An intelligent connected vehicle data classification method supporting hierarchical adjustment

Abstract

The present invention provides an intelligent networked vehicle data grading method supporting hierarchical adjustment, and the method includes: obtaining data to be graded and standard data at the highest level; using the standard data as target prediction data, and calculating the inference strength between the target prediction data and the data to be graded; sorting the inference strength to obtain a gradient sequence; and determining the level of the data to be graded according to the gradient sequence, a preset number of grading levels, and the preset data capacity that can be accommodated at each level. This solution realizes the quantitative grading of intelligent networked vehicle data, and improves the accuracy and effectiveness of data security management.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive information security, and particularly relates to a method for grading intelligent connected vehicle data that supports hierarchical adjustment. Background Art

[0002] With the acceleration of digital transformation, data security has become an increasingly prominent issue. At present, the management and protection of data security often adopt a hierarchical protection strategy, but mostly rely on expert review and qualitative analysis. Although this method can perform reasonable risk assessment based on experience, it often lacks sufficient quantitative analysis ability, resulting in inaccurate division of data security levels and difficulty in adapting to the rapidly changing security threat environment. In particular, there are complex correlations among data in the field of intelligent connected vehicles. Traditional data grading methods usually analyze and grade individual data items, ignoring the interaction between data items, and thus cannot fully reveal the potential connections between data items, which may lead to high-level sensitive data being cracked by lower-level data through logical reasoning, thereby bringing security risks. Summary of the Invention

[0003] The present invention provides a method for grading intelligent connected vehicle data that supports hierarchical adjustment, realizing the quantitative grading of intelligent network vehicle data and improving the accuracy and effectiveness of data security management.

[0004] In a first aspect, the present invention provides a method for grading intelligent connected vehicle data that supports hierarchical adjustment, including:

[0005] Obtain the data to be graded and the standard data at the highest level;

[0006] Use the standard data as the target prediction data, and calculate the inference strength between the target prediction data and the data to be graded;

[0007] Sort the inference strength to obtain a gradient sequence;

[0008] Determine the level of the data to be graded according to the gradient sequence, the preset number of grades, and the preset data volume that can be accommodated at each level.

[0009] Optionally, the step of using the standard data as the target prediction data and calculating the inference strength between the target prediction data and the data to be graded includes:

[0010] For each target prediction data, perform the following:

[0011] S21: Determine whether the data type of the data to be classified is the same as the data type of the target prediction data, and determine whether there is a target rule set corresponding to the data type in the preset rule base; if the judgment results are both yes, execute step S22, otherwise execute step S25;

[0012] S22: The target rule set includes multiple prediction rules, and output the prediction accuracy of each prediction rule within a specified time according to the preset inference accuracy threshold;

[0013] S23: Determine the maximum prediction accuracy among the prediction accuracies corresponding to all the prediction rules as the prediction accuracy of the data to be classified;

[0014] S24: Obtain the risk probability of the data to be classified, and use the product of the risk probability and the prediction accuracy as the inference intensity;

[0015] S25: Determine that the inference intensity is 0.

[0016] Optionally, the determining the level of the data to be classified according to the gradient sequence, the preset number of levels, and the preset data capacity of each level includes:

[0017] S31: Determine whether the current data volume in the highest level is less than the preset data capacity of the highest level; if the judgment result is yes, execute step S32, otherwise execute step S34;

[0018] S32: Determine the data to be classified corresponding to the highest inference intensity from the gradient sequence, and use this data to be classified as the first standard data and classify it into the highest level;

[0019] S33: Use the first standard data as the target prediction data, calculate the inference intensity between the target prediction data and the data to be classified, sort the inference intensity to obtain a gradient sequence, and return to step S31;

[0020] S34: If the judgment result is no, then determine the same number of the inference intensities as the preset data capacity of the sub - highest level from high to low in the gradient sequence, and classify the data to be classified corresponding to this inference intensity into the preset sub - highest level to complete the classification of the preset sub - highest level.

[0021] Optionally, after completing the classification of the preset sub - highest level, it further includes:

[0022] S41: Use the latest completed preset level as the current preset level, determine whether there is data to be classified that has not been classified into the current preset level, if there is, execute step S42, otherwise execute S45;

[0023] S42: Use the current level data in the current preset level as the target prediction data, and calculate the inference strength between the target prediction data and the data to be classified;

[0024] S43: Sort the inference strength to obtain a gradient sequence;

[0025] S44: Determine, from high to low in the gradient sequence, the same number of inference strengths as the accommodation data volume of the preset next level, classify the data to be classified corresponding to the inference strength into the preset next level, and return to step S41; wherein, the preset next level is the next level of the current preset level;

[0026] S45: Complete the complete classification of the data to be classified.

