CN111091231A - Prediction model training method, time prediction method, training device and terminal - Google Patents
Prediction model training method, time prediction method, training device and terminal Download PDFInfo
- Publication number
- CN111091231A CN111091231A CN201911166157.0A CN201911166157A CN111091231A CN 111091231 A CN111091231 A CN 111091231A CN 201911166157 A CN201911166157 A CN 201911166157A CN 111091231 A CN111091231 A CN 111091231A
- Authority
- CN
- China
- Prior art keywords
- time
- target
- sample data
- departure
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present disclosure relates to the technical field of electronic information, and in particular to a prediction model training method, a time prediction method, a training device, and a terminal, which obtain a plurality of sample data, wherein the sample data includes departure time and arrival time of a logistics vehicle from a departure location to a destination, and influence factors affecting the departure time and the arrival time, and the influence factors include traffic information factors, weather information factors, and route information factors; and performing relevance analysis on the plurality of pieces of sample data to obtain a target relevance relation between the target time and other data except the target time in each piece of sample data, wherein the target time is arrival time or departure time, and the probability of deviation of the time prediction model is reduced and the accuracy of the time prediction model is improved by performing relevance analysis on the plurality of pieces of sample data.
Description
Technical Field
The present disclosure relates to the field of electronic information technologies, and in particular, to a prediction model training method, a time prediction method, a training apparatus, and a terminal.
Background
In the production process, logistics transportation is an extremely important link. In the production process, raw materials and (semi) finished products are arranged in each link, and the assembly line cannot normally operate due to the fact that the raw materials cannot reach a destination in time and the (semi) finished products cannot be transferred in time.
In actual production, each production activity is scheduled by time, and the logistics vehicles for transporting materials can influence the production whether arriving in advance or in delay. For example, the logistics vehicles reach in advance, so that warehouse bin explosion can be caused because the front materials are not transferred, each production link in actual production is like a domino, one link makes mistakes, and the subsequent links are affected.
Therefore, in actual production, the prediction of the transportation time of the material transported by the logistics vehicle is crucial.
Disclosure of Invention
In order to solve the above problems, the present disclosure provides a prediction model training method, a time prediction method, a training device, and a terminal, which establish a time prediction model through a plurality of sample data to predict the arrival time or departure time of a target logistics vehicle.
In a first aspect, the present disclosure provides a method for training a temporal prediction model, the method including:
obtaining a plurality of pieces of sample data, wherein the sample data comprises departure time and arrival time of a logistics vehicle from a departure place to a destination and influence factors influencing the departure time and the arrival time, and the influence factors comprise traffic information factors, weather information factors and route information factors;
performing relevance analysis on the plurality of pieces of sample data to obtain a target relevance relation between target time and other data except the target time in each piece of sample data, wherein the target time is arrival time or departure time;
and inputting the departure time, the arrival time and the target relevance relation of each sample data into a preset model for training to obtain the time prediction model.
According to an embodiment of the present disclosure, optionally, in the training method, the step of obtaining a plurality of pieces of sample data specifically includes:
obtaining a plurality of pieces of initial data;
determining whether initial data comprising outlier data exists in the plurality of pieces of initial data, and removing the initial data comprising the outlier data from the plurality of pieces of initial data to obtain a plurality of pieces of updated initial data;
and formatting the updated initial data into a standard format to obtain the sample data.
According to an embodiment of the present disclosure, optionally, in the training method, the step of obtaining a plurality of pieces of sample data specifically includes:
obtaining a plurality of pieces of initial data, and determining the missing level of each piece of initial data;
removing initial data with a first missing level from the initial data, and performing field supplementary processing on initial data with a second missing level from the initial data to obtain updated initial data, wherein the missing degree of the first level is greater than that of the second level;
and formatting the updated initial data into a standard format to obtain the sample data.
According to an embodiment of the present disclosure, optionally, in the training method, the step of performing relevance analysis on the plurality of pieces of sample data to obtain a target relevance relationship between the target time and other data except the target time in each piece of sample data includes:
and inputting the plurality of pieces of sample data into an Apriori model to obtain the target relevance relation between the target time and other data except the target time in each piece of sample data.
According to an embodiment of the present disclosure, optionally, in the training method, the step of inputting the departure time, the arrival time, and the target association relation of each sample data into a preset model for training to obtain the time prediction model specifically includes:
initializing model parameters of the preset model;
circularly inputting the departure time, the arrival time and the target relevance relation of each sample data into the preset model so as to enable the preset model to sequentially output target prediction time;
sequentially determining the difference value between the target prediction time and the corresponding target time;
adjusting the model parameters of the preset model according to the difference;
when the difference value is lower than a first preset threshold value, stopping adjusting the preset model parameters, and outputting the time prediction model;
when the target prediction time is target arrival time, the target time is arrival time, and the target relevance relation is the relation between the arrival time and other data except the arrival time in the sample data; and when the target prediction time is the target departure time, the target time is the departure time, and the target relevance relation is the relation between the departure time and other data except the departure time in the sample data.
In a second aspect, the present disclosure provides a temporal prediction method, including:
acquiring actual influence factors influencing the predicted target time of the target logistics vehicle and the reference time of the target logistics vehicle from the departure place to the destination;
inputting the actual influence factors and the reference time into a time prediction model obtained by a training method of the time prediction model to obtain predicted target time;
when the reference time is the departure time of the target logistics vehicle, the predicted target time is predicted arrival time; and when the reference time is the arrival time of the target logistics vehicle, the predicted target time is the predicted departure time.
According to an embodiment of the present disclosure, optionally, in the time prediction method, the method further includes: and pushing the predicted target time to a management terminal of the target logistics vehicle.
