate frequent k itemset cannot generate a candidate frequent k+1 itemset, the algorithm terminates early. That is, when there is no corresponding directed edge in the directed topology graph for the item record in the last subset of the candidate frequent k itemset, the generation of candidate frequent itemsets stops. For example, suppose...<x,y> It is a candidate frequent binary itemset. <z>For a frequent item set, if there is a directed edge from vertex y to vertex z in a directed topological graph, then<x,y> and <z>Connecting to form candidate frequent three-itemsets<x,y,z> The directed topological graph reduces the probability of generating useless candidate frequent itemsets.

[0078] When generating the candidate frequent k+1 itemset from the candidate frequent k itemset, if any sub-item set in the frequent itemset has appeared in the candidate frequent k itemset, then that sub-item set in the frequent itemset is filtered out, and the next sub-item set in the frequent itemset is selected. By filtering out duplicate sub-itemsets, the algorithm can avoid entering an infinite loop during execution, thus avoiding unnecessary workload.

[0079] In this embodiment, the support can be normalized to the interval [0, 1], which can be represented by the following formula:

[0080]

[0081] Where Support(T) is the support, T is the candidate set, Pi is the directed edge containing the candidate set T, and P is the set of all directed edges.

[0082] Furthermore, step 142 may include:

[0083] Step 1421, if P in the generated candidate frequent k+1 itemset k+1 If the support of an item is greater than a preset support threshold, the candidate frequent k+1 itemset is marked as a maximally frequent itemset, and the candidate frequent k itemset that generates the candidate frequent k+1 itemset is marked as a non-maximally frequent itemset.

[0084] In this embodiment, if the itemset P in the candidate frequent k+1 itemset k+1 If the support is greater than the support threshold, then P k+1 This becomes a subset of the candidate frequent k+1 itemset. Simultaneously, the candidate frequent k+1 itemset is marked as a maximal frequent itemset, while the candidate frequent k-itemset of length k that generates the candidate frequent k+1 itemset is marked as a non-maximal frequent itemset. Deleting one element from any subset of the newly generated candidate frequent k+1 itemset forms a candidate frequent itemset of length k. If this itemset appears in the candidate frequent k-itemset, it is marked as a non-maximal frequent itemset; for example,<x,y,z> It is a candidate frequent three-itemset, and the itemset is formed by removing the first element x.<y,z> If itemsets<y,z> If it is an itemset in a candidate frequent binomial set, then<y,z> It is then marked as a nonmaximal frequent itemset.

[0085] The following will use user location time-series data as an example to illustrate the method described in the above embodiments:

[0086] The location data collected by the mobile phone is a sequence A = [Loc1, Loc2, Loc3, ..., Loc...]. n ], where Loc n =[Tn, LocID n In binary form, Tn represents time, and LocID... n This represents the location. Location data needs to be preprocessed to construct the user's movement path. The definition of a movement path: A user movement path P = [Loc1, Loc2, ..., Loc...]. n [This refers to the sequential sequence of Locs corresponding to the user's location, arranged in chronological order, where Loc1 and Loc2 are...] n This represents the start and end points of path P. After preprocessing the location data, the final path set shown in Table 1 is obtained. Here, the support threshold δsup = 0.5 is set, meaning that candidate frequent itemsets must appear in at least half of the paths.

[0087] Path ID path 1 <![CDATA[[Loc2,Loc1,Loc5,Loc4]]]> 2 <![CDATA[[Loc4,Loc5,Loc2]]]> 3 <![CDATA[[Loc2,Loc1,Loc5]]]> 4 <![CDATA[[Loc5,Loc2,Loc3]]]>

[0088] Table 1 Path Set

[0089] Step 21: Construct a path topology graph based on the path set. The vertices in the topology graph are sets of locations. If locations appear consecutively on a path and satisfy a chronological order, then there is a directed edge connecting them in the graph. By scanning all paths in the path set, the generated directed path graph is as follows: Figure 2 As shown.

[0090] Step 22: Based on the definition of support, find all frequent itemsets, {Loc1, Loc2, Loc4, Loc5}, and then, based on the topological graph constraints, use L... k Itemsets and L1 itemsets generate candidate L k+1 Itemsets, such as the candidate frequent itemset {Loc1, Loc5}, can only generate two candidate frequent itemsets based on the generated directed path graph: {Loc1, Loc5, Loc2} and {Loc1, Loc5, Loc4}. Itemsets like {Loc1, Loc5, Loc3}, which cannot correspond to any path set, do not require processing. When generating candidate frequent binary itemsets from itemset Loc2, {Loc2, Loc5} cannot generate a candidate frequent itemset based on the generated directed graph, so no processing is needed, and its support is obviously zero, thus saving algorithm execution time.

