WO2022247041A1 - 一种航运用户大数据建模分析方法及系统 - Google Patents

一种航运用户大数据建模分析方法及系统 Download PDF

Info

Publication number
WO2022247041A1
WO2022247041A1 PCT/CN2021/115832 CN2021115832W WO2022247041A1 WO 2022247041 A1 WO2022247041 A1 WO 2022247041A1 CN 2021115832 W CN2021115832 W CN 2021115832W WO 2022247041 A1 WO2022247041 A1 WO 2022247041A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
shipowner
record data
activity record
information
Prior art date
Application number
PCT/CN2021/115832
Other languages
English (en)
French (fr)
Inventor
吴键
Original Assignee
海南超船电子商务有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 海南超船电子商务有限公司 filed Critical 海南超船电子商务有限公司
Priority to US17/594,093 priority Critical patent/US20240037485A1/en
Publication of WO2022247041A1 publication Critical patent/WO2022247041A1/zh

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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/21—Design, administration or maintenance of databases
    • G06F16/211—Schema design and management
    • G06F16/212—Schema design and management with details for data modelling support
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G06N5/022—Knowledge engineering; Knowledge acquisition
    • 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
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • 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
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • 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
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • 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
    • 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the invention relates to the technical field of big data modeling and analysis, in particular to a method and system for big data modeling and analysis of shipping users.
  • the online platform cannot provide more comprehensive analysis and matching functions for ship owners or cargo owners, and it is difficult to further improve the transaction completion rate.
  • the object of the present invention is to provide a shipping user big data modeling and analysis method and system to overcome or at least partially solve the above-mentioned problems in the prior art.
  • the first aspect of the present invention provides a big data modeling and analysis method for shipping users, which is applied to a shipping futures order matching platform.
  • the shipping futures order matching platform includes a client and a background system.
  • the client is used for users to manage shipping schedules or Pallet-related information
  • the users include shipowner users and cargo owner users
  • the background system is connected to the client through the network to realize the matching and pushing of shipping schedules and pallet information, including the following steps:
  • the multi-dimensional activity record data of the cargo owner user includes the cargo owner’s first dimension activity record data and the cargo owner’s second dimension activity record data, the cargo owner’s first dimension activity record data is the cargo owner’s local activity record data, and the cargo owner’s second dimension activity record data
  • the recorded data is the record data of the owner's third-party activities;
  • the multi-dimensional activity record data of the shipowner user includes the first dimension activity record data of the shipowner and the second dimension activity record data of the shipowner.
  • the first dimension activity record data of the shipowner is the local activity record data of the shipowner, and the second dimension activity record data of the shipowner
  • the two-dimensional activity record data is the shipowner's third-party activity record data.
  • step S3 the calculation of the owner's evaluation score according to the owner's user model specifically includes: obtaining the corresponding preset weight values of each item of activity record data in the owner's multi-dimensional activity record data, and weighting the owner's multi-dimensional activity record data Calculate and obtain the owner's evaluation score;
  • the calculation of the shipowner's evaluation score according to the shipowner user model specifically includes: obtaining the corresponding preset weight values of each activity record data in the shipowner's multi-dimensional activity record data, performing weighted calculations on the shipowner's multi-dimensional activity record data, and obtaining Owner evaluation score.
  • step S4 the pallet information released by the cargo owner user is paired with the schedule information released by the shipowner user based on the cargo owner user classification and the shipowner user classification, and the matching information is pushed to the cargo owner user and the shipowner user, Specifically include:
  • S401 Screen the shipping schedule information released by the shipowner user according to the pallet information released by the cargo owner user, and remove the schedule information that does not match the time and load;
  • S402. Perform secondary screening on the filtered schedule information according to the cargo owner user's evaluation score demand information for the shipowner user, and remove the shipping schedule information corresponding to the ship schedule information whose evaluation score of the shipowner user is lower than the evaluation score requirement;
  • step S2 there are steps between step S2 and step S3:
  • the method also includes the following steps:
  • a training model for predicting biased information is established based on a machine learning algorithm
  • the weighted information is predicted through the training model, and the weighted calculation is performed on the ship owner's multi-dimensional activity record data according to the prediction result.
  • step S5 specifically includes the following steps:
  • step S6 specifically includes the following steps:
  • the second aspect of the present invention provides a big data modeling and analysis system for shipping users, which is applied to a shipping futures order matching platform.
  • the shipping futures order matching platform includes a client and a background system.
  • the client is used for users to manage shipping schedules or Pallet-related information
  • the users include shipowner users and cargo owner users
  • the background system is connected to the client through the network, and is used to realize the matching and pushing of shipping schedules and pallet information
  • the system includes:
  • the consignor model building module is used to obtain the multi-dimensional activity record data of the consignor user, and establish the consignor user model based on the multi-dimensional activity record data of the consignor user;
  • the shipowner model building module is used to obtain the multi-dimensional activity record data of the shipowner user, and establish the shipowner user model based on the multi-dimensional activity record data of the shipowner user;
  • the evaluation and grading module is used to calculate the cargo owner evaluation score according to the cargo owner user model, and classify the cargo owner user according to the cargo owner evaluation score, and is also used to calculate the ship owner evaluation score according to the ship owner user model, and classify the ship owner user according to the ship owner evaluation score ;
  • the pairing module is used to pair the pallet information released by the cargo owner user with the schedule information released by the shipowner user based on the cargo owner user classification and the shipowner user classification, and push the matching information to the cargo owner user and the shipowner user.
  • a method and system for modeling and analyzing shipping user big data can respectively establish a cargo owner user model and a shipowner user model for the multi-dimensional activity record data of cargo owner users and shipowner users, and respectively calculate based on the two models Cargo owner evaluation scores and ship owner evaluation scores, based on the evaluation scores, the corresponding cargo owner users and ship owner users are classified, and the pallet information and shipping schedule information released by both parties are matched according to the classification, and the corresponding cargo owner users and ship owner users are notified according to the matching results.
  • Shipowner users push matching information, so that based on the needs of both parties, the big data generated based on user activities can further help users select suitable business partners and help improve the transaction completion rate.
  • Fig. 1 is a schematic diagram of the overall flow of a shipping user big data modeling and analysis method provided by an embodiment of the present invention.
  • Fig. 2 is a schematic diagram of the matching process of pallet information and shipping schedule information provided by an embodiment of the present invention.
  • Fig. 3 is a schematic diagram of the overall flow of a shipping user big data modeling and analysis method provided by another embodiment of the present invention.
  • Fig. 4 is a schematic diagram of the overall flow of a shipping user big data modeling and analysis method provided by another embodiment of the present invention.
  • Fig. 5 is a schematic diagram of the overall structure of a shipping user big data modeling and analysis system provided by an embodiment of the present invention.
  • 1 cargo owner model building module 2 ship owner model building module, 3 evaluation and grading module, 4 pairing module.
  • an embodiment of the present invention provides a big data modeling and analysis method for shipping users, the method is applied to a shipping futures order matching platform, and the shipping futures order matching platform includes a client and a background system, and the client is used for Users manage information related to shipping schedules or pallets.
  • the users include shipowner users and cargo owner users. Specifically, shipowner users can publish and manage shipping schedule information and view pallet-related information through the client, while cargo owner users can use client
  • the terminal publishes and manages the pallet information, and checks the relevant information of the shipping schedule;