According to an embodiment of the present disclosure, optionally, in the time prediction method, the method further includes: after the step of pushing the predicted target time to the management terminal of the target logistics vehicle, the time prediction method further comprises the following steps:
acquiring the actual arrival time of the target logistics vehicle;
judging the time difference between the actual arrival time and the inspection time of the target logistics vehicle;
when the time difference between the actual arrival time of the target logistics vehicle and the inspection time is larger than a second preset threshold value, pushing prompt information of target time prediction errors, and sending the actual arrival time, the actual departure time and the actual influence factors of the target logistics vehicle to a memory;
when the predicted target time is predicted arrival time, the inspection time is predicted arrival time, and the actual departure time is the departure time of the target logistics vehicle; and when the predicted target time is predicted departure time, the inspection time is the arrival time of the target logistics vehicle, and the actual departure time is the predicted departure time.
In a third aspect, the present disclosure provides a training apparatus for a temporal prediction model, the training apparatus comprising:
the system comprises an obtaining module, a processing module and a display module, wherein the obtaining module is used for obtaining a plurality of pieces of sample data, the sample data comprises departure time and arrival time of a logistics vehicle from a departure place to a destination and influence factors influencing the departure time and the arrival time, and the influence factors comprise traffic information factors, weather information factors and route information factors;
the relevance determining module is used for carrying out relevance analysis on the plurality of pieces of sample data to obtain a target relevance relation between target time and other data except the target time in each piece of sample data, wherein the target time is arrival time or departure time;
the training module is used for inputting the departure time, the arrival time and the target relevance relation of each sample data into a preset model for training to obtain the time prediction model;
the obtaining module is further used for obtaining actual influence factors influencing the predicted target time of the target logistics vehicle and the reference time of the target logistics vehicle from the departure place to the destination;
and the prediction module is used for inputting the actual influence factors and the reference time into a time prediction model obtained by a training method of the time prediction model so as to obtain the predicted target time.
In a fourth aspect, the present disclosure provides a storage medium storing a computer program executable by one or more processors and operable to implement a method as described above.
In a fifth aspect, the present disclosure provides a terminal comprising a memory and a processor, the memory having stored thereon a computer program which, when executed by the processor, performs the above-mentioned method.
Compared with the prior art, one or more embodiments in the above scheme can have the following advantages or beneficial effects:
according to the prediction model training method, the time prediction method, the training device and the terminal, relevance analysis is performed according to a plurality of pieces of obtained sample data to obtain a target relevance relation between target time and other data except the target time in the sample data; and inputting the data to a preset model for training according to the relation among the departure time, the arrival time and the target relevance of each sample data to obtain the time prediction model, and reducing the probability of the time prediction model of deviation and improving the accuracy of the time prediction model by carrying out relevance analysis on a plurality of sample data.
Drawings
The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings.
Fig. 1 is a schematic flowchart of a training method of a temporal prediction model according to an embodiment of the present disclosure.
Fig. 2 is another schematic flow chart of a training method of a temporal prediction model according to an embodiment of the present disclosure.
Fig. 3 is another schematic flow chart of a training method of a temporal prediction model according to an embodiment of the present disclosure.
Fig. 4 is another schematic flow chart of a training method of a temporal prediction model according to an embodiment of the present disclosure.
Fig. 5 is a flowchart illustrating a time prediction method according to a second embodiment of the disclosure.
Fig. 6 is another schematic flow chart of a time prediction method according to a second embodiment of the disclosure.
Fig. 7 is another schematic flow chart of a time prediction method according to a second embodiment of the disclosure.
Fig. 8 is a block diagram of a training apparatus for a temporal prediction model according to a third embodiment of the present disclosure.
In the drawings, like parts are designated with like reference numerals, and the drawings are not drawn to scale.
Detailed Description
Embodiments of the present disclosure will be described in detail with reference to the accompanying drawings and examples, so that how to apply technical means to solve technical problems and achieve the corresponding technical effects can be fully understood and implemented. The embodiments and the features of the embodiments of the present disclosure can be combined with each other without conflict, and the formed technical solutions are all within the protection scope of the present disclosure.
Example one
Referring to fig. 1, the present disclosure provides a training method of a time prediction model applicable to a terminal such as a mobile phone, a computer, or a tablet computer, where the training method is applied to the terminal and performs the following steps:
step S110: obtaining a plurality of pieces of sample data, wherein the sample data comprises departure time and arrival time of the logistics vehicle from a departure place to a destination, and influence factors influencing the departure time and the arrival time, and the influence factors comprise traffic information factors, weather information factors and route information factors.
Step S120: and performing relevance analysis on the plurality of pieces of sample data to obtain a target relevance relation between the target time and other data except the target time in each piece of sample data, wherein the target time is arrival time or departure time.
Step S130: and inputting the departure time, the arrival time and the target relevance relation of each sample data into a preset model for training to obtain the time prediction model.
In step S110, the plurality of sample data may be historical driving data of the logistics vehicle from the departure point to the destination, the historical driving data may be manually recorded, and the plurality of sample data may be manually set according to historical conditions. Each sample data includes multiple kinds of data, specifically including departure time, arrival time, traffic information factors, weather information factors, and route information factors.
The traffic information factors can be divided into a plurality of levels such as smooth, general congestion and severe congestion, the traffic information factors of different levels have different degrees of influence on the departure time and the arrival time, and the severe congestion has the greatest influence on the departure time and the arrival time, namely, the running time of the logistics vehicle from the departure place to the destination when the traffic information factors are severe congestion is longer than that when the traffic information factors are congested. On the contrary, the smoothness has no influence on the departure time and the arrival time, namely, when the traffic information factor is smoothness, the logistics vehicles arrive at the destination according to the normal driving time.
The weather information factors can be classified into good and bad, when the weather information factors are good, the departure time and the arrival time are not influenced, and conversely, when the weather information factors are bad, the departure time and the arrival time are influenced, namely, the running time of the logistics vehicle from the departure place to the destination is prolonged. Wherein the good weather can be sunny weather, and the bad weather can be typhoon, rainstorm weather, etc.
The route information factor is different routes from the departure place to the destination, the distances of the different routes have different degrees of influence on the departure time and the arrival time, and the longer the distance is, the longer the driving time of the logistics vehicle from the departure place to the destination is.