[0027] Optionally, the data to be classified includes single data to be classified and / or a data group to be classified obtained by combining at least two of the single data to be classified;

[0028] When the data to be classified is the data group to be classified, if the data group to be classified has been classified into a known level, each of the single data to be classified included in the data group to be classified is also classified into the known level.

[0029] Optionally, the data to be classified includes single data to be classified and / or a data group to be classified obtained by combining at least two of the single data to be classified;

[0030] When the data to be classified is the data group to be classified and the single data to be classified, obtain the first known level where the data group to be classified is located and the second known levels where each of the single data to be classified included in the data group to be classified is located;

[0031] Determine whether there is a situation where the first known level is higher than the second known level;

[0032] If the judgment result is yes, classify the single data to be classified corresponding to the second known level into the first known level.

[0033] Optionally, after determining the level of the data to be classified according to the gradient sequence, the preset classification quantity, and the accommodation data volume of each preset level, it further includes:

[0034] When the level where the target data to be classified in the data to be classified is higher than the preset expected level, obtain the data accuracy and data density of the data to be classified, and reduce the data accuracy and / or the data density of the data to be classified to obtain new target data to be classified, and classify the new target data to be classified.

[0035] Optionally, the grading of the new target data to be graded includes:

[0036] S71: Obtain the current level of the target data to be graded, and sequentially supplement the data at the next level of the current level into the current level;

[0037] S72: Use the new target data to be graded and other data to be graded except those classified into the current level and the level above the current level as new data to be graded;

[0038] S73: Determine whether there is new data to be graded that has not been classified into the current level. If so, execute step S74; otherwise, execute step S77;

[0039] S74: Use the current level data in the current level as target prediction data, and calculate the inference strength between the target prediction data and the new data to be graded;

[0040] S75: Sort the inference strength to obtain a gradient sequence;

[0041] S76: Determine the same number of inference strengths as the accommodation data volume of the next level from high to low in the gradient sequence, classify the new data to be graded corresponding to the inference strength into the next level, and use the next level as the current level and return to step S73; where the next level is the next level of the current level;

[0042] S77: Complete the complete grading of the new data to be graded.

[0043] In a second aspect, the present invention provides an intelligent connected vehicle data grading device supporting grading adjustment, including:

[0044] An acquisition module, configured to acquire data to be graded and standard data at the highest level;

[0045] An operation module, configured to use the standard data as target prediction data, calculate the inference strength between the target prediction data and the data to be graded, and sort the inference strength to obtain a gradient sequence;

[0046] A grading module, configured to determine the level of the data to be graded according to the gradient sequence, a preset grading quantity, and the accommodation data volume of each preset level.

[0047] In a third aspect, an embodiment of the present invention further provides a computing device, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the method described in any first aspect of this specification is implemented.

[0048] Fourthly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed on a computer, the computer is made to execute the method described in any first aspect of this specification.

[0049] Fifthly, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method described in any first aspect of this specification are implemented.

[0050] An embodiment of the present invention provides an intelligent connected vehicle data grading method supporting hierarchical adjustment, which acquires data to be graded and standard data at the highest level, uses the standard data as target prediction data, calculates the inference strength between the target prediction data and the data to be graded, then sorts the inference strength to obtain a gradient sequence, and finally determines the level of the data to be graded based on the gradient sequence, the preset number of grading levels, and the preset data capacity of each level. In this way, based on the inference strength between the target prediction data and the data to be graded, the quantitative grading of intelligent network vehicle data is realized, and the accuracy and effectiveness of data security management are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0052] Figure 1 is a flowchart of an intelligent connected vehicle data grading method supporting hierarchical adjustment provided by an embodiment of the present invention;

[0053] Figure 2 is a hardware architecture diagram of a computing device provided by an embodiment of the present invention;

[0054] Figure 3 is a structural schematic diagram of an intelligent connected vehicle data grading device supporting hierarchical adjustment provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0056] The specific implementation manners of the concept of the present application are described below.

[0057] Please refer to Figure 1 , an embodiment of the present invention provides an intelligent connected vehicle data grading method supporting hierarchical adjustment, and the method includes:

[0058] Step 100, obtain the data to be graded and the standard data at the highest level;

[0059] Step 102, use the standard data as the target prediction data, and calculate the inference strength between the target prediction data and the data to be graded;

[0060] Step 104, sort the inference strength to obtain a gradient sequence;

[0061] Step 106, determine the level of the data to be graded according to the gradient sequence, the preset number of grades, and the preset data capacity of each level.

