CN114936822A - Vehicle and goods matching method, device, equipment and storage medium - Google Patents
Vehicle and goods matching method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114936822A CN114936822A CN202210615097.1A CN202210615097A CN114936822A CN 114936822 A CN114936822 A CN 114936822A CN 202210615097 A CN202210615097 A CN 202210615097A CN 114936822 A CN114936822 A CN 114936822A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- goods
- cargo
- information
- matching
- 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.)
- Pending
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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Biodiversity & Conservation Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the field of artificial intelligence, and discloses a vehicle and goods matching method, device, equipment and storage medium, which are used for improving the accuracy of vehicle and goods matching. The vehicle and goods matching method comprises the following steps: acquiring initial vehicle and goods information to be processed; extracting characteristics of the initial vehicle and goods information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector; acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods and the vehicles to obtain a plurality of goods and vehicles matching results; respectively inputting the plurality of vehicle and goods matching results into a preset click rate prediction model to predict click rates, and obtaining a plurality of click rate prediction values; and generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a vehicle and goods matching method, device, equipment and storage medium.
Background
The freight industry in China is rapidly developed, the freight industry is rapidly transformed from the traditional industry to the modern logistics and becomes the inevitable trend of the current industry development, the current freight market is usually the trading mode of off-line telephone communication, oral commitment or network waybill, and the trading modes have the big problems that the information of the freight market is asymmetric, the standardization degree of goods sources and vehicle information in the industry is low, and the trading fall exists between the vacant information of the vehicles and the information of the order under the clients. Therefore, the efficiency of finding goods by vehicles and finding vehicles by goods is low, the transaction flow is long, the loading rate/full load rate of the vehicles is reduced, and the logistics cost and the labor cost of a company are greatly increased; and the non-standardized factors in the service scene are more, and the abstract modeling is difficult.
When matching the carrier vehicles and the goods to be carried, the network platform generally performs matching based on the number of the carrier vehicles and the number of goods orders. In this way, the matching accuracy between the transporting vehicle and the goods to be transported is low and the actual transportation requirements cannot be met well because the situation that the transporting vehicle cannot load the goods to be transported, such as the volume of the goods, the goods stacking requirement, the packaging mode, the loading and unloading mode, the transportation route and the like, are not considered.
Disclosure of Invention
The invention provides a vehicle and goods matching method, device, equipment and storage medium, which are used for improving the accuracy of vehicle and goods matching.
The invention provides a vehicle and goods matching method in a first aspect, which comprises the following steps: acquiring initial vehicle and goods information to be processed from a preset database, wherein the initial vehicle and goods information comprises: vehicle information, historical cargo information, and historical environmental information; extracting the characteristics of the initial vehicle and goods information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector; acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods and the vehicles to obtain a plurality of goods and vehicles matching results; respectively inputting the plurality of vehicle and goods matching results into a preset click rate prediction model to predict click rates, and obtaining a plurality of click rate prediction values; and generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence.
Optionally, in a first implementation manner of the first aspect of the present invention, the extracting features of the initial vehicle and cargo information to obtain feature preference information, and performing vector code conversion on the feature preference information to obtain a feature vector includes: determining a plurality of cargo characteristics corresponding to the initial vehicle cargo information according to the historical cargo information and the historical environment information; acquiring historical carrying times corresponding to each cargo feature from the historical cargo information; comparing the historical carrying times corresponding to each cargo feature with a preset time threshold value to generate a comparison result, and determining cargo preference information according to the comparison result; generating vehicle preference information according to the vehicle information, and taking the vehicle preference information and the cargo preference information as characteristic preference information; and carrying out vector coding conversion on the characteristic preference information to obtain a characteristic vector.
Optionally, in a second implementation manner of the first aspect of the present invention, the performing vector code conversion on the feature preference information to obtain a feature vector includes: extracting a plurality of cargo features and a plurality of vehicle features in the feature preference information; respectively carrying out vector coding and feature cross processing on the plurality of cargo features and the plurality of vehicle features according to a preset rule to obtain a plurality of vector element values; and generating a feature vector corresponding to the feature preference information according to the vector element values.
Optionally, in a third implementation manner of the first aspect of the present invention, the performing vector coding and feature intersection processing on the multiple cargo features and the multiple vehicle features respectively according to a preset rule to obtain multiple vector element values includes: extracting characteristic numerical values of the plurality of cargo characteristics to obtain a plurality of cargo characteristic numerical values; carrying out feature coding on the plurality of vehicle features to obtain a plurality of vehicle code numerical values; generating a plurality of vector element values based on the plurality of cargo characteristic values and the plurality of vehicle code values.
Optionally, in a fourth implementation manner of the first aspect of the present invention, the obtaining information of goods to be matched, and inputting the information of goods to be matched and the feature vector into a preset regression prediction model to perform vehicle-to-goods matching, so as to obtain a plurality of vehicle-to-goods matching results, includes: acquiring information of goods to be matched, and determining weight parameters corresponding to the goods to be matched according to the information of the goods to be matched; inputting the characteristic vector and the weight parameter into a preset regression prediction model, and calculating the vehicle and cargo matching probability of the cargo to be carried through the regression prediction model to obtain a plurality of vehicle and cargo matching probabilities; and generating a vehicle and cargo matching result corresponding to each vehicle and cargo matching probability according to the plurality of vehicle and cargo matching probabilities.
Optionally, in a fifth implementation manner of the first aspect of the present invention, the step of inputting the multiple vehicle-cargo matching results into a preset click rate prediction model for click rate prediction to obtain multiple click rate prediction values includes: respectively inputting the vehicle and goods matching results into a preset click rate estimation model; calculating the correlation degree of each vehicle and goods matching result through a neural network in the click rate estimation model; and generating the click rate predicted value of each vehicle and goods matching result according to the relevance of each vehicle and goods matching result to obtain a plurality of click rate predicted values.
