CN114817696A - Feature acquisition, route sorting method, device, electronic device and storage medium - Google Patents

Abstract

Embodiments of the present invention provide a method, apparatus, electronic device, and storage medium for feature acquisition and route sorting. The feature acquisition method includes: acquiring a to-be-processed route, the to-be-processed route includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination; for each road segment combination, acquiring the transition corresponding to the current road segment combination probability, the transition probability corresponding to the current road segment combination is used to indicate the probability of transferring from the previous road segment in the current road segment combination to the latter road segment in the current road segment combination; based on the transition probability corresponding to each road segment combination, Calculate the route feature corresponding to the to-be-processed route. In the embodiment of the present invention, the transition probability can reflect the correlation between two adjacent road sections, and the possibility of transferring from the previous road section to the next road section. Therefore, the route feature calculated based on the transition probability is more accurate.

Description

Feature acquisition, route sorting method, device, electronic device and storage medium

technical field

The present invention relates to the technical field of data processing, and in particular, to a method, device, electronic device and storage medium for feature acquisition and route sorting.

Background technique

With the continuous development of electronic map technology, electronic map services have provided people with great convenience. The user inputs the starting and ending points, and the electronic map can automatically generate multiple routes and recommend routes for the user.

In the process of route recommendation, it is necessary to obtain the route features of the route for analysis. At present, in the process of obtaining the route features of a route, the attributes of each road section included in the route, such as road grade, road length and other attributes, are usually obtained first; Route characteristics of the route, such as route length, straight road length, number of turns, etc.

However, in the above method for obtaining route features, statistical calculation is mainly performed based on independent road segment attributes, and the accuracy of the obtained route features is low.

SUMMARY OF THE INVENTION

In view of the above problems, embodiments of the present invention are proposed to provide a feature acquisition, route sorting method, apparatus, electronic device, and storage medium that overcome the above problems or at least partially solve the above problems.

In a first aspect, an embodiment of the present invention discloses a feature acquisition method, including:

Obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination;

For each road segment combination, the transition probability corresponding to the current road segment combination is obtained; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the next one in the current road segment combination. the probability of a road segment;

Based on the transition probability corresponding to each road segment combination, the route feature corresponding to the to-be-processed route is calculated.

Optionally, calculating the route feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment includes: calculating the sequence statistical feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment. ; and/or, based on the transition probability corresponding to each road segment combination, calculate the likelihood probability feature corresponding to the route to be processed.

Optionally, calculating the sequence statistical features corresponding to the route to be processed based on the transition probabilities corresponding to the combinations of each road segment includes: calculating a statistical value of the transition probability corresponding to the combination of the various road segments, and using the statistical value As the sequence statistical feature corresponding to the route to be processed; wherein, the statistical value includes at least one of the following: maximum value, minimum value, median, average value, and standard deviation.

Optionally, calculating the likelihood probability feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment includes: calculating the product of the transition probability corresponding to the combination of each road segment, and using the product as Likelihood probability feature corresponding to the route to be processed.

Optionally, the acquiring the transition probability corresponding to the current road segment combination includes: querying the transition probability corresponding to the current road segment combination from the pre-stored transition probabilities corresponding to each candidate road segment combination; wherein, each candidate road segment combination The corresponding transition probability is obtained by statistics based on the relevant data of the historical route.

Optionally, the transition probability corresponding to each candidate road segment combination is obtained in the following manner: every two candidate road segments connected in sequence in the historical route form a candidate road segment combination; for each candidate road segment combination, count the current candidate road segments. The first total number of times the previous candidate road segment in the combination appears in the historical route, and the second total number of times that the current candidate road segment combination appears in the historical route is counted; based on the first total number of times and all The second total number of times is used to calculate the transition probability corresponding to the current candidate road segment combination.

In a second aspect, an embodiment of the present invention discloses a method for sorting routes, including:

Obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination;

For each road segment combination, the transition probability corresponding to the current road segment combination is obtained; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the next one in the current road segment combination. the probability of a road segment;

Calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination;

Route sorting is performed based on route features corresponding to the routes to be processed.