[0091] Step 23: According to the definition of maximal frequent itemsets, candidate frequent three-itemsets are generated from the candidate frequent itemsets {Loc2, Loc1} and the frequent one-itemets, as shown in Step 3 of the table. Since the support of other candidate frequent itemsets is less than the threshold, only the frequent itemsets {Loc2, Loc1, Loc5} become candidate frequent itemsets. Therefore, the candidate frequent itemsets {Loc2, Loc1} and {Loc1, Loc5} are marked as non-maximum itemsets. After four execution processes, no new candidate frequent itemsets are generated, and the algorithm execution process ends prematurely. The final generated maximal frequent itemsets are {Loc5, Loc2} and {Loc1, Loc5, Loc2}. The algorithm terminates after the fourth iteration, and the results of the algorithm execution are shown in Table 2.

[0092] step Candidate frequent itemsets Pattern support 1 <![CDATA[{ <loc1> ， <loc2> ， <loc4> ， <loc5>}]]> < / loc5> < / loc4> < / loc2> < / loc1> {0.5，1，0.5，1} 2 <![CDATA[{<Loc1,Loc5>,<Loc2,Loc1>,<Loc5,Loc2>}]]> {0.5，0,5，0.5} 3 <![CDATA[{<Loc2,Loc1,Loc5>}]]> {0.5} 4 <![CDATA[{<Loc2,Loc1,Loc5,Loc4>}]]> {0.25}

[0093] Table 2 Candidate Frequent Itemsets

[0094] In this embodiment, when constructing candidate options, a directed path topology graph is constructed. When generating new candidate options, the rules of the directed path topology graph must be followed, which eliminates many candidate options that cannot correspond to user paths, reduces the algorithm's running time, and improves the efficiency of mining frequent user patterns.

[0095] like Figure 3 As shown, embodiments of the present invention also provide a data processing apparatus 30, the apparatus comprising:

[0096] Module 31 is used to acquire time series data of the target object;

[0097] Processing module 32 is configured to: form a target data set based on the time series data, the target data set including a portion of the subsequences in the subsequence set formed by each element in the time series data; form a directed topology graph based on the target data set; generate a maximal frequent itemset of the time series data based on the directed topology graph; and generate target business data based on the maximal frequent itemset of the time series data.

[0098] Optionally, the processing module 32 is used to form a target data set based on the time series data, including:

[0099] Based on the time series data S = [S1, S2, ..., Sn], form a subsequence set L1 = [P1, P2, ..., Pn] for each element in the time series data. n ]; where P i Includes at least one data point in S, where 1 ≤ i ≤ n;

[0100] Based on the subset of subsequences in the subsequence set that satisfy preset conditions, the target data set L2 = [P1, P2, ..., P] is formed. m ], where m and n are both positive integers, and m <n。

[0101] Optionally, the processing module 32 is used to form the target data set based on a subset of subsequences in the subsequence set that meet preset conditions, including:

[0102] From the subsequence set L1 = [P1, P2, ..., P...] n In the sequence P, select a target subsequence P. k The P k When an element in P represents location information, k Each element in the array stays at its position for a time greater than a preset time threshold, where 1 ≤ i ≤ m;

[0103] The target data set L2 = [P1, P2, ..., P] is formed based on the target subsequence. m ].

[0104] Optionally, the processing module 32 is configured to form a directed topological graph based on the target data set, including:

[0105] Form a vertex set V from all the elements in the target subsequences of the target data set;

[0106] The directed connection relationships of elements in each target subsequence in the target data set are used to form an edge set E;

[0107] Based on the vertex set V and the directed edge set E, the directed topological graph G is obtained.<V,E> .

[0108] Optionally, the processing module 32 is configured to generate a maximal frequent itemset of the time series data based on the directed topological graph, including:

[0109] Based on the directed topology graph, generate a candidate set;

[0110] Based on the candidate itemset, a maximal frequent itemset of the time series data is generated.

[0111] Optionally, the processing module 32 is used to generate a candidate set based on the directed topology graph, including:

[0112] Based on the target data set, obtain a frequent itemset M1 = [P1, P2, ..., P...] with a support greater than a preset support threshold and a length of one. j The support is defined as the presence of any item P in the set of directed edges E. j The ratio of the directed edges to all directed edges, where j is a positive integer, and j <m;

[0113] Based on the set of directed edges E of the directed topological graph and each item in the frequent itemset, candidate frequent k-itemsets of length k are generated in a sequential loop.

[0114] If the last item in the candidate frequent k-items set is recorded in the directed edge of the directed topology graph, then a candidate frequent k+1 itemset of length k+1 is obtained; otherwise, no further candidate frequent itemset is generated.

[0115] Optionally, the processing module 32 is configured to generate a maximally frequent itemset of the time series data based on the candidate itemset, including:

[0116] If P in the generated candidate frequent k+1 itemset k+1 If the support of an item is greater than a preset support threshold, the candidate frequent k+1 itemset is marked as a maximally frequent itemset.