Optionally, in a sixth implementation manner of the first aspect of the present invention, the generating a vehicle recommendation sequence according to the multiple click rate predicted values, and performing vehicle-to-cargo recommendation on the to-be-matched cargo information according to the vehicle recommendation sequence includes: sorting the candidate vehicles according to the click rate predicted values to obtain vehicle recommendation sorting; taking the candidate vehicle with the largest click rate predicted value in the vehicle recommendation sorting as a target vehicle; recommending the cargo information to be matched to the target vehicle.
A second aspect of the present invention provides a vehicle-cargo matching device, including: the acquisition module is used for acquiring initial vehicle and goods information to be processed from a preset database, wherein the initial vehicle and goods information comprises: vehicle information, historical cargo information, and historical environmental information; the characteristic extraction module is used for extracting the characteristics of the initial vehicle and cargo information to obtain characteristic preference information and performing vector coding conversion on the characteristic preference information to obtain a characteristic vector; the matching module is used for acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods to obtain a plurality of goods matching results; the prediction module is used for inputting the plurality of vehicle and goods matching results into a preset click rate prediction model to predict click rates so as to obtain a plurality of click rate prediction values; and the recommendation module is used for generating vehicle recommendation sequencing according to the plurality of click rate predicted values and recommending the goods to be matched according to the vehicle recommendation sequencing.
Optionally, in a first implementation manner of the second aspect of the present invention, the feature extraction module further includes: the processing unit is used for determining a plurality of cargo characteristics corresponding to the initial vehicle-cargo information according to the historical cargo information and the historical environment information; acquiring historical shipper times corresponding to each cargo characteristic from the historical cargo information; comparing the historical carrying times corresponding to each cargo feature with a preset time threshold value to generate a comparison result, and determining cargo preference information according to the comparison result; generating vehicle preference information according to the vehicle information, and taking the vehicle preference information and the cargo preference information as characteristic preference information; and the conversion unit is used for carrying out vector coding conversion on the characteristic preference information to obtain a characteristic vector.
Optionally, in a second implementation manner of the second aspect of the present invention, the conversion unit further includes: the extracting subunit is used for extracting a plurality of cargo features and a plurality of vehicle features in the feature preference information; the coding subunit is used for respectively carrying out vector coding and feature cross processing on the plurality of cargo features and the plurality of vehicle features according to a preset rule to obtain a plurality of vector element values; and the generating subunit is used for generating the feature vector corresponding to the feature preference information according to the vector element values.
Optionally, in a third implementation manner of the second aspect of the present invention, the coding subunit is specifically configured to: extracting characteristic numerical values of the plurality of cargo characteristics to obtain a plurality of cargo characteristic numerical values; carrying out feature coding on the plurality of vehicle features to obtain a plurality of vehicle code numerical values; generating a plurality of vector element values based on the plurality of cargo characteristic values and the plurality of vehicle code values.
Optionally, in a fourth implementation manner of the second aspect of the present invention, the matching module is specifically configured to: acquiring information of goods to be matched, and determining weight parameters corresponding to the goods to be matched according to the information of the goods to be matched; inputting the characteristic vector and the weight parameter into a preset regression prediction model, and calculating the vehicle and cargo matching probability of the cargo to be carried through the regression prediction model to obtain a plurality of vehicle and cargo matching probabilities; and generating a vehicle and cargo matching result corresponding to each vehicle and cargo matching probability according to the plurality of vehicle and cargo matching probabilities.
Optionally, in a fifth implementation manner of the second aspect of the present invention, the prediction module is specifically configured to: respectively inputting the vehicle and goods matching results into a preset click rate estimation model; calculating the correlation degree of each vehicle and goods matching result through a neural network in the click rate estimation model; and generating the click rate predicted value of each vehicle and goods matching result according to the relevance of each vehicle and goods matching result to obtain a plurality of click rate predicted values.
Optionally, in a sixth implementation manner of the second aspect of the present invention, the recommending module is specifically configured to: sorting the candidate vehicles according to the size of the click rate predicted values to obtain vehicle recommendation sorting; taking the candidate vehicle with the largest click rate predicted value in the vehicle recommendation sequence as a target vehicle; recommending the cargo information to be matched to the target vehicle.
A third aspect of the present invention provides a vehicle-cargo matching apparatus comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor calls the instructions in the memory to enable the vehicle-cargo matching device to execute the vehicle-cargo matching method.
A fourth aspect of the present invention provides a computer-readable storage medium having stored therein instructions, which, when run on a computer, cause the computer to execute the above-mentioned vehicle-cargo matching method.
In the technical scheme provided by the invention, initial vehicle and goods information to be processed is obtained from a preset database, wherein the initial vehicle and goods information comprises the following components: vehicle information, historical cargo information, and historical environmental information; extracting the characteristics of the initial vehicle and goods information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector; acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods and the vehicles to obtain a plurality of goods and vehicles matching results; respectively inputting the plurality of vehicle and goods matching results into a preset click rate prediction model to predict click rates, and obtaining a plurality of click rate prediction values; and generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence. The method and the device train the iterative model together with the characteristic preference information and the off-line result of the prediction regression model, recommend the cargo source to the target vehicle according to the click rate sequence estimated by the click rate estimation model, and improve the accuracy rate of vehicle-cargo matching.
Drawings
FIG. 1 is a diagram of an embodiment of a vehicle and cargo matching method according to an embodiment of the invention;
FIG. 2 is a schematic diagram of another embodiment of the vehicle-cargo matching method in the embodiment of the invention;
FIG. 3 is a schematic diagram of an embodiment of a truck-cargo matching device according to the present invention;
FIG. 4 is a schematic diagram of another embodiment of the truck-cargo matching device in the embodiment of the invention;
fig. 5 is a schematic diagram of an embodiment of the vehicle-cargo matching device in the embodiment of the invention.