[0062] In the embodiment of the present invention, based on the obtained data to be graded and the standard data at the highest level, and using the standard data as the target prediction data, calculate the inference strength between the target prediction data and the data to be graded, then sort the inference strength to obtain a gradient sequence, and then determine the level of the data to be graded from the gradient sequence, the preset number of grades, and the preset data capacity of each level. In this way, based on the inference strength between the target prediction data and the data to be graded, the quantitative grading of the intelligent network vehicle data is realized, and the accuracy and effectiveness of data security management are improved.

[0063] The execution manners of the following Figure 1 shown steps are described.

[0064] First, in step 100, the marked graded data is the graded data determined based on documents such as national security regulations and enterprise internal regulations, and the data at the highest level in the data grading is used as the standard data. For example, if level 1 is taken as the highest level, then the standard data D Lv1 ={D Lv1,1 D Lv1,2 D Lv1,3 ...D Lv1,m}.

[0065] In a preferred embodiment, the data to be classified includes single data to be classified and / or a data group to be classified obtained by combining at least two single data to be classified. It should be noted that the data to be classified can all be single classified data, or all be data groups to be classified, or be single data to be classified and data groups to be classified. Specifically, single data to be classified is distinct from a data group to be classified. For example, single data to be classified includes independent single data such as vehicle pedal data and in-vehicle button data, and each data is independently quantified and classified during classification; a data group to be classified can be a data set corresponding to a vehicle speed, that is, the data group to be classified is combined data of vehicle pedal data and in-vehicle button data, and the entire data group to be classified is classified during classification.

[0066] For step 102, taking the standard data as the target prediction data, calculating the inference strength between the target prediction data and the data to be classified includes:

[0067] For each target prediction data, the following operations are all executed:

[0068] S21: Determine whether the data type of the data to be classified is the same as the data type of this target prediction data, and determine whether there is a target rule set corresponding to the data type in the preset rule library; if both judgment results are yes, execute step S22, otherwise execute step S25;

[0069] S22: The target rule set includes multiple prediction rules, and the prediction accuracy of each prediction rule within a specified time is output according to the preset inference accuracy threshold;

[0070] S23: Determine the maximum prediction accuracy from the prediction accuracies corresponding to all prediction rules as the prediction accuracy of the data to be classified;

[0071] S24: Obtain the risk probability of the data to be classified, and take the product of the risk probability and the prediction accuracy as the inference strength;

[0072] S25: Determine that the inference strength is 0.

[0073] It should be noted that the preset rule library stores rule sets corresponding to different data types, where the data types include numerical types, radar point clouds, video types, audio types, and text types; each rule set contains one or more prediction rules. Specifically, the rule sets corresponding to different data groups to be classified are also different. Here, a prediction rule refers to prediction data obtained within the same specified time. By comparing the deviation between the prediction data and the actual data at the same moment, the probability that the deviation meets the preset inference accuracy threshold within the specified time is the prediction accuracy.

[0074] Specifically, continuing with the previous example, the target prediction data D Lv1 ={D Lv1,1 DLv1,2 D Lv1,3 ...D Lv1,m}, the data to be classified is a single piece of data to be classified D db,n ={D 1 D 2 D 3 ...,D n}, for each target prediction data among them, taking D Lv1,1 as an example, the following operations are all performed: In the first loop, determine whether the data types of D 1 and D Lv1,1 are the same, and determine whether there is a target rule set corresponding to this data type in the preset rule library. The two results after determination are respectively: 1) If both judgment results are yes, determine the target rule set from the preset rules, and then output the prediction accuracies of each prediction rule (from r1 to rn) in the target rule set within the specified time according to the preset inference accuracy threshold Then take the maximum value among them as the prediction accuracy of D 1 for D Lv1,1 . Then calculate the risk probability of the data to be classified D 1 based on the TARA method in the field of information security (risk probability = probability of leakage event * probability of attack occurrence), and take the product of the risk probability and the prediction accuracy as the inference strength between D 1 and D Lv1,1 . 2) If one or both judgment results are no, determine that the inference strength between D 1 and D Lv1,1 is 0. Then repeat the above loop to calculate the inference strengths between D 2 , D 3 , …… D n and D Lv1,1 in turn.

[0075] Then, following the previous example, successively take D Lv1,2 , D Lv1,3 , …… D Lv1,m as the target prediction data and continue to repeat the above loop to obtain the inference strength between each target prediction data and each data to be classified.