  • the background system is connected to the client through the network, and is used to realize the matching of the shipping schedule and the pallet information between different clients—that is, corresponding users of different clients and pushing, the method includes the following steps:
  • the multi-dimensional activity record data of the cargo owner user includes the cargo owner's first dimension activity record data and the cargo owner's second dimension activity record data.
  • the first dimension activity record data of the cargo owner is the cargo owner's local activity record data;
  • the cargo owner's second dimension activity record data is the cargo owner's third-party activity record data.
  • the local activity record data of the shipper is the activity record data generated when the user of the shipper performs various operations on the ship futures order matching platform, such as the number of logins, the number of communications, the turnover, and the failure of the supply of goods, etc. The number of online communications between users and shipowner users through the platform.
  • the third-party activity record data of the consignor is the activity record data generated by the consignor user on the third-party website or platform, such as the number of phone calls received by the consignor user through the operator, the number of overdue credit inquiries of the consignor user through the bank, and the industrial and commercial information query of the consignor.
  • the platform obtains the business operations of the cargo owners and users, etc.
  • a cargo owner user model is established for each cargo owner user, and the model is used to record the user name, name, multi-dimensional activity record data, cargo owner evaluation scores, grade information, etc. of the cargo owner user, reflecting from multiple dimensions Comprehensive information about the owner and user of the cargo.
  • the multi-dimensional activity record data of the shipowner user includes the first dimension activity record data of the shipowner and the second dimension activity record data of the shipowner.
  • the first dimension activity record data of the shipowner is the shipowner local activity record data;
  • the shipowner second dimension activity record data is the shipowner third party activity record data.
  • the shipowner’s local activity record data is the activity record data generated when the shipowner user performs various operations on the ship futures order matching platform, such as the number of logins, the number of times of self-updated shipping schedules, the number of communications, and the transaction volume. The number of times refers to the number of online communications between the shipowner user and the cargo owner user through the platform.
  • the shipowner’s third-party activity record data is the activity record data generated by the shipowner user on the third-party website or platform, such as the number of phone calls received by the shipowner user through the operator, and the ship owner’s ship operation data obtained through the shipping company or related websites , Obtain the number of overdue credit investigations of shipowner users through the bank.
  • a shipowner user model is established for each shipowner user, which is used to record the user name, name, multi-dimensional activity record data, shipowner evaluation scores, grade information and other content of the shipowner user, from multiple This dimension reflects the comprehensive situation of the shipowner user.
  • a shipping user big data modeling and analysis method provided in this embodiment is to establish a cargo owner user model and a shipowner user model for the multi-dimensional activity record data of cargo owner users and shipowner users, and calculate cargo owner evaluations based on the two models Scores and shipowner evaluation scores, based on the evaluation scores, the corresponding cargo owner users and shipowner users are classified, and the pallet information and shipping schedule information released by both parties are matched according to the classification, and the corresponding cargo owner users and shipowner users are notified according to the matching results.
  • Users push pairing information, so that based on the needs of both parties, a corresponding user model is established based on the big data generated by user activities, which further helps users to filter suitable business partners and helps to improve the transaction completion rate.
  • step S3 the calculation of the owner's evaluation score according to the owner's user model specifically includes: obtaining the corresponding preset weight value of each item of activity record data in the owner's multi-dimensional activity record data, Carry out weighted calculation on the owner's multi-dimensional activity record data to obtain the owner's evaluation score.
  • the calculation of the shipowner's evaluation score according to the shipowner user model specifically includes: obtaining the corresponding preset weight values of each activity record data in the shipowner's multi-dimensional activity record data, and performing weighted calculation on the shipowner's multi-dimensional activity record data , to obtain the ship owner evaluation score.
  • each data in the cargo owner's multi-dimensional activity record data and the shipowner's multi-dimensional activity record data corresponds to a preset weight value, and each data is added after being weighted according to the corresponding preset weight value, and divided Based on the number of data items, the result is the cargo owner evaluation score/ship owner evaluation score.
  • activity record data whose original meaning is not expressed by numbers, such as business operations or ship operations, different numbers can be used to refer to different business operations to facilitate calculation.
  • weighted calculations are performed according to the importance of different activity record data, so that the overall evaluation of cargo owner users/shipowner users can be carried out through the single index of cargo owner evaluation score/shipowner evaluation score, and the evaluation score is evaluated according to the level of the evaluation score.
  • Cargo owner users and shipowner users are graded. The higher the grade, the better the overall business situation, credit situation, and communication efficiency of the corresponding users, so that users can be screened according to the grade in the future.
  • step S4 the pallet information released by the cargo owner user is paired with the schedule information issued by the ship owner user based on the classification of the cargo owner user and the classification of the ship owner user. , and push pairing information to cargo owner users and shipowner users, including:
  • S401 Screen the shipping schedule information released by the shipowner user according to the pallet information released by the cargo owner user, and remove the sailing schedule information that does not match the time and load.
  • the shipping schedule information released by the shipowner user cannot meet the time information such as the duration of the freight demand in the pallet information released by the cargo owner user, or the load in the shipping schedule information released by the shipowner user cannot meet the time information in the pallet information.
  • the weight of the cargo filter out such shipping schedule information that cannot meet the demand.
  • S402. Perform secondary screening on the filtered sailing schedule information according to the evaluation score demand information of the shipowner user by the cargo owner user, and remove the sailing schedule information corresponding to the ship schedule information whose evaluation score of the shipowner user is lower than the evaluation score requirement.
  • the evaluation score demand information of the cargo owner user for the shipowner user reflects the requirements of the cargo owner user for the corresponding evaluation score of the shipowner user.
  • the evaluation score demand information reflects the requirements of the cargo owner user for the corresponding evaluation score of the shipowner user.
  • the unsatisfactory shipping schedule information will be screened out.
  • step S403 the shipowner user's evaluation score demand information for the cargo owner user corresponding to the shipping schedule information after the secondary screening is obtained in advance.
  • the evaluation score information reflects the shipowner user's requirements for the cargo owner user's evaluation score. If the evaluation score also meets the requirements of the second-screened shipowner user, the second-screened shipping schedule information can be sent to the cargo owner user for selection and entrustment by the main user.
  • the shipping schedule and the shipowner’s comprehensive information are obtained through double screening, and the shipping schedule information that meets the requirements of the cargo owner user is selected by the supplier owner user. After the cargo owner user selects the entrusted object, further judgment is judged by the evaluation score of the cargo owner user. Whether the comprehensive situation of the cargo owner and the user meets the requirements of the shipowner and the user, so that the comprehensive situation of both parties in the final transaction meets the requirements of the other party, making it easier to facilitate the transaction, improve transaction efficiency, and avoid the occurrence that it is difficult for the cargo owner to find a suitable shipping date, resulting in the loss of the cargo. Frustrated, or the owner's vessel lay idle for a long time.
  • step S2 As an optional implementation manner of this embodiment, referring to FIG. 3 , further steps are included between step S2 and step S3:
  • the weight information is used to describe the degree of importance that cargo owner users attach to various activity record data in the multi-dimensional activity records of shipowner users. The higher the importance attached to a certain activity record data, the The higher the weight of the activity record data.