Detailed Description
The embodiment of the invention provides a vehicle and goods matching method, device, equipment and storage medium, which are used for improving the accuracy of vehicle and goods matching. The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be implemented in other sequences than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
For convenience of understanding, a detailed flow of an embodiment of the present invention is described below, and referring to fig. 1, a first embodiment of a vehicle-cargo matching method in an embodiment of the present invention includes:
101. acquiring initial vehicle and goods information to be processed from a preset database, wherein the initial vehicle and goods information comprises: vehicle information, historical cargo information, and historical environmental information;
it is to be understood that the executing subject of the present invention may be a vehicle and goods matching device, and may also be a terminal or a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.
It should be noted that the vehicle information may be attribute information of the vehicle, the vehicle information may include vehicle length information, vehicle type information, approved load information, and the like of the target vehicle, the historical cargo information may be regarded as cargo information that the vehicle has carried in a completed historical carrying order, the historical cargo information may include information about a loading and unloading manner of the historical carrying cargo, a departure place city, a destination city, and the like, the server may obtain initial vehicle cargo information of the vehicle through a preset database, and specifically, the server obtains initial vehicle cargo information to be processed from the preset database, where the initial vehicle cargo information includes: vehicle information, historical cargo information, and historical environmental information.
102. Extracting characteristics of the initial vehicle and goods information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector;
specifically, the characteristic preference information may be cargo characteristic information acceptable to the vehicle. For example, the target vehicle may accept cargo characteristics such as dirty vehicles, short haul, long haul, etc. After the server acquires the initial vehicle and cargo information to be processed, the characteristic preference information of the vehicle for carrying cargo can be determined from the initial vehicle and cargo information. For example, the loaded goods a have goods features that may make the vehicle dirty, and at this time, the dirty vehicle features may be regarded as the goods features preferred by the target vehicle, and the server performs vector code conversion according to the feature preference information to obtain the feature vector.
103. Acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods to obtain a plurality of goods matching results;
after the server acquires the characteristic preference information of the vehicle and the vehicle information, the characteristic preference information and the vehicle information can be combined to determine whether the vehicle is matched with goods to be carried. For example, the server may first determine whether a dirty vehicle characteristic exists for the cargo to be shipped, and if not, may further determine whether the characteristic preference information indicates a vehicle that is capable of receiving the dirty vehicle characteristic. If so, the determined vehicle can be considered to be matched with the goods to be carried. Then, the owner corresponding to the vehicle and the owner corresponding to the goods to be carried can be informed so as to transport the goods to be carried, and the server determines the matching result between the vehicle and the goods to be matched based on the characteristic preference information and the vehicle information of the vehicle. In this way, the matching accuracy between the vehicle and the goods to be carried can be improved, so as to better meet the actual freight requirement.
Optionally, when the information of the goods to be matched and the feature vector are input into the preset regression prediction model for vehicle-to-goods matching, the feature preference information of the vehicle may be analyzed, for example, when the goods to be carried are non-stolen vehicle features, the server matches nearby vehicles, and if a vehicle of a non-stolen vehicle feature type exists nearby at this time, it may be considered that the current vehicle is determined to be matched with the goods to be carried.
If vehicles with non-dirty vehicle characteristic types do not exist nearby, scanning and judging the vehicles within a preset distance of goods to be carried at the moment to obtain a judgment result, if vehicles which are matched with the non-dirty vehicle characteristic types and are not matched exist simultaneously, performing path analysis on the vehicles which are matched with the non-dirty vehicle characteristic types to obtain a goods receiving path set, performing goods receiving time analysis according to the current road congestion degree and weather conditions to obtain a plurality of corresponding goods receiving times which are used as a first goods receiving time set, performing cleaning time and path analysis on the vehicles which are not matched with the non-dirty vehicle characteristic types simultaneously to obtain cleaning time of each vehicle and goods receiving time of each vehicle in the plurality of vehicles which are not matched with the non-dirty vehicle characteristic types, summing the vehicle cleaning time and the vehicle goods receiving time corresponding to each vehicle to obtain a second goods receiving time set, and then the server merges the first goods receiving time set and the second goods receiving time set, compares the first goods receiving time set and the second goods receiving time set, takes the short goods receiving time as the target goods receiving time, takes the vehicle corresponding to the target goods receiving time as the target vehicle, and at the moment, the determined vehicle can be considered to be matched with the goods to be carried.
104. Respectively inputting the plurality of vehicle and goods matching results into a preset click rate estimation model to predict click rates, so as to obtain a plurality of click rate predicted values;
it should be noted that the click-through rate estimation model includes a first level, a second level, a third level and a fourth level, collects the matching result of the user's vehicle and goods, inputting the matching result of the user's vehicle and goods into the click rate estimation model for matrixing to obtain a user sparse feature matrix, inputting the user sparse feature matrix, converting a user sparse feature matrix into a dense embedded matrix through a first level of a click rate estimation model, inputting the dense embedded matrix into a second level, learning low-order interactive features to obtain an interactive feature relationship between the low-order features, using the interactive feature relationship between the low-order features as the input of a third level containing a residual error network to learn high-order interactive features to obtain an interactive feature relationship between the high-order features, and optimizing the interactive feature relation among the high-order features, and finally outputting a click rate predicted value by a fourth level.
105. And generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence.
Specifically, the server recommends the goods sources to the appropriate drivers according to the click rate sequence estimated by the click rate prediction model, so that accurate pairing is realized between a large number of goods sources and the drivers on the platform, the waiting time of both vehicles and goods is shortened, and the transaction efficiency is improved. The asymmetry of information of supply and demand parties is reduced and eliminated, and the transportation cost is reduced; and the enterprise efficiency is comprehensively improved.