It can be understood that, when a plurality of pieces of sample data are obtained from the historical data, in order to ensure the accuracy of the obtained sample data and further ensure the accuracy of the model obtained by the subsequent training, in this embodiment, the plurality of pieces of sample data obtained from the plurality of pieces of historical data may be cleaned or screened to obtain more accurate sample data.
Referring to fig. 2, one embodiment of cleaning or filtering a plurality of sample data obtained from a plurality of historical data may be that the step S110 includes steps S1111, S1112, and S1113:
step S1111: a plurality of pieces of initial data are obtained.
In the embodiment, the obtained plurality of pieces of initial data are historical traveling data which can be recorded manually for the logistics vehicle from the departure place to the destination, and the plurality of pieces of initial data can also be set manually according to historical conditions.
Step S1112: determining whether initial data including outlier data exists in the plurality of pieces of initial data, and removing the initial data including the outlier data from the plurality of pieces of initial data to obtain a plurality of pieces of updated initial data.
In this embodiment, the culling process may be performed on initial data including outlier data among the plurality of pieces of initial data. Illustratively, the departure point is a place a, the destination is a place B, and in normal cases (good traffic, good weather, shortest route), the time of use from the place a to the place B is two hours, and the acquired initial data indicate that the departure time is 13 points, the arrival time is 19 points, the traffic information factor is clear, the weather information factor is good, and the route information factor is the shortest route, and the obtained initial data indicate that 6 hours are required from the place a to the place B, which is far greater than the time of use in normal cases, indicating that the arrival time is 19 points is obviously outlier data, and therefore, the initial data are deleted from the initial data.
Step S1113: and formatting the updated initial data into a standard format to obtain the sample data.
In this embodiment, the acquired initial data may be from different sources, for example, there may be data stored in a database, and there may be data stored in an Excel table, so that formatting a plurality of pieces of initial data into a standard format is convenient for the processing of the subsequent steps. Wherein, the standard format can be csv file.
In this embodiment, when outlier data exists in each piece of initial data, the initial data with the outlier data is removed, so that the influence of error data on model training is reduced, and the accuracy of prediction of a time prediction model obtained by training is improved; and formatting the plurality of pieces of updated initial data, so that the plurality of pieces of initial data can be conveniently processed in the subsequent steps.
Referring to fig. 3, an implementation manner of cleaning or screening a plurality of sample data obtained from a plurality of historical data may be that the step S110 includes a step S1121, a step S1122, and a step S1123:
step S1121: a plurality of pieces of initial data are obtained, and a deletion level of each piece of the initial data is determined.
In the embodiment, the obtained plurality of pieces of initial data are historical traveling data which can be recorded manually for the logistics vehicle from the departure place to the destination, and the plurality of pieces of initial data can also be set manually according to historical conditions. After the pieces of initial data are obtained, determination of the deletion level is made thereon.
Step S1122: removing initial data with a first missing level from the initial data, and performing field supplementary processing on initial data with a second missing level from the initial data to obtain updated initial data, wherein the missing degree of the first level is greater than that of the second level.
In this embodiment, the initial data with the missing level of the first level in the plurality of pieces of initial data is removed, and the data with the missing level of the first level may affect the accuracy of the time prediction model, so that the initial data is deleted from the plurality of pieces of initial data to avoid affecting the training of the time prediction model. The case that the missing degree is the first grade includes the missing of any data of departure time, arrival time and influencing factors of the logistics vehicle. Illustratively, where the initial data includes a departure time: the arrival time: 9:00, traffic information factor: congestion, weather information factor: good, route information factor: distance shortest route, material information: rice, logistics vehicle type: and the vehicle A is lack of departure time in the data, and the departure time is a factor influencing the arrival time, so that the missing degree of the data is of a first level, and the data is removed from the initial data.
And performing field supplement processing on initial data with the missing level of the second level in the plurality of pieces of initial data, and performing field supplement on the data with the missing level of the second level to be used as normal initial data. When the acquired initial data comprises the material information and the type of the logistics vehicle, the missing degree of the data is the second level because the material information and the type of the logistics vehicle are not factors which can influence the arrival time or the departure time, and when the data is missing, the data can be supplemented, and the supplementing mode can be a mode of adding a preset default field. Illustratively, where the initial data includes a departure time: 7:00, arrival time: 9:00, traffic information factor: congestion, weather information factor: good, route information factor: distance shortest route, material information: and the type of the logistics vehicle is as follows: the vehicle A can convert' material information: "modified to" material information: rice ", wherein rice may be a preset default field.
Step S1123: and formatting the updated initial data into a standard format to obtain the sample data.
In this embodiment, the implementation process is similar to the implementation process of the step S1113, and reference may be made to the implementation process, which is not described herein again.
In this embodiment, the initial data with the missing level as the first level in the multiple pieces of initial data is removed, so that the influence of error data on model training is reduced, and the prediction accuracy of the trained time prediction model is improved; performing field supplementary processing on initial data with a missing level as a second level in the plurality of pieces of initial data so as to ensure the number of training samples to the maximum extent and improve the prediction accuracy of the time prediction model; and formatting the plurality of pieces of updated initial data, so that the plurality of pieces of initial data can be conveniently processed in the subsequent steps.
It is understood that, in an embodiment of cleaning or screening multiple pieces of initial data obtained from multiple pieces of historical data, when multiple pieces of sample data are obtained, the above steps S1112 and S1122 may be performed on the multiple pieces of initial data, that is, the initial data including outlier data in the multiple pieces of initial data is removed, the initial data missing at the first level in the multiple pieces of initial data is removed, the initial data missing at the second level in the multiple pieces of initial data is field-supplemented, the updated multiple pieces of initial data are obtained by performing the above steps S1112 and S1122, and finally the updated multiple pieces of initial data are formatted. The present embodiment does not limit the execution sequence of step S1112 and step S1122.
In step S120, the target relevance relationship refers to a factor that may affect the target time, and different target relevance relationships affect the target time to different extents.