  • the shipowner evaluation score calculated in this embodiment is calculated based on the degree of importance of cargo owner users to different shipowner activity record data after adjusting the temporary weight. The degree of emphasis on different aspects is different. By calculating the owner's evaluation score based on the shipper's biased information, the final evaluation score and classification can adapt to the different needs of different shippers, realize personalized evaluation, and help further promote transactions.
  • the method further includes the following steps:
  • a training model for predicting weight information is established based on a machine learning algorithm.
  • the weighted information is predicted through the training model, and the weighted calculation is performed on the ship owner's multi-dimensional activity record data according to the prediction result.
  • the purpose of this embodiment is to establish a training model that can be used to predict biased information based on a machine learning algorithm, so that the corresponding biased information can be automatically predicted according to the pallet information, without the need for the owner and user to repeatedly provide biased information, which not only speeds up the calculation and processing efficiency, It can also improve the user experience.
  • step S5 specifically includes the following steps:
  • different shippers may have different emphasis on shipowner multi-dimensional activity record data, which may be due to the commercial considerations of the shippers themselves. For example, some shippers pay more attention to the efficiency of communication with shipowners, while others The shipper pays more attention to the credit of the shipowner, which may directly lead to the fact that when certain shipper IDs appear in the pallet information, the corresponding information will usually place more emphasis on the record data of a certain shipowner's activities; it may also be based on the information reflected in the pallet Cargo, time and other information are considered, thus affecting the emphasis on information.
  • the training model predicts the biased information based on the significant factors, so that the cargo owner does not need to repeatedly provide biased information.
  • Described step S6 specifically comprises the following steps:
  • the shipowner’s evaluation score is calculated based on the weighted prediction information. On the one hand, it can make the shipowner’s evaluation score better reflect the evaluation level of the shipowner under the difference of different cargo owners. On the other hand, it does not require the cargo owner Repeatedly providing biased information can not only make the results meet the needs of users, but also make users feel indifferent during this process, which can further improve the user experience.
  • another embodiment of the present invention provides a big data modeling and analysis system for shipping users.
  • the system is applied to a shipping futures order matching platform, and the shipping futures order matching platform includes a client and a background system, the client is used for users to manage information related to shipping schedules or pallets, and the users include shipowner users and cargo owner users.
  • the background system is connected to the client through the network to realize shipping schedules and pallet information matching and pushing. Referring to Figure 5, the system specifically includes:
  • Consignor model building module 1 used to obtain consignor user multi-dimensional activity record data, establish consignor user model based on consignor user multi-dimensional activity record data, said consignor user multi-dimensional activity record data includes consignor first dimension activity record data and consignor second Two-dimensional activity record data, the first dimension activity record data of the cargo owner is the cargo owner’s local activity record data, and the cargo owner’s second dimension activity record data is the cargo owner’s third-party activity record data;
  • the shipowner model building module 2 is used to obtain the multi-dimensional activity record data of the shipowner user, and establish the shipowner user model based on the multi-dimensional activity record data of the shipowner user.
  • the multi-dimensional activity record data of the shipowner user includes the first dimension of the shipowner Activity record data and shipowner's second-dimensional activity record data, the shipowner's first-dimensional activity record data is the shipowner's local activity record data, and the shipowner's second-dimensional activity record data is the shipowner's third-party activity record data;
  • the evaluation and classification module 3 is used to calculate the cargo owner's evaluation score according to the cargo owner's user model, classify the cargo owner's user according to the cargo owner's evaluation score, and also calculate the shipowner's evaluation score according to the shipowner's user model, and classify the shipowner's user according to the shipowner's evaluation score Grading;
  • the pairing module 4 is used to pair the pallet information released by the cargo owner user with the schedule information released by the shipowner user based on the cargo owner user classification and the shipowner user classification, and push the matching information to the cargo owner user and the shipowner user.
  • the evaluation and grading module 3 specifically includes:
  • the cargo owner evaluation and grading module is used to obtain the corresponding preset weight value of each activity record data in the cargo owner's multi-dimensional activity record data, perform weighted calculation on the cargo owner's multi-dimensional activity record data, and obtain the cargo owner's evaluation score;
  • the shipowner evaluation and grading module is used to obtain the corresponding preset weight values of each activity record data in the shipowner's multi-dimensional activity record data, carry out weighted calculation on the shipowner's multi-dimensional activity record data, and obtain the shipowner evaluation score.
  • the pairing module 4 specifically includes:
  • the primary screening module is used to filter the shipping schedule information released by the shipowner user according to the pallet information released by the cargo owner user, and remove the schedule information that does not match the time and load;
  • the secondary screening module is used to perform secondary screening on the screened shipping schedule information according to the demand information of the evaluation score of the shipowner user by the cargo owner user, and remove the ship whose evaluation score of the ship owner user corresponding to the shipping schedule information is lower than the evaluation score requirement period information;
  • the first sending module is used to obtain the evaluation score demand information of the ship owner user for the cargo owner user corresponding to the shipping schedule information after the secondary screening, and judge whether the cargo owner user's evaluation score meets the ship owner user's score demand information requirement. , send the second-screened shipping schedule information to the cargo owner user;
  • the second sending module is used to acquire the shipping schedule information selected by the shipper user from the second-screened shipping schedule information, and send the pallet information to the shipowner user corresponding to the selected shipping schedule information.
  • the system further includes an acquisition module and an adjustment module.
  • the obtaining module is used to obtain information on the emphasis of the cargo owner user on each activity record data in the multi-dimensional activity record data of the shipowner user;
  • the adjustment module is used to adjust the corresponding preset weight value of each activity record data in the shipowner's multi-dimensional activity record data according to the bias information to a temporary weight value, and perform weighted calculation on the shipowner's multi-dimensional activity record data based on the temporary weight value, Earn shipowner rating points.
  • system further includes a training model building module and a prediction weighting module.
  • the training model building module includes:
  • the acquisition sub-module is used to acquire the historical pallet information released by multiple shipper users and the historical bias information of each activity record data in the multi-dimensional activity record data of shipowner users;
  • the analysis sub-module is used to analyze the historical pallet information and the corresponding historical weight information, and determine the significant factors in the historical pallet information, and the significant factor is the content of the historical pallet information that affects the historical weight information;
  • the modeling sub-module is used to construct a classifier according to significant factors and historical weight information, and establish a training model based on a machine learning algorithm, and the training model is used to predict corresponding weight information according to significant factors in pallet information.
  • the prediction weighting module specifically includes:
  • the prediction sub-module is used to analyze and extract the significant factors in the pallet information released by the cargo owner user when calculating the shipowner evaluation score according to the shipowner user model next time, and input the extracted significant factors into the training model to obtain the prediction bias information ;
  • the weighting calculation sub-module is used to adjust the corresponding preset weight value of each activity record data in the multi-dimensional activity record data of the shipowner based on the forecast bias information to a temporary weight value, and to weight the multi-dimensional activity record data of the shipowner based on the temporary weight value Calculate and obtain the ship owner evaluation score.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Operations Research (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Game Theory and Decision Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种航运用户大数据建模分析方法及系统，该方法包括：S1、获取货主用户多维度活动记录数据，基于货主用户多维度活动记录数据建立货主用户模型；S2、获取船东用户多维度活动记录数据，基于船东用户多维度活动记录数据建立船东用户模型；S3、根据货主用户模型计算货主评价分数，根据货主评价分数对货主用户进行分级，根据船东用户模型计算船东评价分数，根据船东评价分数对船东用户进行分级；S4、基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息。该方法基于用户活动产生的大数据建模以帮助用户筛选合适的业务合作对象，有助于提高交易达成率。