[0076] It should be noted that the preset number of classification levels and the data capacity that can be accommodated at each preset level are designed by the user according to the actual application. Among them, the protection schemes corresponding to different levels are different. For example, the data capacity that can be accommodated at each level can be determined by optimizing the problem, and its constraint conditions are: the amount of resources that can be processed by the processor per unit time; the number of resources consumed by each data at each level; the total data capacity of each level should be greater than the number of data items to be classified. Its optimization goal is: the data at a higher level should be as many as possible, max Q lv1 , max Q lv2 (when max Qlv1 At this time). Its decision variables are: the amount of data configured at each level {Q lv1 , Q lv2 , Q lv3}. It should be noted that the data capacity of each level includes but is not limited to being calculated only by the above method.

[0077] Regarding step 104, sorting the inference intensities to obtain a gradient sequence means sorting all the inference intensities between each target prediction data of the highest level obtained and each data to be classified from high to low.

[0078] In step 106, determining the level of the data to be classified according to the gradient sequence, the preset number of classification levels, and the preset data capacity of each level includes:

[0079] S31: Determine whether the current data volume in the highest level is less than the preset data capacity of the highest level; if the judgment result is yes, execute step S32, otherwise execute step S34;

[0080] S32: Determine the data to be classified corresponding to the highest inference intensity from the gradient sequence, and use this data to be classified as the first standard data and classify it into the highest level;

[0081] S33: Use the first standard data as the target prediction data, calculate the inference intensity between the target prediction data and the data to be classified, sort the inference intensities to obtain a gradient sequence, and return to step S31;

[0082] S34: If the judgment result is no, then determine from the gradient sequence the same number of inference intensities as the preset data capacity of the second-highest level from high to low, and classify the data to be classified corresponding to this inference intensity into the preset second-highest level to complete the classification of the preset second-highest level.

[0083] It should be noted that the standard data of the highest level is stipulated by regulations, that is, the data in the highest level is not empty before classification, and the current data volume in the highest level is actually the number of standard data. Therefore, when the preset data capacity of the highest level is greater than this current data volume, the highest level is not full, and it is necessary to determine the data that can be classified into the highest level from the data to be classified.

[0084] Specifically, for example, the current data volume is 2, the preset highest level is the first level with a data volume capacity of 5, the preset second-highest level is the second level with a data volume capacity of 5. First, it is necessary to determine the data to be classified corresponding to the highest inference strength from the gradient sequence, and use this data to be classified as the first standard data and classify it into the highest level. Then, calculate the inference strength between the first standard data and the data to be classified, and then update the gradient sequence (i.e., reorder all the current inference strengths). At this time, return to S31, and the current data volume is 3; repeat the above steps S31 - S33 until the data volume of the first level reaches 5, filling up this highest level. Then, for the most recently updated gradient sequence, select the top 5 inference strength corresponding data to be classified from high to low in this gradient sequence, and classify these 5 data to be classified into the second level.

[0085] In the present invention, classifying the data to be classified corresponding to the highest value among all the calculated inference strengths each time into the highest level can further ensure the accuracy of the data selected each time through the correlation between the data, and avoid potential safety hazards caused by some data being classified into lower levels. At the same time, the data that can best predict the highest level is classified into the second-highest level, making the corresponding data security level higher and reducing the security risk.

[0086] In a preferred embodiment, after completing the classification of the preset second-highest level, it further includes:

[0087] S41: Take the latest completed classified preset level as the current preset level, and determine whether there is data to be classified that has not been classified into the current preset level. If there is, execute step S42; otherwise, execute S45;

[0088] S42: Take the current level data in the current preset level as the target prediction data, and calculate the inference strength between the target prediction data and the data to be classified;

[0089] S43: Sort this inference strength to obtain a gradient sequence;

[0090] S44: Determine the same number of inference strengths as the data volume capacity of the preset next level from high to low in this gradient sequence, and classify the data to be classified corresponding to this inference strength into the preset next level, and return to step S41; where the preset next level is the next level of the current preset level;

[0091] S45: Complete the complete classification of the data to be classified.

[0092] Specifically, for example, if the preset number of levels is 5, the levels are, from highest to lowest, the first level, the second level, the third level, the fourth level, and the fifth level, and the data capacity of each level is preset respectively. After determining the data classified into the first level and the second level according to the foregoing steps S31 to S34, take the second level as the current preset level, and determine whether there is any data to be classified that has not been classified into the second level. If there is still data to be classified that has not been classified into the second level, then use the data in the second level as the target prediction data, and then calculate the inference strength between the remaining data to be classified and the target prediction data, and sort the inference strength from highest to lowest to obtain a gradient sequence. Select from the gradient sequence the same number of inference strengths as the data capacity of the third level from highest to lowest, and classify the corresponding data to be classified into the third level; then take the third level as the current preset level, and determine whether there is any data that has not been classified into the third level. If so, continue to repeat steps S41 to S45. If there is no data to be classified that has not been classified into the second level, then all the data to be classified have been completely classified.