In the embodiment of the invention, initial vehicle and goods information to be processed is obtained from a preset database, wherein the initial vehicle and goods information comprises the following components: vehicle information, historical cargo information, and historical environmental information; extracting the characteristics of the initial vehicle and cargo information to obtain characteristic preference information, and performing vector coding conversion on the characteristic preference information to obtain a characteristic vector; acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods and the vehicles to obtain a plurality of goods and vehicles matching results; respectively inputting the plurality of vehicle and goods matching results into a preset click rate estimation model to predict click rates, so as to obtain a plurality of click rate predicted values; and generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence. The method and the device train the iterative model together with the characteristic preference information and the off-line result of the prediction regression model, recommend the cargo source to the target vehicle according to the click rate sequence estimated by the click rate estimation model, and improve the accuracy rate of vehicle-cargo matching.
Referring to fig. 2, a second embodiment of the vehicle-cargo matching method according to the embodiment of the present invention includes:
201. acquiring initial vehicle and goods information to be processed from a preset database, wherein the initial vehicle and goods information comprises: vehicle information, historical cargo information, and historical environmental information;
specifically, the initial vehicle cargo information includes historical order information, and the server acquires historical carrying information of the vehicle according to the historical order information. Here, the historical order information may include personal information such as name, sex, contact address, etc. of a target driver corresponding to the vehicle, vehicle information of the vehicle and historical source information corresponding to the historical order information, and status information of the historical order information. The status information here can be used to characterize whether the historical order was cancelled, committed, completed, etc.
202. Determining a plurality of cargo features corresponding to the initial vehicle-cargo information according to the historical cargo information and the historical environment information;
specifically, the server may obtain historical cargo information and historical environmental information, which may correspond to one historical shipments, respectively. The server may then determine the cargo characteristics for each of the historical shipments. For example, for historically transported cargo a carried by vehicle a, the server may determine characteristics of the historically transported cargo a such as dirty-car characteristics, short-range transportation characteristics, etc.; for historical shipper items B, the server may determine characteristics of the historical shipper items B such as dirty car characteristics, long distance transport characteristics, etc.; for historical shipments C, the server may determine characteristics of the historical shipments C, such as characteristics pertaining to the vehicle, dirty vehicle characteristics, short-distance transportation characteristics, etc., and the server then determines a plurality of cargo characteristics corresponding to the initial vehicle cargo information based on the historical cargo information and the historical environmental information.
203. Acquiring historical carrying times corresponding to each cargo feature from historical cargo information;
specifically, after determining the cargo characteristics of each piece of historical cargo information, the server may count, for each cargo characteristic, the number of times the vehicle has taken a cargo with the cargo characteristic. For example, after determining the cargo characteristics of the historical transported cargo a, the historical transported cargo B, and the historical transported cargo C, the server may count the historical transported number of the cargo having the cargo characteristic with respect to the dirty vehicle characteristic, the short-distance transportation characteristic, the long-distance transportation characteristic, and the characteristic belonging to the vehicle. For example, the historical carrying times of goods with dirty vehicle characteristics can be counted to be 3 times; the historical carrying times of the goods with the short-distance transportation characteristics is 2 times; the historical carrying times of the goods with the long-distance transportation characteristics is 1; the number of historical shipments in which goods belonging to the characteristics of the vehicle exist is 1.
204. Comparing the historical carrying times corresponding to each cargo feature with a preset time threshold value to generate a comparison result, and determining cargo preference information according to the comparison result;
specifically, after the server counts the historical shipping times corresponding to each cargo feature, the preference feature can be further determined according to the historical shipping times. Specifically, the determination may be made based on whether or not the historical shipment count exceeds a count threshold. The number threshold may include a preset threshold such as 10 times, 20 times, etc., or may include a threshold determined according to a certain rule according to the total number of times of carrying the vehicle. The certain rule may include that the threshold of the number of times is half of the total number of times of delivery, for example, when the threshold of the number of times is determined to be 2 times, the dirty vehicle feature may be determined as a preference feature corresponding to the vehicle a.
205. Generating vehicle preference information according to the vehicle information, and taking the vehicle preference information and the cargo preference information as characteristic preference information;
specifically, after the server acquires the characteristic preference information of the vehicle and the vehicle information, whether the target vehicle is matched with the cargo information to be matched or not can be determined by combining the characteristic preference information and the vehicle information. For example, the server may first determine whether a dirty vehicle characteristic exists for the cargo to be shipped, and if not, may further determine whether the characteristic preference information indicates a vehicle that is capable of receiving the dirty vehicle characteristic. And if so, determining that the determined vehicle is matched with the cargo information to be matched. The server can inform the owner of the target vehicle and the owner of the goods to be transported, so that the goods to be transported can be transported conveniently, the vehicle preference information is generated according to the vehicle information, and the vehicle preference information and the goods preference information are used as feature preference information.
206. Carrying out vector coding conversion on the characteristic preference information to obtain a characteristic vector;
specifically, the server extracts a plurality of cargo features and a plurality of vehicle features in the feature preference information; the server respectively carries out vector coding and feature cross processing on the plurality of cargo features and the plurality of vehicle features according to a preset rule to obtain a plurality of vector element values; and the server generates a feature vector corresponding to the feature preference information according to the vector element values.
The server firstly judges whether the plurality of cargo features and the plurality of vehicle features can be quantified by numerical values, and in some application scenarios, the server can determine the vector of the cargo features by judging whether the cargo features can be quantified by numerical values. For example, when the weight of the cargo is 10 tons and the transport distance is 50 km, it may be considered that the weight characteristic and the distance characteristic can be numerically quantified, when the cargo belongs to the vehicle category, the characteristic belonging to the vehicle cannot be numerically quantified, if the plurality of cargo characteristics and the plurality of vehicle characteristics can be numerically quantified, the characteristic value of the cargo characteristic is determined as the characteristic value of the vector, if it is determined that the cargo characteristic can be numerically quantified, the cargo characteristic may be vectorized, and the characteristic value of the cargo characteristic may be determined as the characteristic value of the vector. For example, after it is determined that a weight feature can be numerically quantified, the weight feature may be vectorized, and a eigenvalue "10" of the weight feature may be determined as an eigenvalue of the vector. It should be noted that, for the weight unit, it may be stated at an adaptive position, if the cargo feature cannot be numerically quantized, the feature value of the vector corresponding to the cargo feature is determined according to a preset encoding rule, if it is determined that the cargo feature cannot be numerically quantized, the cargo feature may also be vectorized, and the feature value of the vector corresponding to the cargo feature may be determined according to a preset encoding. The preset encoding rule herein may include a one-hot encoding rule. For example, after determining that the features belonging to the vehicle cannot be numerically quantized, the features belonging to the vehicle may be vectorized, and a value "1" may be determined as a feature value of a vector based on the one-hot code, and then the server performs vector coding and feature cross processing on the plurality of cargo features and the plurality of vehicle features respectively according to a preset rule to obtain a plurality of vector element values; and the server generates a feature vector corresponding to the feature preference information according to the vector element values.