Description

一种航运用户大数据建模分析方法及系统 技术领域
本发明涉及大数据建模分析技术领域，尤其涉及一种航运用户大数据建模分析方法及系统。
背景技术
随着航运业的快速发展，国内沿海干散货航运市场日益扩大，货主和船东的交易需求量日益增长，为了促进船东和货主达成交易，市面上出现了用于实现船期和货盘匹配的网络平台，船东用户可以通过平台发布船期信息供有需求的货主选择委托，而货主用户也可以通过平台发布货盘信息供船东用户选择承接，但现有的此类网络平台通常只能提供船期信息和货盘信息的发布功能，对于船期与货盘的匹配需要船东和货主自行浏览选择，而部分网络平台虽然能实现船期与货盘的自动匹配，但其匹配的实现仅基于船期信息中的载重、时间等是否能满足货盘信息的需求，而船东或货主自身的经营情况等信息对于另一方选择业务合作对象而言也是十分重要的参考，现有的网络平台并不能为船东或货主提供更全面的分析和匹配功能，难以进一步提高交易的达成率。
发明内容
鉴于此，本发明的目的在于提供一种航运用户大数据建模分析方法及系统，以克服或至少部分解决现有技术所存在的上述问题。
本发明第一方面提供一种航运用户大数据建模分析方法，应用于船期货盘匹配平台，所述船期货盘匹配平台包括客户端和后台系统，所述客户端用于用户管理船期或货盘相关信息，所述用户包括船东用户和货主用户，所述后台系统与客户端通过网络相连接，用于实现船期与货盘信息的匹配和推送，包括以下步骤：
S1、获取货主用户多维度活动记录数据，基于货主用户多维度活动记录数据建立货主用户模型；
S2、获取船东用户多维度活动记录数据，基于船东用户多维度活动记录数据建立船东用户模型；
S3、根据货主用户模型计算货主评价分数，根据货主评价分数对货主用户进行分级，根据船东用户模型计算船东评价分数，根据船东评价分数对船东用户进行分级；
S4、基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息。
进一步的，所述货主用户多维度活动记录数据包括货主第一维度活动记录数据和货主第二维度活动记录数据，所述货主第一维度活动记录数据为货主本地活动记录数据，货主第二维度活动记录数据为货主第三方活动记录数据；
所述船东用户多维度活动记录数据包括船东第一维度活动记录数据和船东第二维度活动记录数据，所述船东第一维度活动记录数据为船东本地活动记录数据，船东第二维度活动记录数据为船东第三方活动记录数据。
进一步的，步骤S3中，所述根据货主用户模型计算货主评价分数，具体包括：获取货主多维度活动记录数据中各项活动记录数据的相应预设权重值，对货主多维度活动记录数据进行加权计算，获得货主评价分数；
所述根据船东用户模型计算船东评价分数，具体包括：获取船东多维度活动记录数据中各项活动记录数据的相应预设权重值，对船东多维度活动记录数据进行加权计算，获得船东评价分数。
进一步的，步骤S4中，所述基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息，具体包括：
S401、根据货主用户发布的货盘信息对船东用户发布的船期信息进行筛选，去除时间、载重不匹配的船期信息；
S402、根据货主用户对船东用户的评价分数需求信息对经过筛选的船期信息进行二次筛选，去除船期信息所对应的船东用户评价分数低于评价分数需求的船期信息；
S403、获取二次筛选后的船期信息所对应的船东用户对货主用户的评价分数需求信息，判断货主用户的评价分数是否符合船东用户的分数需求信息要求，若符合，则向货主用户发送二次筛选后的船期信息；
S404、获取货主用户从二次筛选后的船期信息中所选择的船期信息，并向所选择的船期信息所对应的船东用户发送货盘信息。
进一步的，在步骤S2与步骤S3之间还包括步骤：
S21、获取货主用户对于船东用户多维度活动记录数据中各项活动记录数据的偏重信息；
S22、根据偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
进一步的，所述方法还包括以下步骤：
S5、根据历史货盘信息和历史偏重信息，基于机器学习算法建立用于预测偏重信息的训练模型；
S6、根据货盘信息通过训练模型预测偏重信息，根据预测结果对船东多维度活动记录数据进行加权计算。
进一步的，所述步骤S5具体包括以下步骤：
S501、获取多个货主用户发布的历史货盘信息和对船东用户多维度活动记录数据中各项活动记录数据的历史偏重信息；
S502、分析历史货盘信息和对应的历史偏重信息，确定历史货盘信息中的显著因子，所述显著因子为影响历史偏重信息的历史货盘信息内容；
S503、根据显著因子和历史偏重信息构建分类器，基于机器学习算法建立训练模型，所述训练模型用于根据货盘信息中的显著因子预测相应的偏重信息。
进一步的，所述步骤S6具体包括以下步骤：
S601、在下一次根据船东用户模型计算船东评价分数时，分析并提取货主用户发布的货盘信息中的显著因子，将提取的显著因子输入到训练模型中，获得预测偏重信息；
S602、基于预测偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
本发明第二方面提供一种航运用户大数据建模分析系统，应用于船期货盘匹配平台，所述船期货盘匹配平台包括客户端和后台系统，所述客户端用于用户管理船期或货盘相关信息，所述用户包括船东用户和货主用户，所述后台系统与客户端通过网络相连接，用于实现船期与货盘信息的匹配和推送，所述系统包括：
货主模型建立模块，用于获取货主用户多维度活动记录数据，基于货主用户多维度活动记录数据建立货主用户模型；
船东模型建立模块，用于获取船东用户多维度活动记录数据，基于船东用户多维度活动记录数据建立船东用户模型；
评价分级模块，用于根据货主用户模型计算货主评价分数，根据货主评价分数对货主用户进行分级，还用于根据船东用户模型计算船东评价分数，根据船东评价分数对船东用户进行分级；
配对模块，用于基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息。
与现有技术相比，本发明的有益效果是：
本发明所提供的一种航运用户大数据建模分析方法及系统，能够分别针对货主用户和船东用户的多维度活动记录数据建立货主用户模型和船东用户模型，并基于两种模型分别计算货主评价分数和船东评价分数，基于评价分数对相应的货主用户和船东用户进行分级，根据分级对双方发布的货盘信息 和船期信息进行配对，并根据配对结果向相应的货主用户和船东用户推送配对信息，从而在基于双方需求的基础上，基于用户活动产生的大数据更进一步帮助用户筛选合适的业务合作对象，有助于提高交易达成率。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的优选实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明一实施例提供的一种航运用户大数据建模分析方法整体流程示意图。
图2是本发明一实施例提供的货盘信息与船期信息配对流程示意图。
图3是本发明另一实施例提供的一种航运用户大数据建模分析方法整体流程示意图。
图4是本发明又一实施例提供的一种航运用户大数据建模分析方法整体流程示意图。
图5是本发明一实施例提供的一种航运用户大数据建模分析系统整体结构示意图。