In a third aspect, an embodiment of the present invention discloses a feature acquisition device, including:

a first obtaining module, configured to obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and each two road segments connected in sequence constitutes a road segment combination;

The second acquiring module is configured to acquire, for each road segment combination, the transition probability corresponding to the current road segment combination; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the The probability of the next road segment in the current road segment combination;

The first calculation module is configured to calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination.

Optionally, the first calculation module includes: a first feature calculation unit, configured to calculate the sequence statistical feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment; and/or, a second feature A calculation unit, configured to calculate the likelihood probability feature corresponding to the route to be processed based on the transition probability corresponding to each road segment combination.

Optionally, the first feature calculation unit is specifically configured to calculate the statistical value of the transition probability corresponding to the combination of each road segment, and use the statistical value as the sequence statistical feature corresponding to the route to be processed; wherein, the Statistical values include at least one of the following: maximum, minimum, median, mean, standard deviation.

Optionally, the second feature calculating unit is specifically configured to calculate the product of the transition probabilities corresponding to the combinations of each road segment, and use the product as the likelihood probability feature corresponding to the route to be processed.

Optionally, the second obtaining module is specifically configured to query the transition probability corresponding to the current road segment combination from the pre-stored transition probabilities corresponding to each candidate road segment combination; wherein, the transition corresponding to each candidate road segment combination The probability is based on the statistics of the relevant data of the historical route.

Optionally, the transition probability corresponding to each candidate road segment combination is obtained by the following modules: a determination module, used for forming a candidate road segment combination for every two candidate road segments connected in sequence in the historical route; a statistics module, used for For each candidate road segment combination, count the first total number of times the previous candidate road segment in the current candidate road segment combination appears in the historical route, and count the second total number of times the current candidate road segment combination appears in the historical route ; a second calculation module, configured to calculate the transition probability corresponding to the current candidate road segment combination based on the first total number of times and the second total number of times.

In a fourth aspect, an embodiment of the present invention discloses a route sorting device, including:

The third obtaining module is used to obtain the route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination;

The fourth obtaining module is configured to obtain, for each road segment combination, the transition probability corresponding to the current road segment combination; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the The probability of the next road segment in the current road segment combination;

a third calculation module, configured to calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination;

A sorting module, configured to sort the routes based on the route features corresponding to the routes to be processed.

In a fifth aspect, an embodiment of the present invention discloses an electronic device, comprising: one or more processors; and one or more machine-readable media on which instructions are stored; When executed by each processor, the processor is caused to execute the feature acquisition method described in any one of the above, or execute the route sorting method described above.

In a sixth aspect, an embodiment of the present invention discloses a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, implements the feature acquisition method described in any of the above, or implements the above-mentioned feature acquisition method. Route sorting method.

In the embodiment of the present invention, a to-be-processed route is obtained, the to-be-processed route includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination; for each road segment combination, the transition probability corresponding to the current road segment combination is obtained ; Calculate the route feature corresponding to the route to be processed based on the transition probability corresponding to each road segment combination. It can be seen from this that, in the embodiment of the present invention, since the transition probability corresponding to the current road segment combination is used to indicate the probability of transitioning from the previous road segment in the current road segment combination to the latter road segment in the current road segment combination, the transition probability can reflect the two phases. The correlation between the connecting sections reflects the possibility of transferring from the previous section to the next section, so the route features calculated based on the transition probability are more accurate.

Description of drawings

FIG. 1 is a flowchart of steps of a feature acquisition method according to an embodiment of the present invention.

FIG. 2 is a flow chart of steps of a method for obtaining transition probability according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of a road section according to an embodiment of the present invention.

FIG. 4 is a flowchart of steps of another feature acquisition method according to an embodiment of the present invention.

FIG. 5 is a flow chart of steps of a route sorting method according to an embodiment of the present invention.

FIG. 6 is a structural block diagram of a feature acquisition apparatus according to an embodiment of the present invention.

FIG. 7 is a structural block diagram of another feature acquisition apparatus according to an embodiment of the present invention.

FIG. 8 is a structural block diagram of a route sorting apparatus according to an embodiment of the present invention.

