Method and device for mining regular path in driving record

Abstract

Embodiments of the present application disclose a method and device for mining regular paths in driving records, and relate to the technical field of driving records, which can match a driving record dataset to a road network dataset, merge the driving record dataset according to a multi-scale FP-Growth algorithm, and mine regular paths corresponding to the driving record dataset. At this time, the multi-scale FP-Growth algorithm is called to mine frequent item sets in the driving record dataset, that is, frequent nodes are searched from the whole to the local, and frequent driving paths are searched and mined as the corresponding nodes are taken as a "subspace", and then all frequent driving paths are gradually mined and merged from the local to the whole, so that mining of giant frequent item sets is realized, and mining of regular paths in driving records is facilitated, the mining efficiency of regular paths in driving records is improved, and the efficient mining demand is met.

Description

Technical Field

[0001] This invention relates to the technical field of vehicle recording, and more particularly to a method and apparatus for identifying regular paths in vehicle recording. Background Technology

[0002] With the development of the information age, dashcams constantly generate massive amounts of data. Extracting valuable information from this vast amount of data requires data mining techniques. Data mining of large, dense, long-transaction datasets is particularly relevant, and dashcam data is essentially a type of dense, long-transaction dataset. Pattern path mining based on dashcam data involves extracting giant frequent itemsets. However, for giant frequent itemset mining, the FP-Growth algorithm inevitably causes a large number of function call stacks, severely consuming machine memory and failing to meet efficient mining requirements. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and apparatus for mining regular paths in driving records. This method matches a driving record dataset to a road network dataset, merges the driving record dataset using a multi-scale FP-Growth algorithm, and mines regular paths corresponding to the driving record dataset. Then, the multi-scale FP-Growth algorithm is invoked to mine frequent itemsets in the driving record dataset. Specifically, it first searches for frequent nodes from the whole to the local, and then searches and mines frequent driving paths using the corresponding nodes as "subspaces." Next, it gradually mines and merges all frequent driving paths from the local to the whole, thereby achieving the mining of giant frequent itemsets. This facilitates the mining of regular paths in driving records, improves the efficiency of regular path mining in driving records, and meets the need for high-efficiency mining.

[0004] In a first aspect, embodiments of the present invention provide a method for mining regular paths in driving records, comprising:

[0005] Obtain the dataset of driving records;

[0006] Match the driving record dataset to the road network dataset;

[0007] The multi-scale FP-Growth algorithm is used to merge the dataset of driving records and to discover the regular paths corresponding to the dataset of driving records.

[0008] According to a specific implementation of an embodiment of the present invention, obtaining the dataset of driving records includes:

[0009] Collect daily driving record data;

[0010] The daily driving record data is aggregated into a single dataset, and the driving record dataset is output.

[0011] According to a specific implementation of an embodiment of the present invention, the matching of the driving record dataset to the road network dataset includes:

[0012] Map matching is performed on daily driving records based on map data from driving records, including location data from the driving records.

[0013] The driving record sequence is matched to the road network based on the map data of the city, and the driving record dataset is matched to the road network dataset.

[0014] According to a specific implementation of an embodiment of the present invention, the step of merging the dataset of driving records according to the multi-scale FP-Growth algorithm and mining the regular paths corresponding to the dataset of driving records includes:

[0015] The first scan of the driving record dataset was performed, and it was divided into multiple type spaces;

[0016] Perform multi-scale partitioning on multiple spatial types and output the corresponding scale partitioning;

[0017] The second scan of the driving record dataset is performed, and set operations are performed sequentially to obtain a set consisting of frequent items;

[0018] Mining and set operations are performed on the set of frequent items to obtain new transaction items;

[0019] The FP-Growth algorithm is used to mine closed frequent itemsets of transaction items and to uncover regular paths in the dataset of vehicle travel records.