图中，1货主模型建立模块，2船东模型建立模块，3评价分级模块，4配对模块。
具体实施方式
以下结合附图对本发明的原理和特征进行描述，所列举实施例只用于解释本发明，并非用于限定本发明的范围。
参照图1，本发明实施例提供一种航运用户大数据建模分析方法，该方法应用于船期货盘匹配平台，所述船期货盘匹配平台包括客户端和后台系统，所述客户端用于用户管理船期或货盘相关信息，所述用户包括船东用户和货 主用户，具体的，船东用户可以通过客户端发布、管理船期信息，查看货盘相关信息，而货主用户可以通过客户端发布、管理货盘信息，查看船期相关信息；所述后台系统与客户端通过网络相连接，用于实现不同客户端——即不同客户端对应用户之间船期与货盘信息的匹配和推送，所述方法包括以下步骤：
S1、获取货主用户多维度活动记录数据，基于货主用户多维度活动记录数据建立货主用户模型。
示例性地，所述货主用户多维度活动记录数据包括货主第一维度活动记录数据和货主第二维度活动记录数据。所述货主第一维度活动记录数据为货主本地活动记录数据；货主第二维度活动记录数据为货主第三方活动记录数据。其中，货主本地活动记录数据为货主用户在船期货盘匹配平台上进行各种操作行为时产生的活动记录数据，例如登录次数、沟通次数、成交额、货源落空情况等，所述沟通次数为货主用户通过平台与船东用户的线上通信次数。货主第三方活动记录数据为货主用户在第三方网站或平台上产生的活动记录数据，例如通过运营商获得货主用户的电话接听次数、通过银行获取货主用户的征信逾期次数、通过企业工商信息查询平台获取货主用户对应的企业经营情况等。
本实施例中对于每个货主用户分别建立一货主用户模型，该模型用于记录货主用户的用户名、姓名、多维度活动记录数据、货主评价分数、等级信息等内容，从多个维度反映了该货主用户的综合情况。
S2、获取船东用户多维度活动记录数据，基于船东用户多维度活动记录数据建立船东用户模型。
示例性地，所述船东用户多维度活动记录数据包括船东第一维度活动记录数据和船东第二维度活动记录数据。所述船东第一维度活动记录数据为船东本地活动记录数据；船东第二维度活动记录数据为船东第三方活动记录数 据。其中，船东本地活动记录数据为船东用户在船期货盘匹配平台上进行各种操作行为时产生的活动记录数据，例如登录次数、自主更新船期次数、沟通次数、成交额等，其中沟通次数指船东用户通过平台与货主用户的线上通信次数。船东第三方活动记录数据为船东用户在第三方网站或平台上产生的活动记录数据，例如通过运营商获得船东用户的电话接听次数、通过船舶公司或相关网站获取船东的船舶经营数据、通过银行获取船东用户的征信逾期次数等。
本实施例中对于每个船东用户分别建立一船东用户模型，该模型用于记录船东用户的用户名、姓名、多维度活动记录数据、船东评价分数、等级信息等内容，从多个维度反映了该船东用户的综合情况。
S3、根据货主用户模型计算货主评价分数，根据货主评价分数对货主用户进行分级，根据船东用户模型计算船东评价分数，根据船东评价分数对船东用户进行分级。
S4、基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息。
本实施例所提供的一种航运用户大数据建模分析方法，分别针对货主用户和船东用户的多维度活动记录数据建立货主用户模型和船东用户模型，并基于两种模型分别计算货主评价分数和船东评价分数，基于评价分数对相应的货主用户和船东用户进行分级，根据分级对双方发布的货盘信息和船期信息进行配对，并根据配对结果向相应的货主用户和船东用户推送配对信息，从而在基于双方需求的基础上，基于用户活动产生的大数据建立对应的用户模型，更进一步帮助用户筛选合适的业务合作对象，有助于提高交易达成率。
作为本实施例的一种可选实施方式，步骤S3中，所述根据货主用户模型计算货主评价分数，具体包括：获取货主多维度活动记录数据中各项活动记录数据的相应预设权重值，对货主多维度活动记录数据进行加权计算，获 得货主评价分数。同时，所述根据船东用户模型计算船东评价分数，具体包括：获取船东多维度活动记录数据中各项活动记录数据的相应预设权重值，对船东多维度活动记录数据进行加权计算，获得船东评价分数。
示例性地，货主多维度活动记录数据和船东多维度活动记录数据中每项数据均对应一预设权重值，在根据相应的预设权重值对每项数据进行加权后相加，并除以数据项数，结果即为货主评价分数/船东评价分数。对于原始含义不通过数字表达的活动记录数据，例如企业经营情况或船舶经营情况，可以通过不同的数字指代不同的企业经营情况，从而便于计算。
该实施方式中根据不同活动记录数据的重要性进行加权计算，从而使得可以通过货主评价分数/船东评价分数这一单一指标对于货主用户/船东用户进行整体评价，并根据评价分数的高低对货主用户和船东用户进行分级，级别越高可以反映出对应的用户整体经营情况、信用情况、沟通效率越好，以便于后续可以根据分级对用户进行筛选。
作为本实施例的一种可选实施方式，参照图2，步骤S4中，所述基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息，具体包括：
S401、根据货主用户发布的货盘信息对船东用户发布的船期信息进行筛选，去除时间、载重不匹配的船期信息。
示例性地，当船东用户发布的船期信息无法满足货主用户发布的货盘信息中的货运需求时长等时间信息，或者船东用户发布的船期信息中的载重无法满足货盘信息中的货物重量时，将此类无法满足需求的船期信息筛去。
S402、根据货主用户对船东用户的评价分数需求信息对经过筛选的船期信息进行二次筛选，去除船期信息所对应的船东用户评价分数低于评价分数需求的船期信息。
示例性地，该步骤中需要预先获取货主用户对于船东用户的评价分数需 求信息，所述评价分数需求信息反映了货主用户对于船东用户的对应评价分数的要求，对于经过步骤S401筛选后的船期信息，当其对应的船东用户的评价分数不满足货主用户的要求时，将不满足要求的船期信息筛去。
S403、获取二次筛选后的船期信息所对应的船东用户对货主用户的评价分数需求信息，判断货主用户的评价分数是否符合船东用户的分数需求信息要求，若符合，则向货主用户发送二次筛选后的船期信息。
步骤S403中，预先获取二次筛选后的船期信息对应的船东用户对于货主用户的评价分数需求信息，该评价分数信息反映了船东用户对于货主用户的评价分数的要求，若货主用户的评价分数也满足二次筛选后的船东用户的要求，则可以向货主用户发送二次筛选后的船期信息供货主用户选择委托。
S404、获取货主用户从二次筛选后的船期信息中所选择的船期信息，并向所选择的船期信息所对应的船东用户发送货盘信息。船东用户可以根据货盘信息决定是否接受委托或与货主用户通过平台进行线上通信确定交易细节。
本实施例中通过双重筛选获得船期、船东综合情况均满足货主用户要求的船期信息，以供货主用户选择，当货主用户选定委托对象后，进一步判断通过货主用户的评价分数判断货主用户的综合情况是否满足船东用户的要求，从而使得最终交易双方的综合情况均满足对方的要求，更容易促成交易，提高交易效率，避免出现货主不容易找到合适的船期，导致货盘落空，或船东船舶长时间闲置的情况。
作为本实施例的一种可选实施方式，参照图3，在步骤S2与步骤S3之间还包括步骤：
S21、获取货主用户对于船东用户多维度活动记录数据中各项活动记录数据的偏重信息。