Detailed ways

In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

The mining of route features is of great significance to the electronic map business. For example, in the route recommendation process, multiple routes are recalled first, and then the recalled routes are sorted. During the route sorting process, the sorting can be performed based on the route characteristics of the routes. The feature acquisition method and the route sorting method in the embodiments of the present invention can be applied to various services with route planning requirements. For example, route planning in food delivery business, route planning in taxi-hailing business, route planning in shared bicycle and shared tram business, and so on. The embodiments of the present invention can be applied to a server with a route planning function. For example, the server of the map business, etc.

Hereinafter, detailed description will be given through the following examples.

Referring to FIG. 1 , a flowchart of steps of a feature acquisition method according to an embodiment of the present invention is shown.

As shown in Figure 1, the feature acquisition method may include the following steps:

Step 101, acquiring the route to be processed.

The to-be-processed route can be a route with route feature mining requirements. For example, in the route recommendation business, the routes are recalled first, and then the recalled routes are sorted. After sorting, some routes in the top order are selected for recommendation. Among them, each recalled route can be used as a route to be recommended.

The to-be-processed route may contain multiple links (Links) connected in sequence, and each link can be represented by a link ID (Link ID) corresponding to the link. Therefore, a to-be-processed route consists of an ordered segment of link IDs. In the embodiment of the present invention, every two road segments connected in sequence are formed into a road segment combination, so a road segment combination includes two ordered road segment identifiers.

Step 102, for each road segment combination, obtain the transition probability corresponding to the current road segment combination.

The transition probability corresponding to the current road segment combination is used to indicate the probability of transitioning from the previous road segment in the current road segment combination to the latter road segment in the current road segment combination.

Step 103: Calculate the route feature corresponding to the route to be processed based on the transition probability corresponding to each road segment combination.

Each road segment combination in the route to be processed corresponds to a transition probability, and based on the transition probability corresponding to each road segment combination, the route feature corresponding to the to-be-processed route can be calculated. For example, the route feature may include representative values (such as the maximum value, minimum value, median, etc.) selected from the transition probabilities corresponding to each road segment combination, and the route feature may also include the statistical transition probability corresponding to each road segment combination. The value obtained after the calculation, and so on.

In this embodiment of the present invention, since the transition probability corresponding to the current road segment combination is used to indicate the probability of transitioning from the previous road segment in the current road segment combination to the latter road segment in the current road segment combination, the transition probability can reflect the difference between the two adjacent road segments. The correlation of , reflects the possibility of transferring from the previous section to the next section, so the route features calculated based on the transition probability are more accurate.

Considering that the route features in the prior art are mainly derived from the original road segment attributes, the large amount of historical route related data accumulated on the service side is not fully utilized, and the road segment sequence level features are not utilized. Therefore, in the embodiment of the present invention, the transition probability information corresponding to each road segment combination can be obtained by statistics based on a large amount of historical route related data in advance, and the historical route data accumulated by the business can be fully utilized to improve the accuracy of route features.

Referring to FIG. 2 , a flowchart of steps of a method for acquiring transition probability according to an embodiment of the present invention is shown.

As shown in Figure 2, the transition probability acquisition method may include the following steps:

Step 201: In the historical route, every two candidate road segments connected in sequence form a candidate road segment combination.

Get tons of historical routes. For example, historical routes can be takeaway delivery routes, online car-hailing routes, shared bicycle routes, shared tram routes, and so on. A separate analysis can be performed for each type of historical route.

A historical route contains multiple candidate road segments connected in sequence, so a historical route is composed of a sequence of candidate road segment identifiers, each candidate road segment identifier represents a specific road segment passed through. For each historical route, every two candidate road segments connected in sequence from the beginning to the end included in the historical route form a candidate road segment combination.

比如，一条历史路线包含的有序的候选路段标识可以为[747084859，747084847，762429485，747084864，747084846，747084862，747084860，747084867，57216329，762429480，747084851，57216328，747084854，667768651，667768650，57197633，762429491， 747099272, 747099270]. Therefore, from this historical route, the candidate road segment combinations that can be obtained include: (747084859, 747084847), (747084847, 762429485), (762429485, 747084864), (747084864, 747084847), (747084846, 74708462868), (874708462868), 8 )、(747084860，747084867)、(747084867，57216329)、(57216329，762429480)、(762429480，747084851)、(747084851，57216328)、(57216328，747084854)、(747084854，667768651)、(667768651，667768650)、 (667768650, 57197633), (57197633, 762429491), (762429491, 747099272), (747099272, 747099270).