[0093] It should be noted that after the data to be classified is determined to be classified into a certain level, it becomes data of a known level, and at the same time, this data is correspondingly removed from the gradient sequence.

[0094] In the present invention, by using the data of the previous level as the target prediction data and further calculating the inference strength between the target prediction data and the remaining data to be classified, the analysis of the correlation between data is realized, avoiding the sensitive data of the higher level being cracked by the data of the lower level through logical reasoning, making the data classification more scientific, and improving the accuracy and effectiveness of data security management.

[0095] In some preferred embodiments, due to the different compositions of the data to be classified, the specific classification results are also different, specifically including the following two cases:

[0096] The first case: The data to be classified is a single piece of data to be classified, and the data to be classified is classified in turn according to the above steps.

[0097] The second case: The data to be classified is a group of data to be classified. If the group of data to be classified has been classified into a known level, then each single piece of data to be classified included in the group of data to be classified is also classified into the known level. For example, if a group of data to be classified including vehicle pedal data and in-vehicle button data is classified into the first level, then the single vehicle pedal data and the single in-vehicle button data are also both classified into the first level.

[0098] The third case: When the data to be classified is a data group to be classified and a single data to be classified, obtain the first known level where the data group to be classified is located and the second known levels where each single data to be classified included in the data group to be classified is located; determine whether there is a situation where the first known level is higher than the second known level. If the determination result is yes, classify the single data to be classified corresponding to the second known level into the first known level. For example, the data to be classified is a single automotive pedal data and a data group to be classified including automotive pedal data and in-vehicle button data. The single automotive pedal data is classified into the second level, the data group to be classified is classified into the first level, and the first level is higher than the second level. At this time, the final classification result of the data to be classified is that both the single automotive pedal data and the data group to be classified are classified into the first level. It should be noted here that the preset data capacity of each level is a standard value, and there is actually a margin in the capacity of each level, that is, the maximum data capacity is greater than the standard value. Therefore, even if the preset data capacity of the level is full as described above, the level can still continue to accommodate data.

[0099] After step 106, it further includes: when the level where the target data to be classified in the data to be classified is higher than the preset expected level, obtain the data precision and data density of the data to be classified, and reduce the data precision and / or data density of the data to be classified to obtain new target data to be classified, and classify the new target data to be classified.

[0100] For example, the precision of numerical values in the data to be classified is 0.001 m / s, and the update frequency (i.e., data density) is 100 Hz. After reducing its data precision, it can be 0.01 m / s, 0.1 m / s, 1 m / s, or 10 m / s. After reducing its data density, it can be 10 Hz or 1 Hz. For data of the radar point cloud type, the data precision is the angular resolution, and the data density is the update frequency; for video data, the data precision is the resolution, and the data density is the video frame rate; for audio data, the data precision is the sampling depth, and the data density is the sampling rate.

[0101] In the present invention, after the classification is completed, if the level where the target data to be classified is higher than the preset expected level, the target data to be classified can be processed by three methods: reducing data precision, reducing data density, and reducing both data precision and density, to obtain new target data to be classified. Since the new target data to be classified needs to be re-classified, and this data will be classified into any level other than the current level of the target data to be classified, all data other than the current level where the target data to be classified is located needs to be re-classified. Therefore, the above classification method of steps S41 - S45 needs to be executed for both the new target data to be classified and the data set that has been currently known to be classified into the next level of the target data to be classified, to achieve the level adjustment of downgrading.

[0102] In a specific embodiment, grading the new target data to be graded includes:

[0103] S71: Obtain the current level of the target data to be graded, and sequentially supplement the data at the next level of the current level into the current level;

[0104] S72: Use the new target data to be graded and the other data to be graded except for those classified into the current level and the level above the current level as the new data to be graded;

[0105] S73: Determine whether there is new data to be graded that has not been classified into the current level. If so, execute step S74; otherwise, execute step S77;

[0106] S74: Use the current level data in the current level as the target prediction data, and calculate the inference strength between the target prediction data and the new data to be graded;

[0107] S75: Sort the inference strength to obtain a gradient sequence;

[0108] S76: Determine the same number of inference strengths as the accommodation data volume of the next level from high to low in the gradient sequence, classify the new data to be graded corresponding to the inference strength into the next level, and use the next level as the current level and return to step S73; where the next level is the next level of the current level;

[0109] S77: Complete the complete grading of the new data to be graded.