示例性地，所述偏重信息用于描述货主用户对于船东用户的多维度活动 记录中各项活动记录数据的重视程度，对于某项活动记录数据的重视程度越高，则在货主用户的角度该项活动记录数据的权重越高。
S22、根据偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
该实施方式中所计算得到的船东评价分数是基于货主用户对于不同船东活动记录数据的偏重程度调整临权重后计算得来，不同货主对于船东的沟通效率、船舶经营情况、诚信度等方面的注重程度不同，通过基于货主的偏重信息计算船东评价分数，使得最后的评价分数和分级能够适应不同货主的不同需求，实现个性化评价，有助于进一步促成交易。
作为一种进一步可选的实施方式，参照图4，所述方法还包括以下步骤：
S5、根据历史货盘信息和历史偏重信息，基于机器学习算法建立用于预测偏重信息的训练模型。
S6、根据货盘信息通过训练模型预测偏重信息，根据预测结果对船东多维度活动记录数据进行加权计算。
本实施例的目的在于，基于机器学习算法建立能够用于预测偏重信息的训练模型，从而能够根据货盘信息自动预测相应的偏重信息，而无需货主用户反复提供偏重信息，既加快计算处理效率，又能提高用户的使用体验。
具体的，所述步骤S5具体包括以下步骤：
S501、获取多个货主用户发布的历史货盘信息和对船东用户多维度活动记录数据中各项活动记录数据的历史偏重信息。每条历史偏重信息对应一条历史货盘信息。
S502、分析历史货盘信息和对应的历史偏重信息，确定历史货盘信息中的显著因子，所述显著因子为影响历史偏重信息的历史货盘信息内容。
S503、根据显著因子和历史偏重信息构建分类器，基于机器学习算法建 立训练模型，所述训练模型用于根据货盘信息中的显著因子预测相应的偏重信息。
示例性地，不同货主对于船东多维度活动记录数据可能会有不同的偏重，导致这种不同的原因可能是出于货主自身的商业考虑，例如有些货主更注重与船东的沟通效率，有些货主更注重船东的信用，这容易直接导致在货盘信息中出现某些货主ID时，对应的偏重信息通常也会更偏重某项船东活动记录数据；也可能是根据货盘所体现的货物、时间等信息考虑，从而影响偏重信息。本实施例通过确定不同偏重信息下，其所对应的历史货盘信息中重复出现次数最多的是哪些信息，这些信息可以作为影响历史偏重信息的显著因子，即当货盘信息中带有这些显著因子时，其相应的偏重信息容易确定，训练模型基于此原理根据显著因子对偏重信息进行预测，从而无需货主反复提供偏重信息。
所述步骤S6具体包括以下步骤：
S601、在下一次根据船东用户模型计算船东评价分数时，分析并提取货主用户发布的货盘信息中的显著因子，将提取的显著因子输入到训练模型中，获得预测偏重信息。
S602、基于预测偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
本实施例中，船东评价分数基于预测偏重信息加权计算获得，其一方面能够使船东评价分数更好地反映在不同货主的偏重区别下对于船东的评价等级，另一方面不需要货主反复提供偏重信息，既能使结果符合用户的需求，在这一过程中用户又是无感的，能够进一步提高用户的使用体验。
基于与前述实施例相同的发明构思，本发明另一实施例提供一种航运用户大数据建模分析系统，该系统应用于船期货盘匹配平台，所述船期货盘匹 配平台包括客户端和后台系统，所述客户端用于用户管理船期或货盘相关信息，所述用户包括船东用户和货主用户，所述后台系统与客户端通过网络相连接，用于实现船期与货盘信息的匹配和推送。参照图5，所述系统具体包括：
货主模型建立模块1，用于获取货主用户多维度活动记录数据，基于货主用户多维度活动记录数据建立货主用户模型，所述货主用户多维度活动记录数据包括货主第一维度活动记录数据和货主第二维度活动记录数据，所述货主第一维度活动记录数据为货主本地活动记录数据，货主第二维度活动记录数据为货主第三方活动记录数据；
船东模型建立模块2，用于获取船东用户多维度活动记录数据，基于船东用户多维度活动记录数据建立船东用户模型，所述船东用户多维度活动记录数据包括船东第一维度活动记录数据和船东第二维度活动记录数据，所述船东第一维度活动记录数据为船东本地活动记录数据，船东第二维度活动记录数据为船东第三方活动记录数据；
评价分级模块3，用于根据货主用户模型计算货主评价分数，根据货主评价分数对货主用户进行分级，还用于根据船东用户模型计算船东评价分数，根据船东评价分数对船东用户进行分级；
配对模块4，用于基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息。
可选地，所述评价分级模块3具体包括：
货主评价分级模块，用于获取货主多维度活动记录数据中各项活动记录数据的相应预设权重值，对货主多维度活动记录数据进行加权计算，获得货主评价分数；
船东评价分级模块，用于获取船东多维度活动记录数据中各项活动记录 数据的相应预设权重值，对船东多维度活动记录数据进行加权计算，获得船东评价分数。
可选地，所述配对模块4具体包括：
一次筛选模块，用于根据货主用户发布的货盘信息对船东用户发布的船期信息进行筛选，去除时间、载重不匹配的船期信息；
二次筛选模块，用于根据货主用户对船东用户的评价分数需求信息对经过筛选的船期信息进行二次筛选，去除船期信息所对应的船东用户评价分数低于评价分数需求的船期信息；
第一发送模块，用于获取二次筛选后的船期信息所对应的船东用户对货主用户的评价分数需求信息，判断货主用户的评价分数是否符合船东用户的分数需求信息要求，若符合，则向货主用户发送二次筛选后的船期信息；
第二发送模块，用于获取货主用户从二次筛选后的船期信息中所选择的船期信息，并向所选择的船期信息所对应的船东用户发送货盘信息。
可选地，所述系统还包括获取模块和调整模块。
所述获取模块用于获取货主用户对于船东用户多维度活动记录数据中各项活动记录数据的偏重信息；
所述调整模块用于根据偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
可选地，所述系统还包括训练模型建立模块和预测加权模块。
具体的，所述训练模型建立模块包括：
获取子模块，用于获取多个货主用户发布的历史货盘信息和对船东用户多维度活动记录数据中各项活动记录数据的历史偏重信息；
分析子模块，用于分析历史货盘信息和对应的历史偏重信息，确定历史货盘信息中的显著因子，所述显著因子为影响历史偏重信息的历史货盘信息 内容；
建模子模块，用于根据显著因子和历史偏重信息构建分类器，基于机器学习算法建立训练模型，所述训练模型用于根据货盘信息中的显著因子预测相应的偏重信息。
所述预测加权模块具体包括：
预测子模块，用于在下一次根据船东用户模型计算船东评价分数时，分析并提取货主用户发布的货盘信息中的显著因子，将提取的显著因子输入到训练模型中，获得预测偏重信息；
加权计算子模块，用于基于预测偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
所述系统实施例用于执行前述方法实施例所述的方法，其工作原理和技术效果可以参照前述方法实施例，在此不再赘述。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (9)