[0020] According to a specific implementation of an embodiment of the present invention, the first scanning of the vehicle record dataset and the division into multiple type spaces includes:

[0021] The first scan of the vehicle recording dataset is performed, and the datasets for each vehicle's records are then tagged with their types:

[0022] Obtain the data type of each vehicle's records, and analyze the spatial composition based on the data type of each vehicle's records to divide it into multiple type spaces.

[0023] According to a specific implementation of an embodiment of the present invention, the step of performing multi-scale partitioning on multiple types of space and outputting the corresponding scale partitioning includes:

[0024] Both V and W are one-dimensional continuous intervals and cover multiple types of spaces;

[0025] For a two-dimensional rectangular region: E = V × W, V and W are partitioned into n equal parts using compression mapping and composite mapping respectively, and the following is defined:

[0026]

[0027] In the formula, π m (E) is a scaled division of E.

[0028] According to a specific implementation of an embodiment of the present invention, the second scan of the vehicle record dataset, and the sequential performance of set operations to obtain a set consisting of frequent items, includes:

[0029] Transaction itemset T j (j=1,2,…,l) respectively in π m The dataset is vertically partitioned into several subsets under the constraint (E)(m=1,2,…,m0), and the resulting dataset is denoted as . At the same time, the partitioned transaction itemset Θ was also obtained. m (T j ).

[0030] According to a specific implementation of an embodiment of the present invention, the second scanning of the vehicle record dataset and the sequential performance of set operations to obtain a set composed of frequent items further include:

[0031] The value of m increases by 1 successively from 2 until m0 is reached, and the recursion is performed.

[0032] Using the partitioned transaction itemset Θ m (T j ) and Ω m-1 get

[0033] Re-statistical transaction dataset The support of each term is used to obtain the set Ω consisting of its frequent terms. m .

[0034] According to a specific implementation of an embodiment of the present invention, the step of mining the closed frequent itemsets of transaction items using the FP-Growth algorithm and mining the regular paths of the vehicle record dataset further includes:

[0035] For the space as a whole, first set each transaction item T j Represented as several consistent itemsets These consistent itemsets form new "transaction itemsets";

[0036] From the original transaction dataset {T j |j=1,2,…,l}, resulting in a new “transaction dataset”;

[0037] Then, the FP-Growth algorithm is used to mine the closed frequent itemsets in the above transaction dataset, and set operations are performed on each "transaction item" in the closed frequent itemsets to obtain the original itemsets respectively.

[0038] The algorithm uses the above-mentioned consistent itemset decomposition algorithm to decompose the data into consistent itemsets, and at the same time filters out maximal frequent itemsets. The algorithm ends after outputting the consistent itemsets and maximal frequent itemsets, and then mines the regular paths of the traffic record dataset.

[0039] Secondly, embodiments of the present invention provide a device for detecting regular paths in driving records, comprising:

[0040] The acquisition module is used to acquire the dataset of driving records;

[0041] The matching module is used to match the driving record dataset to the road network dataset;

[0042] The mining module is used to merge the driving record dataset according to the multi-scale FP-Growth algorithm and mine the regular paths corresponding to the driving record dataset.

[0043] Thirdly, embodiments of the present invention provide a device comprising: a housing, a processor, a memory, a circuit board, and a power supply circuit, wherein the circuit board is disposed within the space enclosed by the housing, and the processor and the memory are disposed on the circuit board; the power supply circuit is used to supply power to various circuits or devices of the device; the memory is used to store executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, for executing the method described in any of the foregoing implementations.

[0044] Fourthly, embodiments of the present invention also provide an application program that is executed to implement the method described in any embodiment of the present invention.

[0045] This invention provides a method, apparatus, and device for mining regular paths in driving records. It matches driving record datasets to road network datasets, merges the driving record datasets using a multi-scale FP-Growth algorithm, and mines regular paths corresponding to the driving record datasets. Then, it calls the multi-scale FP-Growth algorithm to mine frequent itemsets in the driving record datasets. Specifically, it first searches for frequent nodes from the whole to the local, and then searches and mines frequent driving paths using the corresponding nodes as "subspaces." Next, it gradually mines and merges all frequent driving paths from the local to the whole, thereby achieving the mining of giant frequent itemsets. This facilitates the mining of regular paths in driving records, improves the efficiency of regular path mining in driving records, and meets the need for high-efficiency mining. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart illustrating the method for mining regular paths in driving records according to an embodiment of the present invention.