Optionally, the server performs vector coding and feature cross processing on the plurality of cargo features and the plurality of vehicle features according to a preset rule, to obtain a plurality of vector element values, and the method may include: the server extracts the characteristic numerical values of the plurality of goods characteristics to obtain a plurality of goods characteristic numerical values; the server carries out feature coding on the plurality of vehicle features to obtain a plurality of vehicle coding numerical values; the server generates a plurality of vector element values according to the plurality of cargo characteristic values and the plurality of vehicle code values.
If the goods characteristics cannot be quantified numerically, the goods characteristics can be vectorized, and the characteristic value of the vector corresponding to the goods characteristics can be determined according to the preset code. The preset encoding rule herein may include a one-hot encoding rule. For example, after it is determined that the feature belonging to the vehicle cannot be numerically quantized, the feature belonging to the vehicle may be vectorized, and a value "1" may be determined as a feature value of the vector based on the unique hot code, and then feature-coding is performed on a plurality of vehicle features by the server to obtain a plurality of vehicle coded values; the server generates a plurality of vector element values according to the plurality of cargo characteristic values and the plurality of vehicle code values.
207. Acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods and the vehicles to obtain a plurality of goods and vehicles matching results;
specifically, the server acquires information of goods to be matched, and determines a weight parameter corresponding to the goods to be matched according to the information of the goods to be matched; the server inputs the characteristic vectors and the weight parameters into a preset regression prediction model, and the vehicle and cargo matching probability of the cargo to be carried is calculated through the regression prediction model to obtain a plurality of vehicle and cargo matching probabilities; and the server generates a vehicle and cargo matching result corresponding to each vehicle and cargo matching probability according to the plurality of vehicle and cargo matching probabilities.
Optionally, the vehicle characteristics corresponding to the information of the goods to be matched and the characteristics of the goods to be carried corresponding to the goods to be carried are vectorized respectively to obtain the vehicle characteristic vector and the characteristic vector of the goods to be carried, and the vehicle characteristics and the characteristics of the goods to be carried can be vectorized to determine the matching result more conveniently. In these application scenarios, for example, the one-hot coding may be adopted for vectorization, and the matching result is determined according to the vector corresponding to the preference feature, the vehicle feature vector, and the feature vector of the goods to be carried. After the vehicle characteristic vector and the characteristic vector of the goods to be carried are determined, a plurality of vehicle and goods matching probabilities can be determined by combining the vectorized vectors of the preference characteristics, and vehicle and goods matching results corresponding to the vehicle and goods matching probabilities are generated. The characteristic vectors and the weight parameters can be input into a preset regression prediction model, and the vehicle and cargo matching probability of the cargo to be carried is calculated through the regression prediction model to obtain a plurality of vehicle and cargo matching probabilities; and the server generates a vehicle and goods matching result corresponding to each vehicle and goods matching probability according to the plurality of vehicle and goods matching probabilities.
208. Respectively inputting the plurality of vehicle and goods matching results into a preset click rate prediction model to predict click rates, and obtaining a plurality of click rate prediction values;
specifically, the server respectively inputs a plurality of vehicle and goods matching results into a preset click rate estimation model; the server calculates the relevance of each vehicle and goods matching result through a neural network in the click rate estimation model; and the server generates the click rate predicted value of each vehicle and goods matching result according to the relevance of each vehicle and goods matching result to obtain a plurality of click rate predicted values.
The learning feature interaction is very important for sparse and huge user behavior features, the problem of learning feature interaction is more or less considered in many existing click rate estimation models, and the importance of learning feature interaction can be reflected in the experimental result. The characteristic interaction comprises low-order characteristic interaction and high-order characteristic interaction, in the technical scheme, for the click rate estimation model, the interaction characteristic relationship between low-order interaction characteristics and the interaction characteristic relationship between high-order interaction characteristics are considered, the correlation degree of each vehicle and goods matching result is calculated through a neural network in the click rate estimation model, the server generates the click rate estimation value of each vehicle and goods matching result according to the correlation degree of each vehicle and goods matching result, a plurality of click rate estimation values are obtained, compared with other traditional estimation models, the click rate estimation model can be obviously seen, meanwhile, the low-order interaction characteristics and the high-order interaction characteristics are considered, and the prediction accuracy of the click rate estimation model can be obviously improved.
209. And generating vehicle recommendation sorting according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sorting.
Specifically, the server sorts a plurality of candidate vehicles according to the size of a plurality of click rate predicted values to obtain vehicle recommendation sorting; taking the candidate vehicle with the largest click rate predicted value in the vehicle recommendation sequence as a target vehicle; and recommending the information of the goods to be matched to the target vehicle.
The server recommends the goods sources to the appropriate drivers according to the click rate sequence estimated by the click rate estimation model, so that accurate pairing is realized between a large number of goods sources and the drivers on the platform, waiting time of both vehicles and goods is shortened, and the transaction efficiency is improved. The information asymmetry of the supply and demand parties is reduced and eliminated, the capacity cost is reduced, the candidate vehicles are ranked according to the predicted values of the click rate, the vehicle recommendation ranking is obtained, and the candidate vehicle with the largest predicted value of the click rate in the vehicle recommendation ranking is used as the target vehicle; and recommending the cargo information to be matched to the target vehicle.