Step 202, for each candidate road segment combination, count the first total number of times that the previous candidate road segment in the current candidate road segment combination appears in the historical route, and count the current candidate road segment combination that appears in the historical route. The second total.

Step 203 , based on the first total number of times and the second total number of times, calculate a transition probability corresponding to the current candidate road segment combination.

In implementation, the second total number of times the current candidate road segment combination appears in the historical route can be calculated, and divided by the first total number of times the previous candidate road segment in the current candidate road segment combination appears in the historical route, to obtain The quotient of is the transition probability corresponding to the current candidate road segment combination. The transition probability corresponding to the current candidate road segment combination is used to indicate the probability of transitioning from a previous candidate road segment in the current candidate road segment combination to a subsequent candidate road segment in the current road segment combination.

For example, for the candidate link combination (Link _i , Link _j ), count the total number of times the candidate link Link _i appears in all historical routes, denoted as C(Link _{i ) ;} The total number of occurrences in the historical route, denoted as C(Link _i , Link _j ). Therefore, the transition probability P(Link _j | Link _i ) corresponding to the candidate link combination (Link _i , Link _j ) can be calculated by the following formula 1, that is, the transition probability from Link _i to Link _j :

The following description will be given by taking the road section shown in FIG. 3 as an example. FIG. 3 is a schematic diagram of a road section according to an embodiment of the present invention. As shown in Figure 3, taking link_0 as the previous road section, and the adjacent road sections are link_1, link_2, link_3, and link_4 respectively, then the combination of road sections where link_0 is located can include (link_0, link_1), (link_0, link_2), ( link_0, link_3), (link_0, link_4). Therefore, for each road segment combination where link_0 is located, the transition probability corresponding to the combination is calculated separately. For example, the transition probability corresponding to (link_0, link_1) is 0.3277, the transition probability corresponding to (link_0, link_2) is 0.1434, the transition probability corresponding to (link_0, link_3) is 0.5210, and the transition probability corresponding to (link_0, link_4) is 0.0079. It should be noted that FIG. 3 is only used for illustration and is not intended to limit the embodiments of the present invention.

After obtaining the transition probability corresponding to each candidate road segment combination through the method shown in FIG. 2 and statistics based on the relevant data of the historical route, the transition probability corresponding to each candidate road segment combination can be stored, so that in the process of obtaining the route characteristics in the subsequent process, it is possible to From the stored transition probabilities corresponding to each candidate road segment combination, query the transition probability corresponding to the required road segment combination.

Optionally, the storage method may include, but is not limited to, a table storage method, an index storage method, a KV (key-value, key-value pair) storage method, and the like. For example, the ordered candidate road segment identifiers included in the candidate road segment combination may be used as the key, and the transition probability corresponding to the candidate road segment combination may be used as the value, and stored in a KV type memory, and so on.

Referring to FIG. 4 , a flowchart of steps of another feature acquisition method according to an embodiment of the present invention is shown.

As shown in Figure 4, the feature acquisition method may include the following steps:

Step 401, acquiring the route to be processed.

Step 402 , for each road segment combination, query the transition probability corresponding to the current road segment combination from the pre-stored transition probabilities corresponding to each candidate road segment combination.

Through the method for obtaining the transition probability shown in FIG. 2, the relevant data of the historical route is analyzed, the transition probability corresponding to each candidate road segment combination is obtained by statistics, and the transition probability corresponding to each candidate road segment combination is stored. Therefore, for each road segment combination in the route to be processed, the transition probability corresponding to the current road segment combination can be queried from the pre-stored transition probabilities corresponding to each candidate road segment combination.

For example, if the transition probability corresponding to each candidate road segment combination is stored in a table storage manner, for each candidate road segment combination, the number of rows and columns of the transition probability storage corresponding to the candidate road segment combination is recorded. Therefore, according to the number of rows and columns stored in the transition probability corresponding to the candidate road segment combination, the transition probability corresponding to the candidate road segment combination is queried.

For another example, if the transition probability corresponding to each candidate road segment combination is stored in an index storage manner, then for each candidate road segment combination, an index for storing the transition probability corresponding to the candidate road segment combination is established. Therefore, according to the index stored in the transition probability corresponding to the candidate road segment combination, the transition probability corresponding to the candidate road segment combination is queried.