[0117] It should be noted that this device is a device corresponding to the above-described data processing method. All implementations of the above method are applicable to the embodiments of this device and can achieve the same technical effect.

[0118] Embodiments of the present invention also provide a data processing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the data processing method as described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0119] Embodiments of the present invention also provide a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the data processing method as described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0120] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0121] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0122] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0123] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0124] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0125] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0126] Furthermore, it should be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Moreover, the steps performing the above-described series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This is something that those skilled in the art can achieve by using their basic programming skills after reading the description of the present invention.

[0127] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0128] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.< / z> < / z>

Claims

1. A data processing method, characterized by, The method includes: Obtain time-series data of the target object; the time-series data is data with a chronological order, and the time-series data includes user location data; wherein, the user location data is collected through a mobile phone; Based on the time series data, a target data set is formed, which includes a portion of the subsequences in the subsequence set formed by each element in the time series data. Based on the target data set, a directed topological graph is formed; Based on the directed topological graph, generate the maximal frequent itemsets of the time series data; Target business data is generated based on the maximally frequent itemsets of the time series data. The step of forming a target data set based on the time series data includes: Based on the time series data S=[S1, S2, ..., Sn], form a subsequence set L1=[P1, P2, ..., Pn] for each element in the time series data. n ]; where P i Includes at least one data point in S, where 1 ≤ i ≤ n; According to the part of the subsequence set satisfying the preset condition, the target data set L2=[P1, P2, …, P m ] is formed, wherein m, n are positive integers, and m 2. The data processing method according to claim 1, characterized in that, The target data set is formed based on a subset of subsequences from the subsequence set that satisfy preset conditions, including: From the subsequence set L1=[P1, P2, ..., P... n In the sequence P, select a target subsequence P. k The P k When an element in P represents location information, k Each element in the array stays at its position for a time greater than a preset time threshold, where 1 ≤ i ≤ m; According to the target sub-sequence, the target data set L2=[P1, P2, …, P m ].

3. The data processing method of claim 1, wherein, Based on the target data set, a directed topological graph is formed, including: Form a vertex set V from all the elements in the target subsequences of the target data set; The directed connection relationships of elements in each target subsequence in the target data set are used to form a set of directed edges E; Based on the vertex set V and the directed edge set E, the directed topological graph G is obtained.<V,E> .

4. The data processing method according to claim 3, characterized in that, Based on the directed topological graph, a maximal frequent itemset of the time series data is generated, including: Based on the directed topology graph, generate a candidate set; Based on the candidate itemset, a maximal frequent itemset of the time series data is generated.

5. The data processing method according to claim 4, characterized in that, Based on the directed topology graph, a candidate set is generated, including: Based on the target data set, obtain a frequent itemset M1=[P1, P2, ..., P] with a support greater than a preset support threshold and a length of one. j The support is defined as the presence of any item P in the set of directed edges E. j The ratio of the directed edges to all directed edges, where j is a positive integer, and j <m; Based on the set of directed edges E of the directed topological graph and each item in the frequent itemset, candidate frequent k-itemsets of length k are generated in a sequential loop. If the last item in the candidate frequent k-items set is recorded in the directed edge of the directed topology graph, then a candidate frequent k+1 itemset of length k+1 is obtained; otherwise, no further candidate frequent itemset is generated.

6. The data processing method according to claim 5, characterized in that, Based on the candidate itemset, a maximal frequent itemset of the time series data is generated, including: If the support of P k+1 items in the generated candidate frequent k+1 item set is greater than a preset support threshold, the candidate frequent k+1 item set is marked as a maximal frequent item set.

7. A data processing apparatus, characterized by, The device includes: The acquisition module is used to acquire time-series data of the target object; the time-series data is data with a chronological order, and the time-series data includes user location data; wherein, the user location data is collected through a mobile phone; The processing module is configured to form a target data set based on the time series data, wherein the target data set includes a portion of the subsequences in the subsequence set formed by each element in the time series data; Based on the target data set, a directed topology graph is formed; based on the directed topology graph, a maximal frequent itemset of the time series data is generated; based on the maximal frequent itemset of the time series data, target business data is generated. The processing module is used to form a target data set based on the time series data, including: According to time series data S=[S1, S2, …, Sn], a sub-sequence set L1=[P1, P2, …, P n ] of each element in the time series data is formed; wherein P i includes at least one data in S, 1≤i≤n; According to the part of the subsequence set satisfying the preset condition, the target data set L2=[P1, P2, …, P m ] is formed, wherein m, n are positive integers, and m 8. A data processing device, characterized by include: A processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the data processing method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, - storing instructions which, when run on a computer, cause the computer to perform the data processing method of any one of claims 1 to 6.

Images