  1. 一种航运用户大数据建模分析方法，其特征在于，应用于船期货盘匹配平台，所述船期货盘匹配平台包括客户端和后台系统，所述客户端用于用户管理船期或货盘相关信息，所述用户包括船东用户和货主用户，所述后台系统与客户端通过网络相连接，用于实现船期与货盘信息的匹配和推送，所述方法包括以下步骤：
    S1、获取货主用户多维度活动记录数据，基于货主用户多维度活动记录数据建立货主用户模型；
    S2、获取船东用户多维度活动记录数据，基于船东用户多维度活动记录数据建立船东用户模型；
    S3、根据货主用户模型计算货主评价分数，根据货主评价分数对货主用户进行分级，根据船东用户模型计算船东评价分数，根据船东评价分数对船东用户进行分级；
    S4、基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息。
  2. 根据权利要求1所述的一种航运用户大数据建模分析方法，其特征在于，所述货主用户多维度活动记录数据包括货主第一维度活动记录数据和货主第二维度活动记录数据，所述货主第一维度活动记录数据为货主本地活动记录数据，货主第二维度活动记录数据为货主第三方活动记录数据；
    所述船东用户多维度活动记录数据包括船东第一维度活动记录数据和船东第二维度活动记录数据，所述船东第一维度活动记录数据为船东本地活动记录数据，船东第二维度活动记录数据为船东第三方活动记录数据。
  3. 根据权利要求1或2所述的一种航运用户大数据建模分析方法，其特征在于，步骤S3中，所述根据货主用户模型计算货主评价分数，具体包括：获取货主多维度活动记录数据中各项活动记录数据的相应预设权重值， 对货主多维度活动记录数据进行加权计算，获得货主评价分数；
    所述根据船东用户模型计算船东评价分数，具体包括：获取船东多维度活动记录数据中各项活动记录数据的相应预设权重值，对船东多维度活动记录数据进行加权计算，获得船东评价分数。
  4. 根据权利要求1所述的一种航运用户大数据建模分析方法，其特征在于，步骤S4中，所述基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息，具体包括：
    S401、根据货主用户发布的货盘信息对船东用户发布的船期信息进行筛选，去除时间、载重不匹配的船期信息；
    S402、根据货主用户对船东用户的评价分数需求信息对经过筛选的船期信息进行二次筛选，去除船期信息所对应的船东用户评价分数低于评价分数需求的船期信息；
    S403、获取二次筛选后的船期信息所对应的船东用户对货主用户的评价分数需求信息，判断货主用户的评价分数是否符合船东用户的分数需求信息要求，若符合，则向货主用户发送二次筛选后的船期信息；
    S404、获取货主用户从二次筛选后的船期信息中所选择的船期信息，并向所选择的船期信息所对应的船东用户发送货盘信息。
  5. 根据权利要求3所述的一种航运用户大数据建模分析方法，其特征在于，在步骤S2与步骤S3之间还包括步骤：
    S21、获取货主用户对于船东用户多维度活动记录数据中各项活动记录数据的偏重信息；
    S22、根据偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
  6. 根据权利要求5所述的一种航运用户大数据建模分析方法，其特征在于，所述方法还包括以下步骤：
    S5、根据历史货盘信息和历史偏重信息，基于机器学习算法建立用于预测偏重信息的训练模型；
    S6、根据货盘信息通过训练模型预测偏重信息，根据预测结果对船东多维度活动记录数据进行加权计算。
  7. 根据权利要求6所述的一种航运用户大数据建模分析方法，其特征在于，所述步骤S5具体包括以下步骤：
    S501、获取多个货主用户发布的历史货盘信息和对船东用户多维度活动记录数据中各项活动记录数据的历史偏重信息；
    S502、分析历史货盘信息和对应的历史偏重信息，确定历史货盘信息中的显著因子，所述显著因子为影响历史偏重信息的历史货盘信息内容；
    S503、根据显著因子和历史偏重信息构建分类器，基于机器学习算法建立训练模型，所述训练模型用于根据货盘信息中的显著因子预测相应的偏重信息。
  8. 根据权利要求7所述的一种航运用户大数据建模分析方法，其特征在于，所述步骤S6具体包括以下步骤：
    S601、在下一次根据船东用户模型计算船东评价分数时，分析并提取货主用户发布的货盘信息中的显著因子，将提取的显著因子输入到训练模型中，获得预测偏重信息；
    S602、基于预测偏重信息调整船东多维度活动记录数据中各项活动记录数据的相应预设权重值为临时权重值，基于临时权重值对船东多维度活动记录数据进行加权计算，获得船东评价分数。
  9. 一种航运用户大数据建模分析系统，其特征在于，应用于船期货盘匹配平台，所述船期货盘匹配平台包括客户端和后台系统，所述客户端用于 用户管理船期或货盘相关信息，所述用户包括船东用户和货主用户，所述后台系统与客户端通过网络相连接，用于实现船期与货盘信息的匹配和推送，所述系统具体包括：
    货主模型建立模块，用于获取货主用户多维度活动记录数据，基于货主用户多维度活动记录数据建立货主用户模型；
    船东模型建立模块，用于获取船东用户多维度活动记录数据，基于船东用户多维度活动记录数据建立船东用户模型；
    评价分级模块，用于根据货主用户模型计算货主评价分数，根据货主评价分数对货主用户进行分级，还用于根据船东用户模型计算船东评价分数，根据船东评价分数对船东用户进行分级；
    配对模块，用于基于货主用户分级和船东用户分级对货主用户发布的货盘信息与船东用户发布的船期信息进行配对，并向货主用户和船东用户推送配对信息。
PCT/CN2021/115832 2021-05-28 2021-08-31 一种航运用户大数据建模分析方法及系统 WO2022247041A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/594,093 US20240037485A1 (en) 2021-05-28 2021-08-31 Big data modeling and analyzing method and system for shipping user