For another example, if the KV storage method is used to store the transition probability corresponding to each candidate road segment combination, then for each candidate road segment combination, the ordered candidate road segment identifiers included in the candidate road segment combination are used as the key, and the corresponding value is queried. The transition probability corresponding to the candidate road segment combination.

Step 403: Calculate the route feature corresponding to the route to be processed based on the transition probability corresponding to each road segment combination.

In an optional embodiment, the route features include sequence statistics. Therefore, the process of calculating the route feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment may include: calculating the sequence statistical feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment.

In an optional embodiment, the route features include likelihood probability features. Therefore, the process of calculating the route feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment may include: calculating the likelihood probability feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment.

In an optional embodiment, the route features include sequence statistics features and likelihood probability features. Therefore, the process of calculating the route feature corresponding to the route to be processed based on the transition probability corresponding to each road segment combination may include: based on the transition probability corresponding to each road segment combination, calculating the sequence statistical features and similarities corresponding to the to-be-processed route. probabilistic features.

The process of calculating the sequence statistical features corresponding to the route to be processed based on the transition probabilities corresponding to the combinations of each road segment may include: calculating a statistical value of the transition probability corresponding to the combination of each road segment, and using the statistical value as the Sequence statistics corresponding to the route to be processed. Wherein, the statistical value may include at least one of the following: maximum value, minimum value, median, average value, and standard deviation.

The process of calculating the likelihood probability feature corresponding to the route to be processed based on the transition probabilities corresponding to the combinations of each road segment may include: calculating a product of the transition probabilities corresponding to the combinations of each road segment, and using the product as the to-be-processed route The likelihood probability feature corresponding to the processing route.

For example, the sequence of road segments contained in a route to be processed is [Link ₀ , Link ₁ , Link ₂ , Link ₃ , Link ₄ , Link ₅ ], so the combination of road segments corresponding to the route to be processed may include (Link ₀ , Link ₁ ), (Link ₁ ,Link ₂ ), (Link ₂ ,Link ₃ ), (Link ₃ ,Link ₄ ), (Link ₄ ,Link ₅ ). The transition probability corresponding to (Link ₀ , Link ₁ ) is obtained as p ₀₁ , the transition probability corresponding to (Link ₁ , Link ₂ ) is p ₁₂ , the transition probability corresponding to (Link ₂ , Link ₃ ) is p ₂₃ , (Link _{3 )} , Link ₄ ) corresponds to the transition probability p ₃₄ , (Link ₄ , Link ₅ ) corresponds to the transition probability p ₄₅ , therefore, the sequence of transition probabilities corresponding to the route to be processed is [p ₀₁ ,p ₁₂ ,p ₂₃ , p ₃₄ , p ₄₅ ].

For the transition probability sequence, at least one of the five numerical features of the maximum value, the minimum value, the median, the average value, and the standard deviation is counted as the sequence statistical feature of the route to be processed.

The maximum value is max(p ₀₁ ,p ₁₂ ,p ₂₃ ,p ₃₄ ,p ₄₅ );

The minimum value is min(p ₀₁ , p ₁₂ , p ₂₃ , p ₃₄ , p ₄₅ );

The median is median(p ₀₁ ,p ₁₂ ,p ₂₃ ,p ₃₄ ,p ₄₅ );

The mean is mean(p ₀₁ ,p ₁₂ ,p ₂₃ ,p ₃₄ ,p ₄₅ );

The standard deviation is std(p ₀₁ , p ₁₂ , p ₂₃ , p ₃₄ , p ₄₅ ).

The road segment sequence corresponding to a route to be processed is regarded as a first-order Markov chain. Therefore, for the transition probability sequence, the product of all transition probabilities in the sequence is calculated to obtain the likelihood probability feature of the route to be processed. The likelihood probability feature is p=p ₀₁ ×p ₁₂ ×p ₂₃ ×p ₃₄ ×p ₄₅ .

In the embodiment of the present invention, based on the relevant data of the historical route, the transition probability of the combination of each road section in the route is mined, so as to provide a basis for the acquisition of the route feature, and the sequence statistical features calculated based on the probability corresponding to the combination of each road section in the route are combined with Likelihood probability feature, as a route feature corresponding to a route, the route feature can more accurately reflect the relevant information of the route. In the business involving the Internet map, the method of the embodiment of the present invention can be used to mine historical route data and calculate route features.