[0048] Figure 2 for Figure 1 A flowchart of S12.

[0049] Figure 3 for Figure 1 A flowchart of S13.

[0050] Figure 4 This is a spatial distribution map of transaction items in the method for mining regular paths in driving records according to an embodiment of the present invention.

[0051] Figure 5 This is a schematic diagram of multi-scale equal partitioning of the method for mining regular paths in driving records according to an embodiment of the present invention.

[0052] Figure 6 This is a schematic diagram of the device composition for the vehicle record regular path excavation device in an embodiment of the present invention.

[0053] Figure 7 This is a hardware diagram of a device according to an exemplary embodiment. Detailed Implementation

[0054] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0055] This embodiment provides a method for mining regular paths in driving records, which solves the technical problem of low efficiency in mining regular paths in current driving records.

[0056] Figure 1 This is a flowchart illustrating the method for mining regular paths in driving records according to Embodiment 1 of the present invention. Figure 1 As shown, this embodiment of the invention provides a method for mining regular paths in driving records, the method comprising:

[0057] S11. Obtain the dataset of driving records;

[0058] S12. Match the driving record dataset to the road network dataset;

[0059] S13. Merge the driving record datasets using the multi-scale FP-Growth algorithm and mine the regular paths corresponding to the driving record datasets.

[0060] This embodiment is applied to the scenario of mining regular paths in driving records. It can match the driving record dataset to a road network dataset, merge the driving record dataset according to the multi-scale FP-Growth algorithm, and mine the regular paths corresponding to the driving record dataset. At this time, the multi-scale FP-Growth algorithm is called to mine frequent itemsets in the driving record dataset. That is, firstly, the frequent nodes are searched from the whole to the local, and the frequent driving paths are searched and mined with the corresponding nodes as "subspaces". Then, all frequent driving paths are gradually mined and merged from the local to the whole, thereby realizing the mining of giant frequent itemsets, which facilitates the mining of regular paths in driving records, improves the mining efficiency of regular paths in driving records, and meets the high-efficiency mining requirements.

[0061] The following is combined Figures 2 to 7 This invention provides a detailed description of a method for mining regular paths in driving records, the method comprising:

[0062] S11. Obtain the dataset of driving records;

[0063] This involves collecting daily driving record data; aggregating the daily driving record data into a single dataset; and outputting the driving record dataset to facilitate data processing based on the driving record dataset.

[0064] S12. Match the driving record dataset to the road network dataset;

[0065] S121, Map matching of daily driving records based on map data from driving records, including location data from driving records;

[0066] S122. Match the driving record sequence to the road network based on the map data of the city, and match the driving record dataset to the road network dataset.

[0067] At this point, the algorithm is applied to the driving record data to mine the regular paths in the driving records. Before calling the multi-scale FP-Growth algorithm to mine frequent itemsets, it is necessary to first use the map data of the driving records to perform map matching on the daily driving records. That is, by combining the location data such as latitude and longitude, timestamp, speed and heading angle of the driving records, the driving record sequence is matched to the road network using the map data of the city. This not only achieves the purpose of standardizing the data and eliminating redundancy, but also can rationally complete the driving path by associating factors such as duration, speed and distance when there is no positioning signal in tunnels.

[0068] S13. Merge the driving record datasets using the multi-scale FP-Growth algorithm and mine the regular paths corresponding to the driving record datasets.