[0110] As Figure 2 、 Figure 3 shown, the embodiments of the present invention provide an intelligent connected vehicle data grading device supporting grading adjustment. The device embodiments can be implemented through software, or through hardware or a combination of software and hardware. From the hardware level, as Figure 2 shown, it is a hardware architecture diagram of a computing device where the intelligent connected vehicle data grading device supporting grading adjustment provided by the embodiments of the present invention is located. In addition to Figure 2 the shown processor, memory, network interface, and non-volatile memory, the computing device where the device is located in the embodiments usually may also include other hardware, such as a forwarding chip responsible for processing packets, etc. Taking software implementation as an example, as Figure 3 shown, as a logically meaningful device, it is formed by the CPU of its computing device reading the corresponding computer program in the non-volatile memory into the memory and running. An intelligent connected vehicle data grading device supporting grading adjustment provided by the present embodiment, the device includes:

[0111] An acquisition module 300, configured to acquire data to be graded and standard data at the highest level;

[0112] The operation module 302 is configured to use the standard data as target prediction data, calculate the inference strength between the target prediction data and the data to be classified, and sort the inference strength to obtain a gradient sequence;

[0113] The classification module 304 is configured to determine the level of the data to be classified according to the gradient sequence, a preset number of classifications, and the data volume accommodated in each preset level.

[0114] In some specific embodiments, the acquisition module 300 can be used to execute the above step 100, the operation module 302 can be used to execute the above steps 102 and 104, and the classification module 304 can be used to execute the above step 106.

[0115] In an embodiment of the present invention, the data to be classified includes single data to be classified and / or a data group to be classified obtained by combining at least two of the single data to be classified.

[0116] In an embodiment of the present invention, the operation module 302 is further configured to perform the following operations:

[0117] For each of the target prediction data, the following is performed:

[0118] S21: Determine whether the data type of the data to be classified is the same as the data type of the target prediction data, and determine whether there is a target rule set corresponding to the data type in a preset rule library; if both judgment results are yes, execute step S22, otherwise execute step S25;

[0119] S22: The target rule set includes multiple prediction rules, and the prediction accuracy of each prediction rule within a specified time is output according to a preset inference accuracy threshold;

[0120] S23: Determine the maximum prediction accuracy among the prediction accuracies corresponding to all the prediction rules as the prediction accuracy of the data to be classified;

[0121] S24: Obtain the risk probability of the data to be classified, and use the product of the risk probability and the prediction accuracy as the inference strength;

[0122] S25: Determine that the inference strength is 0.

[0123] In an embodiment of the present invention, the classification module 304 is further configured to perform the following operations:

[0124] S31: Determine whether the current data volume in the highest level is less than the data volume accommodated in the preset highest level; if the judgment result is yes, execute step S32, otherwise execute step S34;

[0125] S32: Determine the data to be classified corresponding to the highest inference intensity from the gradient sequence, and use this data to be classified as the first standard data and classify it into the highest level;

[0126] S33: Use the first standard data as the target prediction data, calculate the inference intensity between the target prediction data and the data to be classified, sort the inference intensity to obtain a gradient sequence, and return to step S31;

[0127] S34: If the judgment result is no, determine the same number of the inference intensities as the accommodation data volume of the preset second-highest level from high to low in the gradient sequence, and classify the data to be classified corresponding to this inference intensity into the preset second-highest level to complete the classification of the preset second-highest level;

[0128] S41: Use the preset level that has been newly completed with classification as the current preset level, determine whether there is data to be classified that has not been classified into the current preset level. If there is, execute step S42; otherwise, execute S45;

[0129] S42: Use the current level data in the current preset level as the target prediction data, and calculate the inference intensity between the target prediction data and the data to be classified;

[0130] S43: Sort this inference intensity to obtain a gradient sequence;

[0131] S44: Determine the same number of inference intensities as the accommodation data volume of the preset next level from high to low in this gradient sequence, and classify the data to be classified corresponding to this inference intensity into the preset next level, and return to step S41; wherein, the preset next level is the next level of the current preset level;

[0132] S45: Complete the complete classification of the data to be classified.

[0133] In an embodiment of the present invention, when the data to be classified is the data group to be classified, if the data group to be classified has been classified into a known level, each single data to be classified included in the data group to be classified is also classified into the known level;

[0134] In an embodiment of the present invention, when the data to be classified is the data group to be classified and the single data to be classified, obtain the first known level where the data group to be classified is located and the second known levels where each single data to be classified included in the data group to be classified is located;

[0135] Judge whether there is the first known level higher than the second known level;

[0136] If the judgment result is "exists", classify the single data to be classified corresponding to the second known level into the first known level.