Illustratively, among the plurality of pieces of sample data collected, there is a difference between the arrival time and the traffic information factor in the first sample data (departure time: 7:00, arrival time: 9:00, traffic information factor: clear, weather information factor: good, route information factor: shortest distance line), the second sample data (departure time: 7:00, arrival time: 11:00, traffic information factor: congestion, weather information factor: good, route information factor: shortest distance line), the third sample data (departure time: 7:00, arrival time: 13:00, traffic information factor: severe congestion, weather information factor: good, route information factor: shortest distance line), the first sample data, the second sample data, and the third sample data, therefore, the relationship between the arrival time and the traffic information factor is determined as the target relevance relationship. The way of analyzing the relevance of the multiple pieces of sample data may be to calculate the relevance between the parameters in the multiple pieces of sample data by using a relevance calculation expression, or to calculate the relevance between the parameters in the multiple pieces of sample data by using an Apriori model.
In this embodiment, the step S120 may be: and inputting the plurality of pieces of sample data into an Apriori model to obtain the target relevance relation between the target time and other data except the target time in the sample data.
In this embodiment, the Apriori model is an association rule mining algorithm, which finds out the relationship between the target time in a plurality of pieces of sample data and the association relationship between other data in the sample data except the target time by using a layer-by-layer search iteration method, and is used for training the time prediction model.
For example, the first sample data, the second sample data, and the third sample data are used as examples, in the three sample data, only the traffic information factor and the arrival time are different, and therefore, the relationship between the traffic information factor and the arrival time may be determined as the target relevance relationship.
In step S130, the time prediction model is obtained by inputting the departure time, the arrival time, and the target relevance relationship of each sample data into a preset model for training.
For example, still taking the first sample data, the second sample data, and the third sample data as examples, determining that the relationship between the arrival time and the traffic information factor is a target relevance relationship, if the purpose of the time prediction model is to predict the arrival time, determining the required time of the logistics vehicle from the departure location to the destination according to the determined target relevance relationship, and then determining the output result of the time prediction model, that is, the predicted arrival time, according to the departure time and the required time of the logistics vehicle.
For example, still taking the first sample data, the second sample data, and the third sample data as examples, determining that the relationship between the arrival time and the traffic information factor is a target relevance relationship, if the purpose of the time prediction model is to predict the departure time, determining the required time of the logistics vehicle from the departure place to the destination according to the determined target relevance relationship, and then determining the output result of the time prediction model, that is, the predicted departure time, according to the arrival time and the required time of the logistics vehicle.
In this embodiment, a target association relationship between a target time and other data in the sample data except the target time is obtained by performing association analysis on multi-sample data, and the departure time, the arrival time, and the target association relationship of each sample data are input to a preset model for training to obtain the time prediction model, so that the probability of occurrence of a deviation of the time prediction model is reduced, and the accuracy of the time prediction model is improved.
Referring to fig. 4, one embodiment of the method for training the preset model may be that the step S130 specifically includes the following steps S1301, S1302, S1303, and S1404:
step S1301: initializing the model parameters of the preset model.
In this embodiment, the model parameters of the predetermined model may be initialized to selected constant values.
Step S1302: circularly inputting the departure time, the arrival time and the target relevance relation of each sample data into the preset model so as to enable the preset model to sequentially output target prediction time.
In this embodiment, the input cycle may be a plurality of sample data records of the same batch; or a plurality of sample data records of different batches, that is, a plurality of sample data records input each time are different. Generally, a plurality of sample data records of different batches are input into the preset model.
In this embodiment, the model for predicting the arrival time of the logistics vehicle may be trained to obtain the arrival time prediction model, or the model for predicting the departure time of the logistics vehicle may be trained to obtain the departure time prediction model.
Illustratively, taking training target predicted time as predicted arrival time and target time as arrival time, the obtained target association relationship is the relationship between traffic information factors and arrival time as an example, and assuming that the input departure time is 12:00 and the arrival time is 14: 00, in this case, the purpose of the time prediction model is to predict the arrival time, determine that the required elapsed time of the physical distribution vehicle from the departure point to the destination is 180 minutes based on the determined target correlation, and determine that the output result of the time prediction model is 15: 00, the arrival time predicted by the preset model in this case.
Step S1303: and sequentially determining the difference value between the target prediction time and the corresponding target time.
In this embodiment, the difference value may represent the accuracy of the preset model in predicting the target prediction time in this round of training, and the output result of the time prediction model is still 15: 00 for example, the arrival time is 14: 00, and the predicted arrival time (target predicted time) of the preset model is 15: 00, it is determined that the difference between the target predicted time and the corresponding target time is 60 minutes.
Step S1304: and adjusting the model parameters of the preset model according to the difference.
In this embodiment, the model parameters of the preset model are adjusted according to the difference to reduce the difference between the target prediction time and the arrival time in the corresponding sample data, the preset model is updated, and the updated preset model is put into the next round of training.
Step S1305: and when the difference value is lower than a first preset threshold value, stopping adjusting the preset model parameters, and outputting the time prediction model.
In this embodiment, when the difference between the target prediction time and the arrival time in the corresponding sample data is no longer significantly reduced, the adjustment of the model parameters is stopped, and the road surface condition detection model is output.
Illustratively, the result is still output as 15: 00 for example, the arrival time is 14: 00, and the target prediction time is 15: 00, determining that the difference value between the target prediction time and the corresponding target time is 60 minutes, and assuming that the first preset threshold is 10 minutes, because the difference value is greater than the first preset threshold under the condition, adjusting the model parameters, updating the preset model, and training with the new preset model.
And then the output result of the time prediction model is 15: 00 for example, arrival time is 14: for example, the difference between the target prediction time and the arrival time is 4 minutes, and since the difference is smaller than the first preset threshold in this case, the adjustment of the model parameters is stopped, the model parameters of this round are used as the model parameters of the preset model, and the preset model at this time is output as the time prediction model.