[0069] Among them, S131 is the dataset of the first scan of the driving record, which is divided into multiple type spaces;

[0070] S132. Perform multi-scale partitioning on multiple types of space and output the corresponding scale partitioning;

[0071] S133. Scan the driving record dataset a second time and perform set operations sequentially to obtain a set consisting of frequent items;

[0072] S134. Mining and set operations are performed on the set of frequent items to obtain new transaction items;

[0073] S135. Perform FP-Growth algorithm mining on the closed frequent itemsets of the transaction items and discover the regular paths in the dataset of vehicle travel records.

[0074] The first scan of the vehicle record dataset and the division into multiple type spaces include: first scanning the vehicle record dataset and labeling the dataset of each vehicle record: obtaining the type of each vehicle record dataset and analyzing the spatial composition based on the type of each vehicle record dataset to divide into multiple type spaces, so as to facilitate subsequent scale division through multiple type spaces.

[0075] Furthermore, spatial multi-scale partitioning refers to the multi-scale equal partitioning of regular regions and the multi-scale non-equal partitioning of irregular regions. Multi-scale partitioning of regular regions includes one-dimensional continuous intervals, two-dimensional rectangular regions, and multi-dimensional cubes. Taking a two-dimensional rectangular region as an example, the multi-scale partitioning of multiple spatial types and the output of corresponding scale partitions include: V and W are both one-dimensional continuous intervals, covering multiple spatial types; for the two-dimensional rectangular region: E = V × W, V and W are partitioned into n equal parts using compression mapping and composite mapping respectively, and the following is defined:

[0076]

[0077] In the formula, π m (E) is a scaled division of E.

[0078] For any m∈N * Also known as π m (E) is a scale-based partition of E. Let's also denote π0(E) = {E} to form a multi-scale partition of a two-dimensional rectangular (i.e., regular) region. At the same time, we take two points and fill the small regions they belong to under different scale partitions with darker gray shadows.

[0079] Taking the irregular region E as an element, its corresponding set is denoted as π0(E); then taking the sub-regions obtained by the non-uniform partitioning of E as elements, their corresponding sets are denoted as π1(E); further, taking the sub-regions obtained by the above more fine-scale non-uniform partitioning as elements, their corresponding sets are denoted as π2(E); and so on, we can obtain the different scale partitions π of E. m (E)(m∈N * The subdivisions of E with different scales, such as π0(E), π1(E), and π2(E), constitute a multi-scale non-equal subdivision of E.

[0080] To construct a multi-scale spatial partition, each road segment and each node in the map is assigned a unique ID number, and then each road is mapped to its starting and ending nodes. Thus, a node may correspond to multiple road segments. These nodes are used as "subspaces" and two-dimensional regular regions composed of latitude and longitude data are divided equally at multiple scales.

[0081] In addition, the second scan of the vehicle record dataset is used to perform set operations sequentially to obtain a set consisting of frequent items, including:

[0082] Transaction itemset T j (j=1,2,…,l) respectively in π m The dataset is vertically partitioned into several subsets under the constraint (E)(m=1,2,…,m0), and the resulting dataset is denoted as . At the same time, the partitioned transaction itemset Θ was also obtained. m (T j ).

[0083] In addition, the second scan of the vehicle record dataset, and the subsequent set operations to obtain a set consisting of frequent items, also includes:

[0084] The value of m increases by 1 successively from 2 until m0 is reached, and the recursion is performed.

[0085] Using the partitioned transaction itemset Θ m(T j ) and Ω m-1 get

[0086] Re-statistical transaction dataset The support of each term is used to obtain the set Ω consisting of its frequent terms. m .

[0087] Let ^Ω m0 =Ω m0 ; Traverse all e∈Ω m0 Mining transaction datasets using the FP-Growth algorithm If a closed frequent itemset exists, then a consistent itemset can be obtained by using a consistent itemset decomposition algorithm. Otherwise, the corresponding e will be from set ^Ω m0 Delete; and thus derive ^Ω m (m = 1, 2, ..., m0-1).