Referring to FIG. 5 , a flowchart of steps of a route sorting method according to an embodiment of the present invention is shown.

As shown in Figure 5, the route sorting method may include the following steps:

Step 501, acquiring the route to be processed.

The route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence constitute a road segment combination.

Step 502, for each road segment combination, obtain the transition probability corresponding to the current road segment combination.

The transition probability corresponding to the current road segment combination is used to indicate the probability of transitioning from a previous road segment in the current road segment combination to a subsequent road segment in the current road segment combination.

Step 503: Calculate the route feature corresponding to the route to be processed based on the transition probability corresponding to each road segment combination.

Step 504: Perform route ranking based on route features corresponding to the to-be-processed routes.

In the application scenario of route recommendation, the route is first recalled according to the starting point and end point input by the user, and then the recalled routes are sorted. In the process of route sorting, it is necessary to sort based on the route features of the routes. Therefore, the accuracy of route features can affect the accuracy of route ranking results. Since in the process of route sorting, each route that is finally output is also composed of a sequence of road segments, so the relationship between road segments (such as the transition probability feature between road segments) is excavated from the historical route to calculate the route features, which can be used to calculate the final ranking result. generate positive returns.

In an alternative embodiment, the ranking model may be pre-trained. Optionally, in the training process of the ranking model, for the sample route, the route feature corresponding to the sample route (such as the above-mentioned sequence statistic feature, likelihood probability feature, etc.) can be obtained according to the feature acquisition method of the embodiment of the present invention. For example, to obtain a sample route, the sample route includes multiple sample road segments connected in sequence, and every two sample road segments connected in sequence form a sample road segment combination; for each sample road segment combination, the transition probability corresponding to the current sample road segment combination is obtained, Based on the transition probability corresponding to each sample road segment combination, the route feature corresponding to the sample route is calculated. The sorting model is trained based on the route features corresponding to the sample route, thereby improving the sorting accuracy of the sorting model. It should be noted that during the training process, other route features of the sample route, such as route length, straight road length, and number of turns, can also be added for training.

In the route sorting process, based on the route characteristics corresponding to the routes to be processed, a sorting model is used to sort the routes to be processed. For example, for each route to be processed, the route feature of the route to be processed is used as the input of the ranking model, the ranking score of the route to be processed output by the ranking model is obtained, and the ranking of the route to be processed is determined according to the ranking score of each route to be processed. order.

For example, the sorting model may be a LambdaMART (Multiple Additive Regression Tree) model, or the like.

In the embodiment of the present invention, during the route sorting process, by adding the route features calculated by the embodiment of the present invention, the accuracy of the sorting can be effectively improved.

Referring to FIG. 6 , a structural block diagram of a feature acquisition apparatus according to an embodiment of the present invention is shown.

As shown in Figure 6, the feature acquisition device may include the following modules:

The first obtaining module 601 is used to obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence constitute a road segment combination;

The second obtaining module 602 is configured to obtain, for each road segment combination, the transition probability corresponding to the current road segment combination; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the selected road segment combination. the probability of the latter road segment in the current road segment combination;

The first calculation module 603 is configured to calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination.

Referring to FIG. 7 , a structural block diagram of another feature acquisition apparatus according to an embodiment of the present invention is shown.

As shown in Figure 7, the feature acquisition device may include the following modules:

The first obtaining module 701 is used to obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination;

The second obtaining module 702 is configured to obtain, for each road segment combination, the transition probability corresponding to the current road segment combination; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the current road segment combination. the probability of the latter road segment in the current road segment combination;

The first calculation module 703 is configured to calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination.

Optionally, the first calculation module 703 includes: a first feature calculation unit 7031, configured to calculate the sequence statistical features corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment; and/or, the first The second feature calculation unit 7032 is configured to calculate the likelihood probability feature corresponding to the route to be processed based on the transition probability corresponding to each road segment combination.

Optionally, the first feature calculation unit 7031 is specifically configured to calculate the statistical value of the transition probability corresponding to the combination of each road segment, and use the statistical value as the sequence statistical feature corresponding to the route to be processed; The statistical value includes at least one of the following: maximum value, minimum value, median, mean, standard deviation.