[0137] In an embodiment of the present invention, it further includes: a grading adjustment module, and the grading adjustment module is further configured to perform the following operations:

[0138] S70: When the level of the target data to be classified in the data to be classified is higher than the preset expected level, obtain the data accuracy and data density of the data to be classified, and reduce the data accuracy and / or the data density of the data to be classified to obtain new target data to be classified;

[0139] S71: Obtain the current level of the target data to be classified, and sequentially supplement the data at the next level of the current level into the current level;

[0140] S72: Use the new target data to be classified and the other data to be classified except those classified into the current level and the level above the current level as new data to be classified;

[0141] S73: Determine whether there is new data to be classified that has not been classified into the current level. If so, execute step S74; otherwise, execute step S77;

[0142] S74: Use the current level data in the current level as target prediction data, and calculate the inference strength between the target prediction data and the new data to be classified;

[0143] S75: Sort the inference strength to obtain a gradient sequence;

[0144] S76: Determine, from high to low in the gradient sequence, the same number of inference strengths as the accommodation data volume of the next level, and classify the new data to be classified corresponding to the inference strength into the next level, and use the next level as the current level and return to step S73; wherein, the next level is the next level of the current level;

[0145] S77: Complete the complete classification of the new data to be classified.

[0146] It can be understood that the structure illustrated in the embodiments of the present invention does not constitute a specific limitation on an intelligent networked vehicle data grading device supporting grading adjustment. In other embodiments of the present invention, an intelligent networked vehicle data grading device supporting grading adjustment may include more or fewer components than shown in the figure, or combine certain components, or split certain components, or have different component arrangements. The illustrated components can be implemented in hardware, software, or a combination of software and hardware.

[0147] For the information interaction, execution process, etc. between the modules in the above device, since they are based on the same concept as the method embodiments of the present invention, the specific content can be referred to the description in the method embodiments of the present invention, and will not be elaborated here.

[0148] An embodiment of the present invention further provides a computing device, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, a method for grading and adjusting intelligent connected vehicle data in any embodiment of the present invention is implemented.

[0149] An embodiment of the present invention further provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the processor is caused to execute a method for grading and adjusting intelligent connected vehicle data in any embodiment of the present invention.

[0150] An embodiment of the present application further provides a computer program product. The computer program product includes a computer program. The processor of a computer device reads the computer program from a computer-readable storage medium, and the processor executes the computer program, so that the computer device executes a method for grading and adjusting intelligent connected vehicle data described in any one of the above embodiments.

[0151] Specifically, a system or device equipped with a storage medium can be provided. A software program code for implementing the functions in any one of the above embodiments is stored on the storage medium, and the computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0152] In this case, the program code read from the storage medium itself can implement the functions in any one of the above embodiments. Therefore, the program code and the storage medium storing the program code constitute a part of the present invention.

[0153] Embodiments of the storage medium for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Optionally, the program code can be downloaded from a server computer through a communication network.

[0154] In addition, it should be clear that not only can the actual operations be completed in part or in whole by executing the program code read by the computer, but also by instructions based on the program code to cause an operating system, etc. on the computer to complete the actual operations, so as to implement the functions in any one of the above embodiments.

[0155] In addition, it can be understood that the program code read from the storage medium is written into the memory provided in the expansion board inserted into the computer or into the memory provided in the expansion module connected to the computer, and then based on the instructions of the program code, the CPU or the like installed on the expansion board or the expansion module is made to execute part or all of the actual operations, thereby implementing the functions of any one of the above embodiments.

[0156] It should be noted that in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0157] Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes various media such as ROM, RAM, magnetic disk or optical disc that can store program code.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for classifying data of intelligent connected vehicles supporting class adjustment, characterized in that: include: Obtain the data to be classified and the standard data at the highest level; Taking the standard data as target prediction data, calculating the reasoning strength between the target prediction data and the data to be classified; Sorting the inference strengths to obtain a gradient sequence; Determining the level of the data to be classified according to the gradient sequence, the preset number of classifications and the preset amount of data accommodated at each level; Taking the standard data as target prediction data, calculating the reasoning strength between the target prediction data and the data to be classified, including: For each target prediction data, execute: S21: Determine whether the data type of the data to be classified is the same as the data type of the target prediction data, and determine whether there is a target rule set corresponding to the data type in the preset rule base; if the judgment results are both yes, execute step S22, otherwise execute step S25; S22: The target rule set includes a plurality of prediction rules, and outputs the prediction accuracy of each prediction rule within a specified time according to a preset inference accuracy threshold; S23: Determine the maximum prediction accuracy from the prediction accuracies corresponding to all the prediction rules as the prediction accuracy of the data to be classified; S24: Obtaining the risk probability of the data to be classified, and taking the product of the risk probability and the prediction accuracy as the reasoning strength; S25: Determine that the reasoning strength is 0.