In the embodiment of the invention, the characteristic preference information and the off-line result of the prediction regression model are trained together to form the iterative model, the information of the goods to be matched is recommended to the target vehicle according to the click rate sequence estimated by the click rate estimation model, the accuracy of matching the vehicle and the goods is improved, the correlation of each vehicle and goods matching result is calculated through a neural network in the click rate estimation model, the server generates the click rate estimation value of each vehicle and goods matching result according to the correlation of each vehicle and goods matching result to obtain a plurality of click rate estimation values, and compared with other traditional prediction models, the prediction accuracy of the click rate estimation model is obviously improved by considering the low-order interaction characteristic and the high-order interaction characteristic.
In the above description of the vehicle and cargo matching method in the embodiment of the present invention, referring to fig. 3, the vehicle and cargo matching device in the embodiment of the present invention is described below, and the first embodiment of the vehicle and cargo matching device in the embodiment of the present invention includes:
an obtaining module 301, configured to obtain initial vehicle and cargo information to be processed from a preset database, where the initial vehicle and cargo information includes: vehicle information, historical cargo information, and historical environmental information;
a feature extraction module 302, configured to perform feature extraction on the initial vehicle and cargo information to obtain feature preference information, and perform vector coding conversion on the feature preference information to obtain a feature vector;
the matching module 303 is configured to obtain information of goods to be matched, and input the information of goods to be matched and the feature vector into a preset regression prediction model to perform vehicle-goods matching, so as to obtain a plurality of vehicle-goods matching results;
the prediction module 304 is used for inputting the plurality of vehicle and goods matching results into a preset click rate prediction model for click rate prediction to obtain a plurality of click rate prediction values;
and the recommending module 305 is configured to generate a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommend the goods to be matched according to the vehicle recommendation sequence.
Further, the server stores the vehicle recommendation ranking in the blockchain database, which is not limited herein.
In the embodiment of the invention, initial vehicle and goods information to be processed is obtained from a preset database, wherein the initial vehicle and goods information comprises the following components: vehicle information, historical cargo information, and historical environmental information; extracting the characteristics of the initial vehicle and goods information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector; acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods and the vehicles to obtain a plurality of goods and vehicles matching results; respectively inputting the plurality of vehicle and goods matching results into a preset click rate prediction model to predict click rates, and obtaining a plurality of click rate prediction values; and generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence. The invention trains the iterative model together with the characteristic preference information and the off-line result of the prediction regression model, recommends the cargo information to be matched to the target vehicle according to the click rate sequence estimated by the click rate estimation model, and improves the accuracy of vehicle-cargo matching.
Referring to fig. 4, a second embodiment of the vehicle-cargo matching device in the embodiment of the present invention includes:
an obtaining module 301, configured to obtain initial vehicle and cargo information to be processed from a preset database, where the initial vehicle and cargo information includes: vehicle information, historical cargo information, and historical environmental information;
the feature extraction module 302 is configured to perform feature extraction on the initial vehicle and cargo information to obtain feature preference information, and perform vector coding conversion on the feature preference information to obtain a feature vector;
the matching module 303 is configured to obtain information of goods to be matched, and input the information of the goods to be matched and the feature vector into a preset regression prediction model to perform vehicle-goods matching, so as to obtain a plurality of vehicle-goods matching results;
the prediction module 304 is used for inputting the plurality of vehicle and goods matching results into a preset click rate prediction model for click rate prediction to obtain a plurality of click rate prediction values;
and the recommending module 305 is configured to generate a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommend the goods to be matched according to the vehicle recommendation sequence.
Optionally, the feature extraction module 302 further includes:
the processing unit 3021 is configured to determine, according to the historical cargo information and the historical environment information, a plurality of cargo characteristics corresponding to the initial vehicle-cargo information; acquiring historical carrying times corresponding to each cargo feature from the historical cargo information; comparing the historical carrying times corresponding to each cargo feature with a preset time threshold value to generate a comparison result, and determining cargo preference information according to the comparison result; generating vehicle preference information according to the vehicle information, and taking the vehicle preference information and the cargo preference information as characteristic preference information;
a converting unit 3022, configured to perform vector coding conversion on the feature preference information to obtain a feature vector.
Optionally, the conversion unit 3022 further includes:
an extraction subunit 30221 configured to extract a plurality of cargo features and a plurality of vehicle features in the feature preference information;
the encoding subunit 30222 is configured to perform vector encoding and feature intersection processing on the cargo features and the vehicle features according to a preset rule, respectively, to obtain a plurality of vector element values;
a generating subunit 30223, configured to generate a feature vector corresponding to the feature preference information according to the plurality of vector element values.
Optionally, the coding subunit 30222 is specifically configured to: extracting characteristic numerical values of the plurality of cargo characteristics to obtain a plurality of cargo characteristic numerical values; carrying out feature coding on the plurality of vehicle features to obtain a plurality of vehicle coding numerical values; generating a plurality of vector element values based on the plurality of cargo characteristic values and the plurality of vehicle code values.
Optionally, the matching module 303 is specifically configured to: acquiring information of goods to be matched, and determining weight parameters corresponding to the goods to be matched according to the information of the goods to be matched; inputting the characteristic vector and the weight parameter into a preset regression prediction model, and calculating the vehicle and cargo matching probability of the cargo to be carried through the regression prediction model to obtain a plurality of vehicle and cargo matching probabilities; and generating a vehicle and cargo matching result corresponding to each vehicle and cargo matching probability according to the plurality of vehicle and cargo matching probabilities.
Optionally, the prediction module 304 is specifically configured to: respectively inputting the vehicle and goods matching results into a preset click rate estimation model; calculating the correlation degree of each vehicle and goods matching result through a neural network in the click rate estimation model; and generating the click rate predicted value of each vehicle and goods matching result according to the relevance of each vehicle and goods matching result to obtain a plurality of click rate predicted values.