[0088] The recursive process is performed by successively decreasing the value of m by 1 from m0-1 until 1: traversing all e∈^Ω m First, set each transaction item set Represented as several consistent itemsets These consistent itemsets form new "transaction itemsets," which can then be derived from the transaction dataset. A new "transaction dataset" is obtained; then, the FP-Growth algorithm is used to mine closed frequent itemsets in the above "transaction dataset," and the set union operation is performed on each "transaction item" in the itemset to obtain the original itemsets respectively; consistent itemsets are decomposed using a consistent itemset decomposition algorithm to obtain consistent itemsets.

[0089] Furthermore, the method of performing FP-Growth algorithm mining on closed frequent itemsets of transaction items and mining regular paths in the vehicle record dataset also includes:

[0090] For the space as a whole, first set each transaction item T j Represented as several consistent itemsets The union of these consistent itemsets forms a new "transaction item set";

[0091] From the original transaction dataset {T j |j=1,2,…,l}, resulting in a new “transaction dataset”;

[0092] Then, the FP-Growth algorithm is used to mine the closed frequent itemsets in the above transaction dataset, and set operations are performed on each "transaction item" in the closed frequent itemsets to obtain the original itemsets respectively.

[0093] The algorithm decomposes the data into consistent itemsets using the above-mentioned consistent itemset decomposition algorithm, and simultaneously filters out maximal frequent itemsets. The algorithm ends after outputting the consistent itemsets and maximal frequent itemsets, and then mines the regular paths in the data set of driving records.

[0094] Based on the spatial multi-scale partitioning method of the aforementioned dataset, the multi-scale FP-Growth algorithm is used to mine frequent itemsets. This involves first searching for frequent nodes from the global to the local level, and then searching and mining frequent driving paths using the corresponding nodes as "subspaces." Next, all frequent driving paths are gradually mined and merged from the local to the global level. The driving paths corresponding to the consistent itemsets finally output by the algorithm are called regular paths.

[0095] This method applies the multi-scale FP-Growth algorithm to a big data platform and traverses the platform to acquire driving record datasets. The big data platform then connects to street monitoring systems and road monitoring systems to further supplement the vehicle records and fully utilizes its computational capabilities. The multi-scale FP-Growth algorithm is used to merge the driving record datasets and mine regular paths corresponding to them. The algorithm then mines frequent itemsets in the driving record datasets, first searching for frequent nodes from the whole to the part, and then searching and mining frequent driving paths using the corresponding nodes as "subspaces." Finally, it gradually mines and merges all frequent driving paths from the part to the whole, thus achieving the mining of giant frequent itemsets. This facilitates the mining of regular paths in driving records, improves the efficiency of regular path mining, and meets the requirements for high-efficiency mining.

[0096] Furthermore, the driving record dataset is merged using the multi-scale FP-Growth algorithm, and the regular paths corresponding to the driving record dataset are mined. The merging of the driving record dataset is performed on the big data platform, which involves multi-scale partitioning of multiple type spaces and outputting the corresponding scale partitions. The driving record dataset is scanned a second time, and set operations are performed sequentially to obtain sets composed of frequent items. Mining and set operations are performed on the sets composed of frequent items to obtain new transaction items. The FP-Growth algorithm is used to mine the closed frequent itemsets of the transaction items to uncover the regular paths of the driving record dataset.

[0097] The multi-scale FP-Growth algorithm proposed in this application is based on spatial multi-scale partitioning. It progressively associates multi-scale spaces according to the partitioning and then gradually merges frequent itemsets by repeatedly calling the FP-Growth algorithm, thereby achieving the mining of giant frequent itemsets. The more frequent transaction items each transaction contains, the deeper the FP-Growth tree initially constructed from the transaction dataset becomes. Recursively calling the FP-Growth algorithm inevitably leads to a large number of function call stacks, severely consuming machine memory and failing to meet efficient mining requirements. The proposed algorithm combines the FP-Growth algorithm with spatial multi-scale partitioning to achieve the gradual merging of frequent itemsets.