2. The method according to claim 1, characterized in that The step of determining the level of the data to be classified according to the gradient sequence, the preset number of classifications and the preset amount of data accommodated at each level includes: S31: Determine whether the current data volume in the highest level is less than the preset data volume of the highest level; if the determination result is yes, execute step S32, otherwise execute step S34; S32: determining the data to be classified corresponding to the highest reasoning strength from the gradient sequence, and taking the data to be classified as the first standard data and classifying it into the highest level; S33: taking the first standard data as target prediction data, calculating the inference strength between the target prediction data and the data to be graded, sorting the inference strengths to obtain a gradient sequence, and returning to step S31; S34: If the judgment result is no, determine the inference strength of the same amount as the amount of data accommodated at the preset second-highest level from the gradient sequence from high to low, and classify the data to be classified corresponding to the inference strength into the preset second-highest level to complete the classification of the preset second-highest level.

3. The method according to claim 2, characterized in that After completing the preset second-highest level classification, the process further includes: S41: taking the latest preset level of classification as the current preset level, determining whether there is data to be classified that is not classified into the current preset level, if so, executing step S42, otherwise executing step S45; S42: taking the current level data in the current preset level as target prediction data, and calculating the reasoning strength between the target prediction data and the data to be classified; S43: sorting the reasoning strength to obtain a gradient sequence; S44: determining the same number of inference strengths as the amount of data accommodated in the preset next level from high to low in the gradient sequence, and classifying the to-be-classified data corresponding to the inference strength into the preset next level, and returning to step S41; wherein the preset next level is the level below the current preset level; S45: Complete the classification of the data to be classified.

4. The method according to claim 2, characterized in that: The data to be classified includes a single data to be classified and / or a data group to be classified obtained by combining at least two single data to be classified; When the data to be classified is the data group to be classified, if the data group to be classified has been classified into a known level, each of the single data to be classified included in the data group to be classified is also classified into the known level; or, When the data to be classified is the data group to be classified and the single data to be classified, obtaining the first known level of the data group to be classified and the second known levels of the single data to be classified included in the data group to be classified; Determine whether the first known level is higher than the second known level; If the judgment result is yes, the single data to be classified corresponding to the second known level is classified into the first known level.

5. The method according to any one of claims 1 to 4, characterized in that: After determining the level of the data to be classified according to the gradient sequence, the preset number of classifications and the preset amount of data accommodated at each level, the method further includes: When the level of the target data to be classified in the data to be classified is higher than the preset expected level, the data accuracy and data density of the data to be classified are obtained, and the data accuracy and / or the data density of the data to be classified are reduced to obtain new target data to be classified, and the new target data to be classified is classified.

6. The method according to claim 5, characterized in that The step of grading the new target data to be graded includes: S71: obtaining the current level of the target data to be classified, and sequentially adding the data of the next level below the current level to the current level; S72: taking the new target data to be classified and the other data to be classified except those classified into the current level and the level above the current level as new data to be classified; S73: Determine whether there is new data to be classified that is not classified into the current level. If yes, execute step S74; otherwise, execute step S77; S74: taking the current level data in the current level as target prediction data, and calculating the reasoning strength between the target prediction data and the new data to be classified; S75: sorting the reasoning strength to obtain a gradient sequence; S76: Determine the inference strength of the same amount as the amount of data to be accommodated in the next level from high to low in the gradient sequence, and classify the new data to be classified corresponding to the inference strength into the next level, and take the next level as the current level and return to step S73; wherein the next level is the next level of the current level; S77: Complete the complete classification of the new data to be classified.

7. A data classification device for intelligent connected vehicles supporting classification adjustment, characterized in that: Used to implement the method according to any one of claims 1 to 6, comprising: An acquisition module, used to acquire the data to be classified and the standard data at the highest level; A calculation module, used for taking the standard data as target prediction data, calculating the inference strength between the target prediction data and the data to be graded, and sorting the inference strengths to obtain a gradient sequence; The classification module is used to determine the level of the data to be classified according to the gradient sequence, the preset number of levels and the preset amount of data accommodated at each level.

8. A computing device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method according to any one of claims 1 to 6.

Images