Optionally, the recommending module 305 is specifically configured to: sorting the candidate vehicles according to the size of the click rate predicted values to obtain vehicle recommendation sorting; taking the candidate vehicle with the largest click rate predicted value in the vehicle recommendation sorting as a target vehicle; recommending the cargo information to be matched to the target vehicle.
Further, the server stores the vehicle recommendation ranking in the blockchain database, which is not limited herein.
In the embodiment of the invention, initial vehicle and goods information to be processed is obtained from a preset database, wherein the initial vehicle and goods information comprises the following components: vehicle information, historical cargo information, and historical environmental information; extracting the characteristics of the initial vehicle and goods information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector; acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods and the vehicles to obtain a plurality of goods and vehicles matching results; respectively inputting the plurality of vehicle and goods matching results into a preset click rate prediction model to predict click rates, and obtaining a plurality of click rate prediction values; and generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence. The invention trains the iterative model together with the characteristic preference information and the off-line result of the prediction regression model, recommends the cargo information to be matched to the target vehicle according to the click rate sequence estimated by the click rate estimation model, and improves the accuracy of vehicle-cargo matching.
Fig. 3 and fig. 4 describe the vehicle-cargo matching device in the embodiment of the present invention in detail from the perspective of a modular functional entity, and the vehicle-cargo matching device in the embodiment of the present invention is described in detail from the perspective of hardware processing.
Fig. 5 is a schematic structural diagram of a vehicle and cargo matching device according to an embodiment of the present invention, where the vehicle and cargo matching device 500 may have a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) storing applications 533 or data 532. Memory 520 and storage media 530 may be, among other things, transient or persistent storage. The program stored in the storage medium 530 may include one or more modules (not shown), each of which may include a series of instructions operating on the truck-cargo matching apparatus 500. Further, the processor 510 may be configured to communicate with the storage medium 530 to execute a series of instruction operations in the storage medium 530 on the truck-cargo matching device 500.
The truck-to-cargo matching device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input-output interfaces 560, and/or one or more operating systems 531, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, and the like. Those skilled in the art will appreciate that the configuration of the vehicle cargo matching apparatus shown in fig. 5 does not constitute a limitation of the vehicle cargo matching apparatus and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.
The invention further provides vehicle and goods matching equipment which comprises a memory and a processor, wherein computer readable instructions are stored in the memory, and when the computer readable instructions are executed by the processor, the processor executes the steps of the vehicle and goods matching method in the embodiments.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and which may also be a volatile computer readable storage medium, having stored therein instructions, which, when executed on a computer, cause the computer to perform the steps of the vehicle-to-cargo matching method.
Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the blockchain node, and the like.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media that can store program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a random access memory, a magnetic disk, or an optical disk.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. The vehicle and goods matching method is characterized by comprising the following steps:
acquiring initial vehicle and cargo information to be processed from a preset database, wherein the initial vehicle and cargo information comprises: vehicle information, historical cargo information, and historical environmental information;
extracting the characteristics of the initial vehicle and goods information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector;
acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for vehicle and goods matching to obtain a plurality of vehicle and goods matching results;
respectively inputting the plurality of vehicle and goods matching results into a preset click rate estimation model to predict click rates, so as to obtain a plurality of click rate predicted values;
and generating a vehicle recommendation sequence according to the plurality of click rate predicted values, and recommending the goods to be matched according to the vehicle recommendation sequence.
2. The vehicle and cargo matching method according to claim 1, wherein the extracting the characteristics of the initial vehicle and cargo information to obtain characteristic preference information, and performing vector code conversion on the characteristic preference information to obtain a characteristic vector comprises:
determining a plurality of cargo characteristics corresponding to the initial vehicle cargo information according to the historical cargo information and the historical environment information;
acquiring historical shipper times corresponding to each cargo characteristic from the historical cargo information;
comparing the historical carrying times corresponding to each cargo feature with a preset time threshold value to generate a comparison result, and determining cargo preference information according to the comparison result;
generating vehicle preference information according to the vehicle information, and taking the vehicle preference information and the cargo preference information as characteristic preference information;
and carrying out vector coding conversion on the characteristic preference information to obtain a characteristic vector.
3. The vehicle-cargo matching method according to claim 2, wherein the performing vector code conversion on the feature preference information to obtain a feature vector comprises:
extracting a plurality of cargo features and a plurality of vehicle features in the feature preference information;
respectively carrying out vector coding and feature cross processing on the plurality of cargo features and the plurality of vehicle features according to a preset rule to obtain a plurality of vector element values;
and generating a feature vector corresponding to the feature preference information according to the vector element values.
4. The vehicle and cargo matching method according to claim 3, wherein the vector encoding and feature crossing processing are respectively performed on the plurality of cargo features and the plurality of vehicle features according to a preset rule to obtain a plurality of vector element values, and the method comprises the following steps:
extracting characteristic numerical values of the plurality of cargo characteristics to obtain a plurality of cargo characteristic numerical values;
carrying out feature coding on the plurality of vehicle features to obtain a plurality of vehicle code numerical values;
generating a plurality of vector element values based on the plurality of cargo characteristic values and the plurality of vehicle code values.
5. The vehicle and cargo matching method according to claim 1, wherein the step of obtaining information of the cargo to be matched, inputting the information of the cargo to be matched and the feature vector into a preset regression prediction model for vehicle and cargo matching to obtain a plurality of vehicle and cargo matching results comprises:
acquiring information of goods to be matched, and determining weight parameters corresponding to the goods to be matched according to the information of the goods to be matched;
inputting the characteristic vector and the weight parameter into a preset regression prediction model, and calculating the vehicle and goods matching probability of the goods information to be matched through the regression prediction model to obtain a plurality of vehicle and goods matching probabilities;
and generating a vehicle and goods matching result corresponding to each vehicle and goods matching probability according to the plurality of vehicle and goods matching probabilities.