In this embodiment, model parameters of the preset model are continuously adjusted by comparing the magnitude relationship between the difference and the first preset threshold to update the preset model, so as to reduce the difference between the target prediction time and the corresponding target time, so that the target prediction time output by the preset model is continuously close to the target time, and the accuracy of the preset model is improved.
In addition, when the training target prediction time is the predicted departure time, and the target time is the departure time, the training implementation process of the time prediction model is similar to the implementation process, which can be referred to and is not described herein again.
Example two
Referring to fig. 5, in the time prediction method applicable to a terminal such as a mobile phone, a computer, or a tablet computer, the time prediction method performs steps S210 and S220 when applied to the terminal:
step S210: and acquiring actual influence factors influencing the predicted target time of the target logistics vehicle and the reference time of the target logistics vehicle from the departure place to the destination.
In this embodiment, when the reference time is the departure time of the target logistics vehicle, the predicted target time is the predicted arrival time; and when the reference time is the arrival time of the target logistics vehicle, the predicted target time is the predicted departure time.
Step S220: and inputting the actual influence factors and the reference time into a time prediction model obtained by a training method of the time prediction model to obtain predicted target time.
In this embodiment, the time prediction model obtained by the training method of the time prediction model is the time prediction model obtained by the training in the first embodiment. And still taking the reference time as the departure time of the target logistics vehicle and the predicted target time as the predicted arrival time as an example, inputting the three actual influence factors and the departure time of the target logistics vehicle into a trained time prediction model to obtain the predicted arrival time when the actual influence factors include that the actual traffic influence factor is congestion, the actual weather influence factor is good and the actual route influence factor is the shortest distance route.
In addition, when the departure time is predicted, the implementation process is similar to the implementation process of the predicted arrival time, and the implementation process can be referred to, which is not described herein again.
Referring to fig. 6, the present embodiment provides another implementation of the time prediction method, which includes, in addition to the steps S210 and S220, a step S230:
step S230: and pushing the predicted target time to a management terminal of the target logistics vehicle.
In this embodiment, pushing the predicted target time to the management terminal of the target logistics vehicle may be realized by a voice notification, or may be displayed on a screen of the management terminal, and the user holding the management terminal may be a driver of the target logistics vehicle, or a warehouse transfer worker.
In this embodiment, the predicted target time is pushed to the management terminal of the target logistics vehicle, so that a target driver and a warehouse transporter who hold the management terminal can make corresponding measures, and if the logistics vehicle is predicted to arrive in advance, the target driver is reminded to slow down the advancing speed or the warehouse transporter is reminded to prepare in advance, and the situations that goods and materials burst and the target logistics vehicle cannot stop for a long time to wait for unloading due to untimely warehouse logistics turnover are avoided.
Referring to fig. 7, the present embodiment provides another implementation of the time prediction method, which includes, in addition to the above steps S210, S220, and S230, steps S240, S250, and S260:
step S240: and collecting the actual arrival time of the target logistics vehicle.
In this embodiment, after the target logistics vehicle arrives at the destination, the actual arrival time of the target logistics vehicle may be collected, and the actual arrival time may be manually recorded or determined by the time when the monitoring device of the destination captures the picture of the target logistics vehicle.
Step S250: and judging the time difference between the actual arrival time of the target logistics vehicle and the inspection time.
In this embodiment, the time difference between the actual arrival time of the target logistics vehicle and the inspection time can represent the accuracy of the time prediction model, and the smaller the difference is, the higher the accuracy of the time prediction model is represented. When the predicted target time is predicted arrival time, the inspection time is predicted arrival time, and the actual departure time is the departure time of the target logistics vehicle; and when the predicted target time is predicted departure time, the inspection time is the arrival time of the target logistics vehicle, and the actual departure time is the predicted departure time.
Step S260: and when the time difference between the actual arrival time of the target logistics vehicle and the inspection time is greater than a second preset threshold value, pushing prompt information of target time prediction errors, and sending the actual arrival time, the actual departure time and the actual influence factors of the target logistics vehicle to a memory.
In this embodiment, the second preset threshold represents the accuracy of the time prediction model, and the smaller the second preset threshold is, the higher the accuracy of the time prediction model is. When the time difference between the actual arrival time of the target logistics vehicle and the inspection time is greater than a second preset threshold, it is indicated that the time obtained by the time prediction model is far from the time when the target logistics vehicle normally arrives at the destination, that is, the accuracy of the time prediction model is low, and the time prediction model needs to be retrained to improve the accuracy of the time prediction model.
And sending the actual arrival time, the actual departure time and the actual influence factors to a memory to be used as new sample data of the time prediction model so as to retrain or update the time prediction model.
In this embodiment, by collecting the actual arrival time of the target logistics vehicle and determining the time difference between the actual arrival time and the inspection time, when the time difference is greater than the second preset threshold, the actual arrival time, the actual departure time, and the actual influence factor of the target logistics vehicle are sent to the memory, so as to provide data storage for subsequent training or updating of the road surface condition detection model.
EXAMPLE III
Referring to fig. 8, the present disclosure provides a training apparatus for a time prediction model applicable to a terminal such as a mobile phone, a computer, or a tablet computer, the training apparatus including:
the obtaining module 301 is configured to obtain a plurality of pieces of sample data, where the sample data includes departure time and arrival time of the logistics vehicle from a departure location to a destination, and influence factors affecting the departure time and the arrival time, where the influence factors include a traffic information factor, a weather information factor, and a route information factor.
The implementation principle of the obtaining module 301 is similar to that in step S110, and therefore, for specific description of the obtaining module 301, reference may be made to the first embodiment, which is not described herein again.
The relevance determining module 302 is configured to perform relevance analysis on the multiple pieces of sample data to obtain a target relevance relationship between the target time in each piece of sample data and other data except the target time, where the target time is an arrival time or a departure time.
The implementation principle of the relevance determining module 302 is similar to that of step S120, and therefore, for the specific description of the relevance determining module 302, reference may be made to embodiment one, which is not described herein again.
The training module 303 is configured to input the departure time, the arrival time, and the target relevance relationship of each sample data into a preset model for training, so as to obtain the time prediction model.