[0098] like Figure 6 As shown, this embodiment of the invention also provides a device for detecting regular paths in driving records, the device comprising:

[0099] Module 21 is used to acquire the dataset of driving records;

[0100] Matching module 22 is used to match the driving record dataset to the road network dataset;

[0101] Mining module 23 is used to merge the dataset of driving records according to the multi-scale FP-Growth algorithm and mine the regular paths corresponding to the dataset of driving records.

[0102] Figure 7 This is a schematic diagram of the structure of an embodiment of the device of the present invention. To implement the above process, the device includes: a housing 41, a processor 42, a memory 43, a circuit board 44, and a power supply circuit 45. The circuit board 44 is disposed inside the space enclosed by the housing 41, and the processor 42 and the memory 43 are disposed on the circuit board 44. The power supply circuit 45 is used to supply power to the various circuits or devices of the device. The memory 43 is used to store executable program code. The processor 42 runs the program corresponding to the executable program code by reading the executable program code stored in the memory 43, and is used to execute the method for mining regular paths in driving records as described in any of the foregoing embodiments.

[0103] For details on the specific execution process of the above steps by processor 42, and the steps further executed by processor 42 through running executable program code, please refer to the present invention. Figure 1-5 The description of the illustrated embodiments will not be repeated here.

[0104] This device exists in various forms, including but not limited to:

[0105] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and primarily aim to provide voice and data communication. These terminals include: smartphones (e.g., iPhones), multimedia phones, feature phones, and low-end phones, etc.

[0106] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers, possessing computing and processing capabilities, and generally also have mobile internet access features. These terminals include PDAs, MIDs, and UMPCs, such as the iPad.

[0107] (3) Portable entertainment devices: These devices display and play multimedia content. This category includes audio and video players (such as iPods), handheld game consoles, e-books, as well as smart toys and portable car navigation devices.

[0108] (4) Server: A device that provides computing services. The components of a server include a processor, hard disk, memory, system bus, etc. Servers are similar to general computer architectures, but because they need to provide highly reliable services, they have higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0109] (5) Other devices with data interaction functions.

[0110] In an exemplary embodiment, a computer-readable storage medium is also provided, which stores one or more programs that can be executed by one or more processors to implement the aforementioned method for mining regular paths in driving records. For example, the computer-readable storage medium is a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0111] In addition, embodiments of the present invention also provide an application program that is executed to implement the methods provided in any embodiment of the present invention.

[0112] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0113] The various embodiments in this specification are described in a related manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0114] In particular, the device embodiment is basically similar to the method embodiment, so the description is relatively simple. For relevant details, please refer to the description of the method embodiment.

[0115] For ease of description, the above apparatus is described by dividing it into various functional units / modules. Of course, in implementing this invention, the functions of each unit / module are implemented in one or more software and / or hardware.

[0116] Those skilled in the art will understand that implementing all or part of the processes in the above embodiments is accomplished by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0117] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method of mining regular paths in driving records, characterized by, The method comprises the following steps: obtaining a driving record dataset; matching the driving record dataset to a road network dataset; merging the driving record dataset according to a multi-scale FP-Growth algorithm and mining a regular path corresponding to the driving record dataset; the merging the driving record dataset according to the multi-scale FP-Growth algorithm and mining the regular path corresponding to the driving record dataset comprises: firstly scanning the driving record dataset and dividing a plurality of type spaces; multi-scale partitioning the plurality of type spaces and outputting corresponding scale partitions; secondly scanning the driving record dataset and sequentially performing set operations to obtain a set composed of frequent items; mining and performing set operations according to the set composed of frequent items to obtain new transaction items; performing FP-Growth algorithm mining on a closed frequent item set of the transaction items and mining the regular path of the driving record dataset; the first scanning of the driving record dataset and the division of the plurality of type spaces comprises: firstly scanning the driving record dataset and marking the type of each driving record dataset; obtaining the type of each driving record dataset and analyzing the spatial composition according to the type of each driving record dataset to divide the plurality of type spaces; the multi-scale partitioning of the plurality of type spaces and the outputting of the corresponding scale partitions comprise: V and W are both one-dimensional continuous intervals and cover the plurality of type spaces; for a two-dimensional rectangular region: E = V x W, V and W are n-equal-partitioned by using compression mapping and composite mapping respectively, and the following definitions are made: where π m (E) is a bisecting section of a dimension of E.