Optionally, the second feature calculation unit 7032 is specifically configured to calculate the product of the transition probabilities corresponding to the combinations of each road segment, and use the product as the likelihood probability feature corresponding to the route to be processed.

Optionally, the second obtaining module 702 is specifically configured to query the transition probability corresponding to the current road segment combination from the pre-stored transition probabilities corresponding to each candidate road segment combination; wherein, the corresponding candidate road segment combinations are The transition probability is obtained by statistics based on the relevant data of the historical route.

Optionally, the transition probability corresponding to each candidate road segment combination is obtained through the following modules: the determination module 704 is used to form a candidate road segment combination for every two candidate road segments connected in sequence in the historical route; the statistics module 705 is used to use For each candidate road segment combination, count the first total number of times that the previous candidate road segment in the current candidate road segment combination appears in the historical route, and count the second total number of times the current candidate road segment combination appears in the historical route. The total number of times; the second calculation module 706 is configured to calculate the transition probability corresponding to the current candidate road segment combination based on the first total number of times and the second total number of times.

In the embodiment of the present invention, since the transition probability corresponding to the current road segment combination is used to indicate the probability of transitioning from the previous road segment in the current road segment combination to the latter road segment in the current road segment combination, the transition probability can reflect the difference between two adjacent road segments. The correlation of , reflects the possibility of transferring from the previous section to the next section, so the route features calculated based on the transition probability are more accurate.

Referring to FIG. 8 , a structural block diagram of a route sorting apparatus according to an embodiment of the present invention is shown.

As shown in Figure 8, the route sorting device may include the following modules:

The third obtaining module 801 is configured to obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination;

The fourth obtaining module 802 is configured to obtain, for each road segment combination, the transition probability corresponding to the current road segment combination; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the current road segment combination. the probability of the latter road segment in the current road segment combination;

The third calculation module 803 is configured to calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination;

The sorting module 804 is configured to sort the routes based on the route characteristics corresponding to the routes to be processed.

In the embodiment of the present invention, by adding the above calculated route features, the accuracy of sorting can be effectively improved.

As for the apparatus embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and reference may be made to the partial description of the method embodiment for related parts.

In an embodiment of the present invention, an electronic device is also provided. The electronic device may include one or more processors, and one or more machine-readable media having stored thereon instructions, such as an application program. When executed by the one or more processors, the instructions cause the processors to execute the feature acquisition method described in any of the above embodiments, or to execute the route sorting method described in the above embodiments.

In an embodiment of the present invention, there is also provided a non-transitory computer-readable storage medium on which a computer program is stored, and the program can be executed by a processor of an electronic device to perform the features described in any of the above embodiments Obtaining method, or, executing the route sorting method described in the above embodiment. For example, the non-transitory computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments may be referred to each other.

It should be understood by those skilled in the art that the embodiments of the embodiments of the present invention may be provided as a method, an apparatus, or a computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media having computer-usable program code embodied therein, including but not limited to disk storage, CD-ROM, optical storage, and the like.

Embodiments of the present invention are described with reference to flowcharts and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the present invention. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing terminal equipment to produce a machine that causes the instructions to be executed by the processor of the computer or other programmable data processing terminal equipment Means are created for implementing the functions specified in a flow or flows of the flowcharts and/or a block or blocks of the block diagrams.

These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal equipment to operate in a particular manner, such that the instructions stored in the computer readable memory result in an article of manufacture comprising instruction means, the The instruction means implement the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing terminal equipment, so that a series of operational steps are performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby executing on the computer or other programmable terminal equipment The instructions executed on the above provide steps for implementing the functions specified in the flowchart flow or blocks and/or the block diagram block or blocks.

While preferred embodiments of the embodiments of the present invention have been described, additional changes and modifications to these embodiments may be made by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be construed to include the preferred embodiment as well as all changes and modifications that fall within the scope of the embodiments of the present invention.