The implementation principle of the training module 303 is similar to that of step S130, and therefore, for specific description of the training module 303, reference may be made to embodiment one, which is not described herein again.
When the target time is predicted, the obtaining module 301 is further configured to obtain an actual influence factor influencing the predicted target time of the target logistics vehicle and a reference time from the departure point to the destination of the target logistics vehicle.
When predicting the target time, the implementation principle of the obtaining module 301 is also similar to that of step S210, and therefore, for specific description of the obtaining module 301, reference may be made to embodiment two, which is not described herein again.
When the target time is predicted, the prediction module 304 is configured to input the actual influencing factor and the reference time into a time prediction model obtained by a training method of the time prediction model, so as to obtain a predicted target time.
When predicting the target time, the implementation principle of the prediction module 304 is also similar to that of step S220, and therefore, for the specific description of the prediction module 304, reference may be made to embodiment two, which is not described herein again.
Example 4
The present embodiments also provide a computer readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application mall, etc., having stored thereon a computer program that when executed by a processor, performs the method steps of:
step S110: obtaining a plurality of pieces of sample data, wherein the sample data comprises departure time and arrival time of the logistics vehicle from a departure place to a destination, and influence factors influencing the departure time and the arrival time, and the influence factors comprise traffic information factors, weather information factors and route information factors.
Step S120: and performing relevance analysis on the plurality of pieces of sample data to obtain a target relevance relation between the target time and other data except the target time in each piece of sample data, wherein the target time is arrival time or departure time.
Step S130: and inputting the departure time, the arrival time and the target relevance relation of each sample data into a preset model for training to obtain the time prediction model.
The specific embodiment process of the above method steps can be referred to as embodiment one, and the detailed description of this embodiment is not repeated herein.
The computer program may further realize the following method steps when executed by the processor:
step S210: and acquiring actual influence factors influencing the predicted target time of the target logistics vehicle and the reference time of the target logistics vehicle from the departure place to the destination.
Step S220: and inputting the actual influence factors and the reference time into a time prediction model obtained by a training method of the time prediction model to obtain predicted target time.
The specific embodiment process of the above method steps can be referred to as embodiment two, and the detailed description of this embodiment is not repeated here.
EXAMPLE five
The embodiment of the present disclosure provides a terminal, which may be a mobile phone, a computer, a tablet computer, or the like, and includes a memory and a processor, where the memory stores a computer program, and the computer program, when executed by the processor, implements the methods described in the first embodiment and the second embodiment. It is to be understood that the terminal can also include multimedia components, input/output (I/O) interfaces, and communication components.
Wherein the processor is adapted to perform all or part of the steps of the method according to the first and second embodiments. The memory is used to store various types of data, which may include, for example, instructions for any application or method in the terminal, as well as application-related data.
The Processor may be an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic components, and is configured to perform the methods of the first and second embodiments.
The Memory may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk or optical disk.
The multimedia component may include a screen, which may be a touch screen, and an audio component for outputting and/or inputting an audio signal. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may further be stored in a memory or transmitted through a communication component. The audio assembly also includes at least one speaker for outputting audio signals.
The I/O interface provides an interface between the processor and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons.
The communication component is used for carrying out wired or wireless communication between the terminal and other equipment. Wireless Communication, such as Wi-Fi, bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination of one or more of them, so that the corresponding Communication component 405 may include: Wi-Fi module, bluetooth module, NFC module.
In summary, according to the prediction model training method, the time prediction method, the training device and the terminal provided by the present disclosure, relevance analysis is performed according to a plurality of obtained sample data to obtain a target relevance relationship between target time and other data in the sample data except the target time; inputting the sample data to a preset model for training according to the relation among the departure time, the arrival time and the target relevance of each sample data to obtain a time prediction model, performing relevance analysis on a plurality of sample data to reduce the probability of deviation of the time prediction model and improve the accuracy of the time prediction model, predicting the target time by using the trained time prediction model, and pushing the predicted target time to a management terminal of a target logistics vehicle so that a target driver and a warehouse transloader who hold the management terminal can take corresponding measures.
In the several embodiments provided in the embodiments of the present disclosure, it should be understood that the disclosed apparatus and method may be implemented in other manners. The system and method embodiments described above are merely illustrative.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Although the embodiments disclosed in the present disclosure are described above, the descriptions are only for the convenience of understanding the present disclosure, and are not intended to limit the present disclosure. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims.
Claims (11)
1. A method for training a temporal prediction model, the method comprising:
obtaining a plurality of pieces of sample data, wherein the sample data comprises departure time and arrival time of a logistics vehicle from a departure place to a destination and influence factors influencing the departure time and the arrival time, and the influence factors comprise traffic information factors, weather information factors and route information factors;
performing relevance analysis on the plurality of pieces of sample data to obtain a target relevance relation between target time and other data except the target time in each piece of sample data, wherein the target time is arrival time or departure time;
and inputting the departure time, the arrival time and the target relevance relation of each sample data into a preset model for training to obtain the time prediction model.
2. The training method of claim 1, wherein the step of obtaining a plurality of pieces of sample data comprises:
obtaining a plurality of pieces of initial data;
determining whether initial data comprising outlier data exists in the plurality of pieces of initial data, and removing the initial data comprising the outlier data from the plurality of pieces of initial data to obtain a plurality of pieces of updated initial data;
and formatting the updated initial data into a standard format to obtain the sample data.
3. The training method of claim 1, wherein the step of obtaining a plurality of pieces of sample data comprises:
obtaining a plurality of pieces of initial data, and determining the missing level of each piece of initial data;
removing initial data with a first missing level from the initial data, and performing field supplementary processing on initial data with a second missing level from the initial data to obtain updated initial data, wherein the missing degree of the first level is greater than that of the second level;
and formatting the updated initial data into a standard format to obtain the sample data.