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110591706.X 2021-05-28
CN202110591706.XA CN113312334B (zh) 2021-05-28 2021-05-28 一种航运用户大数据建模分析方法及系统

Publications (1)

Publication Number Publication Date
WO2022247041A1 true WO2022247041A1 (zh) 2022-12-01

Family

ID=77376215

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/115832 WO2022247041A1 (zh) 2021-05-28 2021-08-31 一种航运用户大数据建模分析方法及系统

Country Status (3)

Country Link
US (1) US20240037485A1 (zh)
CN (1) CN113312334B (zh)
WO (1) WO2022247041A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113312334B (zh) * 2021-05-28 2023-05-26 海南超船电子商务有限公司 一种航运用户大数据建模分析方法及系统
CN113656676A (zh) * 2021-08-16 2021-11-16 海南超船电子商务有限公司 一种船期智能匹配推送方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130117142A1 (en) * 2011-11-03 2013-05-09 Micky L. Thompson System and method of automatically matching cargo carriers to shippers
CN107622330A (zh) * 2017-09-25 2018-01-23 宜昌润江龙智能物流有限公司 一种长江流域船货物流分配方法及系统
CN109242044A (zh) * 2018-09-30 2019-01-18 江苏满运软件科技有限公司 车货匹配模型的训练方法、装置、存储介质及电子设备
CN110717717A (zh) * 2019-10-11 2020-01-21 惠龙易通国际物流股份有限公司 模型生成方法及系统、交通工具分配方法及装置
CN112435153A (zh) * 2020-11-26 2021-03-02 中远海运科技股份有限公司 一种运输船货搓合平台和方法
CN113312334A (zh) * 2021-05-28 2021-08-27 海南超船电子商务有限公司 一种航运用户大数据建模分析方法及系统

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103218700B (zh) * 2013-04-08 2016-03-09 镇江惠龙长江港务有限公司 一种货物与运力的线上智能配对方法
CN103278833B (zh) * 2013-05-13 2016-08-10 深圳先进技术研究院 一种基于北斗/gps数据的线路推荐系统及方法
CN104318413B (zh) * 2014-10-22 2018-01-09 湖南路联信息科技股份有限公司 物流信息匹配方法和系统
CN104766215B (zh) * 2015-04-13 2018-02-13 南京大学 一种综合性、多维度的货主选择量化方法
CN104866997A (zh) * 2015-06-10 2015-08-26 南京大学 一种用于车货在线配载的智能配对方法
CN105512845A (zh) * 2015-12-14 2016-04-20 杭州仕福信息科技有限公司 一种车船货运隐藏式报价系统及方法
US10467261B1 (en) * 2017-04-27 2019-11-05 Intuit Inc. Methods, systems, and computer program product for implementing real-time classification and recommendations
CN107358386A (zh) * 2017-06-23 2017-11-17 镇江五八到家供应链管理服务有限公司 一种货运司机与订单的匹配方法及匹配系统
CN107357852A (zh) * 2017-06-28 2017-11-17 镇江五八到家供应链管理服务有限公司 一种货运司机对订单意愿的判断方法
CN108510228B (zh) * 2018-04-02 2022-04-05 江苏国镖信息科技有限公司 一种公路运输车货智能匹配方法
US11830012B2 (en) * 2018-12-13 2023-11-28 Target Brands, Inc. System for U.S. customs compliance for overseas importers
CN109658033B (zh) * 2018-12-26 2021-03-16 江苏满运物流信息有限公司 货源路线相似度计算方法、系统、设备及存储介质
US20200380426A1 (en) * 2019-02-12 2020-12-03 Travel Labs, Inc. Systems and methods for creating and maintaining a secure traveler profile for curating travel itineraries
CN110119928B (zh) * 2019-05-07 2021-01-01 宏图物流股份有限公司 一种基于司机特征的车辆匹配推荐方法
CN112418575A (zh) * 2019-08-22 2021-02-26 刘畅 一种基于云计算和人工智能深度学习算法的期货主力合约量化择时决策系统
US10860115B1 (en) * 2019-09-19 2020-12-08 Bao Tran Air transportation systems and methods
CN110852599A (zh) * 2019-11-07 2020-02-28 南京大学 一种基于用户反馈的运输服务质量评价方法
CN112418758A (zh) * 2020-11-17 2021-02-26 国网电子商务有限公司 一种为货主智能推荐承运商的方法及系统
CN112365084A (zh) * 2020-11-26 2021-02-12 中远海运科技股份有限公司 船舶推荐系统和平台

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130117142A1 (en) * 2011-11-03 2013-05-09 Micky L. Thompson System and method of automatically matching cargo carriers to shippers
CN107622330A (zh) * 2017-09-25 2018-01-23 宜昌润江龙智能物流有限公司 一种长江流域船货物流分配方法及系统
CN109242044A (zh) * 2018-09-30 2019-01-18 江苏满运软件科技有限公司 车货匹配模型的训练方法、装置、存储介质及电子设备
CN110717717A (zh) * 2019-10-11 2020-01-21 惠龙易通国际物流股份有限公司 模型生成方法及系统、交通工具分配方法及装置
CN112435153A (zh) * 2020-11-26 2021-03-02 中远海运科技股份有限公司 一种运输船货搓合平台和方法
CN113312334A (zh) * 2021-05-28 2021-08-27 海南超船电子商务有限公司 一种航运用户大数据建模分析方法及系统

Also Published As

Publication number Publication date
CN113312334A (zh) 2021-08-27
CN113312334B (zh) 2023-05-26
US20240037485A1 (en) 2024-02-01

Similar Documents

Publication Publication Date Title
CN109345339B (zh) 电力行业垂直产业链一体化交易服务系统
WO2022247041A1 (zh) 一种航运用户大数据建模分析方法及系统
US7676390B2 (en) Techniques for performing business analysis based on incomplete and/or stage-based data
CN109767032A (zh) 一种基于数据分析的企业财务运营数字化管理优化系统
Awasthi et al. A combined approach integrating gap analysis, QFD and AHP for improving logistics service quality
Rathore et al. Impact of risks in foodgrains transportation system: a system dynamics approach
CN113537807B (zh) 一种企业智慧风控方法及设备
KR20170099078A (ko) 생산관리시스템(ｍｅｓ)의 제조생산설비 수집정보를 활용한 빅데이터 기반 경영예측 플랫폼 시스템
Makarova et al. Dealer-service center competitiveness increase using modern management methods
TWI612485B (zh) 產品服務系統分類與服務轉型方法
Dhanorkar et al. An empirical investigation of transaction dynamics in online surplus networks: A complex adaptive system perspective
Qiu et al. Multiproduct firms, export product scope, and trade liberalization: The role of managerial efficiency
Cheong et al. The rise of accounting: Making accounting information relevant again with exogenous data
Wei [Retracted] A Machine Learning Algorithm for Supplier Credit Risk Assessment Based on Supply Chain Management
CN113421014A (zh) 一种目标企业确定方法、装置、设备和存储介质
Wang An analysis of the optimal customer clusters using dynamic multi-objective decision
CN114819820A (zh) 基于人工智能的智慧供应链管理方法、装置、系统及介质
KR102499182B1 (ko) 인공지능을 이용한 가계대출 사기/부실 상시감사지원시스템
Yembergenov et al. Management accounting in the restaurant business: Organization methodology
Zhang et al. Port capability evaluation from the perspective of supply chain
Nyman An exploratory study of supply chain management it solutions
CN112990550A (zh) 企业估值方法、装置、电子设备及介质
CN115115322A (zh) 目标群组识别方法、风险评估方法、装置、设备及介质
Larson et al. The promise of information sharing and the peril of information overload
Wood et al. Expanding sales and operations planning using sentiment analysis: Demand and sales clarity from social media

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 17594093

Country of ref document: US

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21942608

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21942608

Country of ref document: EP

Kind code of ref document: A1