Finally, it should also 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 these entities or that there is any such actual relationship or sequence between operations. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or terminal device comprising a list of elements includes not only those elements, but also a non-exclusive list of elements. other elements, or also include elements inherent to such a process, method, article or terminal equipment. Without further limitation, an element defined by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

A method, device, electronic device and storage medium for feature acquisition and route sorting provided by the present invention have been described above in detail. Specific examples are used in this paper to illustrate the principles and implementations of the present invention. The description is only used to help understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation and application scope. , the contents of this specification should not be construed as limiting the invention.

Claims (11)

1. a feature acquisition method, is characterized in that, comprises: Obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination; For each road segment combination, the transition probability corresponding to the current road segment combination is obtained; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the next one in the current road segment combination. the probability of a road segment; Based on the transition probability corresponding to each road segment combination, the route feature corresponding to the to-be-processed route is calculated. 2 . The method according to claim 1 , wherein calculating the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination comprises: 3 . Calculate the sequence statistical features corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment; and / or, Based on the transition probability corresponding to each road segment combination, the likelihood probability feature corresponding to the route to be processed is calculated. 3 . The method according to claim 2 , wherein calculating the sequence statistical features corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment comprises: 3 . Calculate the statistical value of the transition probability corresponding to the combination of each road segment, and use the statistical value as the sequence statistical feature corresponding to the route to be processed; Wherein, the statistical value includes at least one of the following: a maximum value, a minimum value, a median, an average value, and a standard deviation. 4 . The method according to claim 2 , wherein, calculating the likelihood probability feature corresponding to the route to be processed based on the transition probability corresponding to the combination of each road segment, comprising: 5 . Calculate the product of the transition probabilities corresponding to the combinations of each road segment, and use the product as the likelihood probability feature corresponding to the route to be processed. 5. The method according to claim 1, wherein the obtaining the transition probability corresponding to the current road segment combination comprises: From the pre-stored transition probabilities corresponding to each candidate road segment combination, query the transition probability corresponding to the current road segment combination; Wherein, the transition probability corresponding to each candidate road segment combination is obtained by statistics based on the relevant data of the historical route. 6. The method according to claim 5, wherein the transition probability corresponding to each candidate road segment combination is obtained in the following manner: In the historical route, every two candidate road segments connected in sequence form a candidate road segment combination; For each candidate road segment combination, count the first total number of times the previous candidate road segment in the current candidate road segment combination appears in the historical route, and count the second total number of times the current candidate road segment combination appears in the historical route frequency; Based on the first total number of times and the second total number of times, a transition probability corresponding to the current candidate road segment combination is calculated. 7. A method for sorting routes, comprising: Obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination; For each road segment combination, the transition probability corresponding to the current road segment combination is obtained; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the next one in the current road segment combination. the probability of a road segment; Calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination; Route sorting is performed based on route features corresponding to the routes to be processed. 8. A feature acquisition device, comprising: The first obtaining module is used to obtain the route to be processed; the route to be processed includes a plurality of road sections connected in sequence, and every two road sections connected in sequence form a road section combination; The second acquiring module is configured to acquire, for each road segment combination, the transition probability corresponding to the current road segment combination; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the The probability of the next road segment in the current road segment combination; The first calculation module is configured to calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination. 9. A route sorting device, comprising: a third obtaining module, configured to obtain a route to be processed; the route to be processed includes a plurality of road segments connected in sequence, and every two road segments connected in sequence form a road segment combination; The fourth obtaining module is configured to obtain, for each road segment combination, the transition probability corresponding to the current road segment combination; the transition probability corresponding to the current road segment combination is used to indicate the transition from the previous road segment in the current road segment combination to the The probability of the next road segment in the current road segment combination; a third calculation module, configured to calculate the route feature corresponding to the to-be-processed route based on the transition probability corresponding to each road segment combination; A sorting module, configured to sort the routes based on the route features corresponding to the routes to be processed. 10. An electronic device, comprising: one or more processors; and one or more machine-readable media having instructions stored thereon; When executed by the one or more processors, the instructions cause the processors to perform the feature acquisition method as claimed in any one of claims 1 to 6, or to perform the route ordering as claimed in claim 7 method. 11. A computer-readable storage medium, characterized in that, a computer program is stored thereon, and when the program is executed by a processor, the feature acquisition method according to any one of claims 1 to 6 is realized, or, the method as claimed in claim 1 is realized. The route sorting method described in claim 7.

Images