4. A training method as claimed in any one of claims 1 to 3, wherein said step of performing a correlation analysis on said plurality of sample data to obtain a target correlation relationship between the target time and data other than the target time in each sample data comprises:
and inputting the plurality of pieces of sample data into an Apriori model to obtain the target relevance relation between the target time and other data except the target time in each piece of sample data.
5. A training method as claimed in any one of claims 1 to 3, wherein the step of inputting the departure time, arrival time and target correlation relationship of each sample data into a preset model for training to obtain the time prediction model specifically comprises:
initializing model parameters of the preset model;
circularly inputting the departure time, the arrival time and the target relevance relation of each sample data into the preset model so as to enable the preset model to sequentially output target prediction time;
sequentially determining the difference value between the target prediction time and the corresponding target time;
adjusting the model parameters of the preset model according to the difference;
when the difference value is lower than a first preset threshold value, stopping adjusting the preset model parameters, and outputting the time prediction model;
when the target prediction time is target arrival time, the target time is arrival time, and the target relevance relation is the relation between the arrival time and other data except the arrival time in the sample data; and when the target prediction time is the target departure time, the target time is the departure time, and the target relevance relation is the relation between the departure time and other data except the departure time in the sample data.
6. A temporal prediction method, characterized in that the temporal prediction method comprises:
acquiring actual influence factors influencing the predicted target time of the target logistics vehicle and the reference time of the target logistics vehicle from the departure place to the destination;
inputting the actual influence factors and the reference time into a time prediction model obtained by the training method of the time prediction model according to any one of claims 1 to 5 to obtain a predicted target time;
when the reference time is the departure time of the target logistics vehicle, the predicted target time is predicted arrival time; and when the reference time is the arrival time of the target logistics vehicle, the predicted target time is the predicted departure time.
7. The method of claim 6, wherein the method further comprises: and pushing the predicted target time to a management terminal of the target logistics vehicle.
8. The method of claim 7, wherein after the step of pushing the predicted target time to the management terminal of the target logistics vehicle, the time prediction method further comprises:
acquiring the actual arrival time of the target logistics vehicle;
judging the time difference between the actual arrival time and the inspection time of the target logistics vehicle;
when the time difference between the actual arrival time of the target logistics vehicle and the inspection time is larger than a second preset threshold value, pushing prompt information of target time prediction errors, and sending the actual arrival time, the actual departure time and the actual influence factors of the target logistics vehicle to a memory;
when the predicted target time is predicted arrival time, the inspection time is predicted arrival time, and the actual departure time is the departure time of the target logistics vehicle; and when the predicted target time is predicted departure time, the inspection time is the arrival time of the target logistics vehicle, and the actual departure time is the predicted departure time.
9. An apparatus for training a temporal prediction model, the apparatus comprising:
the system comprises an obtaining module, a processing module and a display module, wherein the obtaining module is used for obtaining a plurality of pieces of sample data, the sample data comprises departure time and arrival time of a logistics vehicle from a departure place to a destination and influence factors influencing the departure time and the arrival time, and the influence factors comprise traffic information factors, weather information factors and route information factors;
the relevance determining module is used for carrying out relevance analysis on the plurality of pieces of sample data to obtain a target relevance relation between target time and other data except the target time in each piece of sample data, wherein the target time is arrival time or departure time;
the training module is used for inputting the departure time, the arrival time and the target relevance relation of each sample data into a preset model for training to obtain the time prediction model;
the obtaining module is further used for obtaining actual influence factors influencing the predicted target time of the target logistics vehicle and the reference time of the target logistics vehicle from the departure place to the destination;
a prediction module, configured to input the actual influencing factor and the reference time into a time prediction model obtained by the training method of the time prediction model according to any one of claims 1 to 5, so as to obtain a predicted target time.
10. A storage medium storing a computer program executable by one or more processors for performing the method of any one of claims 1 to 8.
11. A terminal, characterized in that it comprises a memory and a processor, said memory having stored thereon a computer program which, when executed by said processor, performs the method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911166157.0A CN111091231B (en) | 2019-11-25 | 2019-11-25 | Prediction model training method, time prediction method, training device and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911166157.0A CN111091231B (en) | 2019-11-25 | 2019-11-25 | Prediction model training method, time prediction method, training device and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111091231A true CN111091231A (en) | 2020-05-01 |
| CN111091231B CN111091231B (en) | 2022-04-15 |
Family
ID=70393730
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911166157.0A Active CN111091231B (en) | 2019-11-25 | 2019-11-25 | Prediction model training method, time prediction method, training device and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111091231B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738517A (en) * | 2020-06-24 | 2020-10-02 | 詹晨 | E-commerce fresh commodity express delivery and distribution management system based on big data |
| CN112215238A (en) * | 2020-10-29 | 2021-01-12 | 支付宝(杭州)信息技术有限公司 | Method, system and device for constructing general feature extraction model |