2. The method of claim 1, wherein, the obtaining of the driving record dataset comprises: collecting daily driving record data; collecting the daily driving record data into the same dataset and outputting the driving record dataset.

3. The method of claim 1, wherein, the matching of the driving record dataset to the road network dataset comprises: performing map matching on the daily driving record based on map data of the driving record, wherein the positioning data in the driving record is comprehensively used; matching the driving record sequence to the road network according to map data of the city where the driving record is located and matching the driving record dataset to the road network dataset.

4. The method of claim 1, wherein, the second scanning of the driving record dataset and the sequentially performing of set operations to obtain the set composed of frequent items comprises: Transaction item set T j are divided into several subsets vertically under the constraint of π m (E) respectively, and the data set composed of them is denoted as Meanwhile, the partitioned transaction item set Θ m (T j ) is obtained.

5. The method of claim 4, wherein, the second scanning of the driving record dataset and the sequentially performing of set operations to obtain the set composed of frequent items further comprises: the value of m is increased by 1 from 2 until m0 is executed recursively; using the divided transaction item set Θ m (T j ) and Ω m-1 get Re-statistics transaction dataset support of each item, obtaining a set Ω composed of frequent items m .

6. The method of claim 5, wherein, the FP-Growth algorithm mining on the new transaction items and the mining of the regular path of the driving record dataset further comprise: For the space as a whole, each transaction item set T j is represented as a number of consistent item sets The new "transaction item sets" are formed with these consistent item sets as new "transaction items"; from the original transaction dataset {T j | j = 1,2,...,l}, a new "transaction dataset"; the FP-Growth algorithm mining on the closed frequent item set in the above transaction dataset and the set operation on each "transaction item" in the closed frequent item set are performed to obtain the original item set respectively; the consistent item set decomposition algorithm is used for decomposition to obtain a consistent item set, and a maximal frequent item set is screened out, and the algorithm is ended after the consistent item set and the maximal frequent item set are output, and the regular path of the driving record dataset is mined.

7. A device for excavating regular paths in a driving record, characterized in that The method comprises the following steps: an obtaining module is configured to obtain a driving record dataset; The matching module is configured to match the driving record data set to a road network data set; The mining module is configured to merge the driving record data set according to a multi-scale FP-Growth algorithm, and to mine regular paths corresponding to the driving record data set; The merging the driving record data set according to the multi-scale FP-Growth algorithm and mining the regular paths corresponding to the driving record data set comprises: The driving record data set is scanned for the first time, and a plurality of type spaces are divided; The plurality of type spaces are multi-scale dissected, and corresponding scale dissections are outputted; The driving record data set is scanned for the second time, and set operations are sequentially performed to obtain a set composed of frequent items; Mining and set operations are performed according to the set composed of frequent items to obtain new transaction items; FP-Growth algorithm mining is performed on a closed frequent item set of the transaction items, and regular paths of the driving record data set are mined; The scanning the driving record data set for the first time and dividing the plurality of type spaces comprises: The driving record data set is scanned for the first time, and each driving record data set is type marked; Types of each driving record data set are obtained, and space compositions are analyzed according to the types of each driving record data set to divide the plurality of type spaces; The multi-scale dissecting the plurality of type spaces and outputting corresponding scale dissections comprises: V and W are both one-dimensional continuous intervals, and cover the plurality of type spaces; For a two-dimensional rectangular region: E = V × W, V and W are n-equal dissectioned by using compression mapping and composite mapping respectively, and are defined as: where π m (E) is a bisecting section of a dimension of E.

Images