6. The vehicle and goods matching method of claim 1, wherein the step of inputting the vehicle and goods matching results into a preset click rate prediction model for click rate prediction to obtain a plurality of click rate prediction values comprises the steps of:
respectively inputting the vehicle and goods matching results into a preset click rate estimation model;
calculating the correlation degree of each vehicle and goods matching result through a neural network in the click rate estimation model;
and generating the click rate predicted value of each vehicle and goods matching result according to the relevance of each vehicle and goods matching result to obtain a plurality of click rate predicted values.
7. The vehicle-cargo matching method according to any one of claims 1-6, wherein the generating of a vehicle recommendation sequence according to the click rate predicted values and the vehicle-cargo recommendation of the cargo information to be matched according to the vehicle recommendation sequence comprise:
sorting the candidate vehicles according to the size of the click rate predicted values to obtain vehicle recommendation sorting;
taking the candidate vehicle with the largest click rate predicted value in the vehicle recommendation sequence as a target vehicle;
and recommending the cargo information to be matched to the target vehicle.
8. The vehicle and goods matching device is characterized by comprising:
the acquisition module is used for acquiring initial vehicle and cargo information to be processed from a preset database, wherein the initial vehicle and cargo information comprises: vehicle information, historical cargo information, and historical environmental information;
the characteristic extraction module is used for extracting the characteristics of the initial vehicle and cargo information to obtain characteristic preference information and performing vector coding conversion on the characteristic preference information to obtain a characteristic vector;
the matching module is used for acquiring information of goods to be matched, and inputting the information of the goods to be matched and the characteristic vector into a preset regression prediction model for matching the goods to obtain a plurality of goods matching results;
the prediction module is used for inputting the plurality of vehicle and goods matching results into a preset click rate prediction model respectively to perform click rate prediction so as to obtain a plurality of click rate prediction values;
and the recommending module is used for generating vehicle recommending sequences according to the plurality of click rate predicted values and recommending the goods to be matched according to the vehicle recommending sequences.
9. A vehicle-cargo matching apparatus characterized by comprising: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invokes the instructions in the memory to cause the vehicle-cargo matching device to perform the vehicle-cargo matching method of any one of claims 1-7.
10. A computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the vehicle-cargo matching method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210615097.1A CN114936822A (en) | 2022-06-01 | 2022-06-01 | Vehicle and goods matching method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210615097.1A CN114936822A (en) | 2022-06-01 | 2022-06-01 | Vehicle and goods matching method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114936822A true CN114936822A (en) | 2022-08-23 |
Family
ID=82865629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210615097.1A Pending CN114936822A (en) | 2022-06-01 | 2022-06-01 | Vehicle and goods matching method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114936822A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116342011A (en) * | 2023-05-23 | 2023-06-27 | 万联易达物流科技有限公司 | Intelligent matching method and system for vehicles and goods in whole vehicle transportation |
| CN117236646A (en) * | 2023-11-10 | 2023-12-15 | 杭州一喂智能科技有限公司 | Vehicle scheduling method, device, electronic equipment and computer readable medium |
-
2022
- 2022-06-01 CN CN202210615097.1A patent/CN114936822A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116342011A (en) * | 2023-05-23 | 2023-06-27 | 万联易达物流科技有限公司 | Intelligent matching method and system for vehicles and goods in whole vehicle transportation |
| CN116342011B (en) * | 2023-05-23 | 2023-07-21 | 万联易达物流科技有限公司 | Intelligent matching method and system for vehicles and goods in whole vehicle transportation |
| CN117236646A (en) * | 2023-11-10 | 2023-12-15 | 杭州一喂智能科技有限公司 | Vehicle scheduling method, device, electronic equipment and computer readable medium |
| CN117236646B (en) * | 2023-11-10 | 2024-03-12 | 杭州一喂智能科技有限公司 | Vehicle scheduling method, device, electronic equipment and computer readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902987B (en) | Method for determining a transport plan, method and device for training a rapid loading model | |
| CN114936822A (en) | Vehicle and goods matching method, device, equipment and storage medium | |
| Eskandarpour et al. | A reverse logistics network for recovery systems and a robust metaheuristic solution approach | |
| Sadri Esfahani et al. | Modeling the time windows vehicle routing problem in cross-docking strategy using two meta-heuristic algorithms | |
| CN110689355A (en) | Client classification method, device, computer equipment and storage medium | |
| CN112862184A (en) | Transportation information prediction method and device, electronic equipment and readable storage medium | |
| Papier et al. | Queuing models for sizing and structuring rental fleets | |
| Jiang et al. | A scheme for determining vehicle routes based on Arc-based service network design | |
| CN111324827A (en) | Method, device, equipment and storage medium for intelligently recommending goods source order information | |
| TW201203166A (en) | Bulk distribution method | |
| CN114444998A (en) | Logistics supply chain management system | |
| CN117709836A (en) | Management system of digital logistics supply chain | |
| CN116523270B (en) | Logistics transportation task automatic scheduling method, equipment, server and medium | |
| Zare Mehrjerdi | A multiple objective stochastic approach to vehicle routing problem | |
| CN108399524A (en) | Information processing method and system | |
| CN115222340B (en) | Goods scheduling management method based on intelligent warehousing and related device | |
| CN116415747A (en) | Method and device for determining carpooling route and electronic equipment | |
| CN111680941A (en) | Premium recommendation method, device, equipment and storage medium | |
| CN115907374A (en) | Intelligent dispatching method, device and equipment for vehicle of distribution center and storage medium | |
| CN114418417A (en) | Vehicle and goods matching method, and transport capacity prediction method, device and electronic equipment based on vehicle and goods matching method | |
| CN113762842B (en) | Warehouse scheduling method, server and system | |
| CN113888096A (en) | Goods management method and system for logistics park | |
| CN111768139B (en) | Stock processing method, apparatus, device and storage medium | |
| CN112734237A (en) | Distributor matching method for transaction service | |
| Behnamian et al. | Optimal design of cross docking supply chain networks with time-varying uncertain demands |
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 |