| CN113704994A (en) * | 2021-08-25 | 2021-11-26 | 福州市规划设计研究院集团有限公司 | Method and system for building urban traffic lifeline in extreme rainstorm weather |
| CN114509996A (en) * | 2022-01-10 | 2022-05-17 | 阿里云计算有限公司 | Equipment operation time length prediction and instruction scheduling method, equipment and storage medium |
| CN114950979A (en) * | 2022-04-28 | 2022-08-30 | 广州艮业信息科技有限公司 | Linear sorting method, system, equipment and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009286222A (en) * | 2008-05-28 | 2009-12-10 | Toyota Motor Corp | Accident predicting device |
| CN102157075A (en) * | 2011-03-15 | 2011-08-17 | 上海交通大学 | Method for predicting bus arrivals |
| CN102610088A (en) * | 2012-03-08 | 2012-07-25 | 东南大学 | Method for forecasting travel time between bus stops |
| CN103207944A (en) * | 2013-02-04 | 2013-07-17 | 国家电网公司 | Electric power statistical index relevance analysis method |
| CN104821080A (en) * | 2015-03-02 | 2015-08-05 | 北京理工大学 | Intelligent vehicle traveling speed and time predication method based on macro city traffic flow |
| CN105006147A (en) * | 2015-06-19 | 2015-10-28 | 武汉大学 | Road segment travel time deducing method based on road space-time incidence relation |
| CN105139656A (en) * | 2015-09-28 | 2015-12-09 | 百度在线网络技术（北京）有限公司 | Road state prediction method and device |
| CN106339514A (en) * | 2015-07-06 | 2017-01-18 | 杜比实验室特许公司 | Method estimating reverberation energy component from movable audio frequency source |
| CN107092988A (en) * | 2017-04-21 | 2017-08-25 | 北方工业大学 | Method for predicting station-parking time of bus on special lane |
| CN107194491A (en) * | 2017-04-06 | 2017-09-22 | 广东工业大学 | A kind of dynamic dispatching method based on Forecasting of Travel Time between bus passenger flow and station |
| CN110245377A (en) * | 2019-05-08 | 2019-09-17 | 暨南大学 | A kind of travel plan recommended method and recommender system |
-
2019
- 2019-11-25 CN CN201911166157.0A patent/CN111091231B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009286222A (en) * | 2008-05-28 | 2009-12-10 | Toyota Motor Corp | Accident predicting device |
| CN102157075A (en) * | 2011-03-15 | 2011-08-17 | 上海交通大学 | Method for predicting bus arrivals |
| CN102610088A (en) * | 2012-03-08 | 2012-07-25 | 东南大学 | Method for forecasting travel time between bus stops |
| CN103207944A (en) * | 2013-02-04 | 2013-07-17 | 国家电网公司 | Electric power statistical index relevance analysis method |
| CN104821080A (en) * | 2015-03-02 | 2015-08-05 | 北京理工大学 | Intelligent vehicle traveling speed and time predication method based on macro city traffic flow |
| CN105006147A (en) * | 2015-06-19 | 2015-10-28 | 武汉大学 | Road segment travel time deducing method based on road space-time incidence relation |
| CN106339514A (en) * | 2015-07-06 | 2017-01-18 | 杜比实验室特许公司 | Method estimating reverberation energy component from movable audio frequency source |
| CN105139656A (en) * | 2015-09-28 | 2015-12-09 | 百度在线网络技术（北京）有限公司 | Road state prediction method and device |
| CN107194491A (en) * | 2017-04-06 | 2017-09-22 | 广东工业大学 | A kind of dynamic dispatching method based on Forecasting of Travel Time between bus passenger flow and station |
| CN107092988A (en) * | 2017-04-21 | 2017-08-25 | 北方工业大学 | Method for predicting station-parking time of bus on special lane |
| CN110245377A (en) * | 2019-05-08 | 2019-09-17 | 暨南大学 | A kind of travel plan recommended method and recommender system |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738517A (en) * | 2020-06-24 | 2020-10-02 | 詹晨 | E-commerce fresh commodity express delivery and distribution management system based on big data |
| CN112215238A (en) * | 2020-10-29 | 2021-01-12 | 支付宝(杭州)信息技术有限公司 | Method, system and device for constructing general feature extraction model |
| CN113704994A (en) * | 2021-08-25 | 2021-11-26 | 福州市规划设计研究院集团有限公司 | Method and system for building urban traffic lifeline in extreme rainstorm weather |
| CN113704994B (en) * | 2021-08-25 | 2023-09-01 | 福州市规划设计研究院集团有限公司 | Urban traffic lifeline construction method and system in extreme stormy weather |
| CN114509996A (en) * | 2022-01-10 | 2022-05-17 | 阿里云计算有限公司 | Equipment operation time length prediction and instruction scheduling method, equipment and storage medium |
| CN114950979A (en) * | 2022-04-28 | 2022-08-30 | 广州艮业信息科技有限公司 | Linear sorting method, system, equipment and storage medium |
| CN114950979B (en) * | 2022-04-28 | 2023-09-05 | 广州艮业信息科技有限公司 | Linear sorting method, system, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111091231B (en) | 2022-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111091231B (en) | Prediction model training method, time prediction method, training device and terminal | |
| US20230259875A1 (en) | Automatically predicting shipper behavior using machine learning models | |
| CN108257378B (en) | Traffic flow prediction method and device | |
| CN108615129B (en) | Transport capacity monitoring method and device and electronic equipment | |
| US20190378180A1 (en) | Method and system for generating and using vehicle pricing models | |
| US11250031B2 (en) | Method of predicting a traffic behaviour in a road system | |
| CN110750571A (en) | Port berth data mining method, device, equipment and storage medium | |
| CN111078760B (en) | Goods source searching method, device, equipment and storage medium | |
| CN111814056A (en) | Supplier recommendation method based on information processing and related equipment | |
| CN105488599B (en) | Method and device for predicting article popularity | |
| US20240046164A1 (en) | Active notification using transportation service prediction | |
| CN112434260A (en) | Road traffic state detection method and device, storage medium and terminal | |
| US20180038703A1 (en) | System and method for recommending an optimal route | |
| CN114581167A (en) | Service abnormity identification method and device, storage medium and electronic equipment | |
| US20190370349A1 (en) | Automatic detection of point of interest change using cohort analysis | |
| US11748424B2 (en) | Visiting destination prediction device and visiting destination prediction method | |
| US20160292610A1 (en) | Method and device for real time prediction of timely delivery of telecom service orders | |
| JPH09233700A (en) | Method of evaluating reliability on estimation of day maximum demand power | |
| CN111008729A (en) | Migration prediction method and device | |
| CN114331568B (en) | Commercial vehicle market segment identification method, equipment and medium based on Internet of vehicles | |
| US20220228886A1 (en) | Missing map data identification system | |
| CN112051843B (en) | Path planning method and device based on order prediction, robot and storage medium | |
| CN114418236A (en) | Information prediction method, information prediction device, storage medium and electronic equipment | |
| CN114186090A (en) | Intelligent quality inspection method and system for image annotation data | |
| US20170160747A1 (en) | Map generation based on raw stereo vision based measurements |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |