CN109583796A - A kind of data digging system and method for Logistics Park OA operation analysis - Google Patents
A kind of data digging system and method for Logistics Park OA operation analysis Download PDFInfo
- Publication number
- CN109583796A CN109583796A CN201910016603.3A CN201910016603A CN109583796A CN 109583796 A CN109583796 A CN 109583796A CN 201910016603 A CN201910016603 A CN 201910016603A CN 109583796 A CN109583796 A CN 109583796A
- Authority
- CN
- China
- Prior art keywords
- data
- demand
- module
- model
- achievement
- 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
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
- G06Q10/06—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
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of data digging systems and method for Logistics Park OA operation analysis, it includes data acquisition module, data preprocessing module, data memory module, data warehouse module, data mining engine module and result display module, data collection system is for acquiring the first achievement data, and data preprocessing module is for pre-processing the first achievement data；Data warehouse module and data memory module are electrically connected, data mining engine module includes business diagnosis module, data mining model library and data mining execution module, second-level model is ranked up to form alternative model queue by data mining model library, and data mining execution module executes data mining algorithm according to second-level model.Second-level model is ranked up to form alternative model queue by the present invention by data mining model library, data mining algorithm is executed according to second-level model by data mining execution module again, to obtain the corresponding achievement data of expected operations objective, to facilitate garden operator to carry out decision.
Description
Technical field
The present invention relates to data mining technology fields, dig more particularly, to a kind of data for Logistics Park OA operation analysis
Dig system and method.
Background technique
Logistics Park provides a variety of services to enter the loglstics enterprise of garden, such as provides logistics information service platform, object
Stream enterprise can obtain delivery side's information, transit route information from platform, while can also (oneself has by the capacity information of oneself
How many free time lorries, will when dispatch a car, route of dispatching a car etc.) on platform, loglstics enterprise utilizes these information energy for publication
It enough reduces the rate of empty ride of lorry, improve income.
Due to often there is the problems such as transaction for the first time, not enough trust between each other, garden between delivery side and loglstics enterprise
Area's operator is many times being used as third party guarantees, ensures that both parties' is fair and just.Therefore, Logistics Park operator needs
It is to be understood that the much informations such as the prestige of its services client, management state, capital turnover situation, reduce the risk of oneself；It solves simultaneously
Needed for client is, more value-added services are provided for client, reach the situation of two-win.
By the retrieval to existing literature, it was found that two kinds of service plans, first is that how to construct logistic information systems, i.e. line
The electronization and networking of lower logistics operating process；Second is that being closed from how enterprise's angle utilizes data mining technology to improve client
System, it is common that the existing data mining scheme for logistic industry considers deficiency to the otherness and complexity of data acquisition,
Data are substantially acquired from the table in database, and in reality, fraction of stream enterprise remains in hand filling document
Stage.
Summary of the invention
Based on this, it is necessary in view of the deficiencies of the prior art, provide a kind of data digging for Logistics Park OA operation analysis
System and method is dug, the collecting method towards papery data source is taken, passes through the user being arranged in business diagnosis module
Requirement validation component identifies the demand of user, the demand of user is identified by natural language processing technique, facilitates garden operation
Fang Jinhang decision.
In order to solve the above technical problems, the technical scheme adopted by the invention is that: one kind being used for Logistics Park OA operation analysis
Data digging system comprising data acquisition module, data preprocessing module, data memory module, data warehouse module, number
According to engine modules and result display module is excavated, for the data collection system for acquiring the first achievement data, the data are pre-
Processing module is used to after pre-processing the first achievement data of data collecting module collected be converted to the index of reference format
Data；The data memory module carries out the index number of the reference format after data prediction for data cached preprocessing module
According to；The data warehouse module and data memory module are electrically connected, and the data warehouse module stores mould for storing data
The achievement data for the reference format that block is sent；The data mining engine module includes business diagnosis module, data mining
The achievement data stored in data warehouse module is classified as by model library and data mining execution module, the business diagnosis module
Loglstics enterprise operations risks, which excavate demand, value-added service precision marketing in garden excavates demand and garden service ability is promoted to excavate needs
Ask, the business diagnosis module by loglstics enterprise operations risks excavate demand, garden value-added service precision marketing excavate demand and
Garden service ability promotes the corresponding achievement data of excavation demand and is mapped in data mining model library, the data mining
Model library includes specific data mining model, and the specific data mining model is made of first-level model, the first-level model by
Second-level model composition, the second-level model under first-level model is ranked up to form alternative model team by the data mining model library
Column, the data mining execution module, which is used to extract logistics enterprise operation risk from data warehouse module, excavates demand, garden
Value-added service precision marketing excavates demand or garden service ability promotes the corresponding achievement data of excavation demand, the data
It excavates execution module and data mining algorithm is executed according to second-level model, the result display module holds data mining execution module
The achievement data obtained after row data mining algorithm shows user.
The method that the data preprocessing module carries out data prediction in one of the embodiments, includes following step
It is rapid:
(1.1), the first achievement data duplicate removal of acquisition and supplement data cleansing: are omitted into achievement data, identification and modification
Abnormal index data complete data cleansing process；
(1.2), data integration: the first achievement data after data cleansing is merged to form third achievement data collection
It closes, completes data integration process；
(1.3), data convert: third achievement data being standardized by modes such as data smoothing, normalization processing
Processing, in the form of the achievement data for the data digging system for being transformed to be suitable for Logistics Park OA operation analysis；
(1.4), data degradation: the achievement data after being transformed the data by PCA Principal Component Analysis carries out dimensionality reduction,
Remove and contribute small dimension variable, to form the achievement data of reference format.
The data mining model library passes through decision tree, neural network or support vector machines in one of the embodiments,
Algorithm excavates the corresponding achievement data of demand to loglstics enterprise operations risks and handles, and the data mining model library passes through pass
Connection rule or collaborative filtering excavate the corresponding achievement data of demand to garden value-added service precision marketing and handle, described
Data mining model library promotes excavation demand pair to garden service ability by decision tree, neural network or algorithm of support vector machine
The achievement data answered is handled.
User demand confirmation component is additionally provided in the business diagnosis module in one of the embodiments, for knowing
The demand of other user.
The method of the user demand confirmation component demand of user for identification includes such as in one of the embodiments,
Lower step:
(2.1), it obtains user and inputs requirement command；
(2.2), natural language processing is carried out to the requirement command of user's input；Wherein, natural language processing include participle,
Form the modes such as term vector；
(2.3), preset standard is intended to template term vector, does Semantic Similarity Measurement based on term vector, obtains defeated to user
The corresponding subdivision demand of the requirement command entered；Wherein, standards sought template term vector is excavated corresponding to loglstics enterprise operations risks
Demand, garden value-added service precision marketing excavate the subdivision demand in demand or garden service ability promotion excavation demand；
(2.4), feedback subdivision demand is to user；
(2.5), user confirms the corresponding subdivision demand of requirement command.
The first-level model is to pass through correlation rule, decision tree, neural network, cluster, branch in one of the embodiments,
Vector machine or collaborative filtering are held to the achievement data that stores in the data warehouse module data set that carries out that treated, described two
Grade model is to be calculated by the classification in correlation rule, decision tree, neural network, cluster, support vector machines or collaborative filtering
Method is to the achievement data that stores in the data warehouse module data set that carries out that treated.
The second-level model by under first-level model is ranked up to form alternative model queue in one of the embodiments,
Method include the following steps:
(3.1), loglstics enterprise operations risks are obtained and excavate demand, garden value-added service precision marketing excavation demand and garden
Service ability promotes the corresponding achievement data of one type in excavation demand；
(3.2), the corresponding number of each second-level model is obtained after being handled the achievement data of the acquisition in step (3.1)
According to collection；
(3.3), the data set in step (3.2) is divided into training set and test set, to second-level model on training set
Expansion training, is verified, the recall rate index, accurate rate index, stability for recording each second-level model refer on test set
Mark, recall rate index, accurate rate index, stability indicator value are in [0,1]；
(3.4), recall rate index, accurate rate index, the weight of stability indicator are obtained by analytic hierarchy process (AHP), definition is called together
The rate of returning, accurate rate, stability weight use α, β and γ to represent respectively, then the score score=α recall+ of each second-level model
βprecisioin+γPSI；
(3.4), it is ranked up by the score size of each second-level model, obtains alternative model queue.
Specific step is as follows for the analytic hierarchy process (AHP) in the step (3.4) in one of the embodiments,
(3.4.1), Analytic Hierarchy Process Model is established；The method for establishing Analytic Hierarchy Process Model includes the following steps:
(3.4.1.1) establishes top, and user demand instructs corresponding subdivision demand for identification；
(3.4.1.2), middle layer is established, the recall rate index, accurate rate index, stability for obtaining second-level model refer to
Mark；
(3.4.1.3), the bottom is established, the concrete scheme for the user demand instruction that achieves a solution；
(3.4.2), construction pairwise comparison matrix；
(3.4.3), recall rate, accurate rate, the weight of stability are calculated and consistency check is done to pairwise comparison matrix.
The data memory module is the computer cluster constructed using Hadoop technology in one of the embodiments,.
A kind of data digging method for Logistics Park OA operation analysis comprising following steps:
It obtains the loglstics enterprise operations risks stored in data warehouse module and excavates demand, garden value-added service precision marketing
Excavation demand and garden service ability promote the corresponding achievement data of excavation demand；
The corresponding data set of each second-level model is obtained after the achievement data of acquisition is handled；
Data set is divided into training set and test set, second-level model is unfolded to train on training set, on test set
It is verified, records recall rate index, the accurate rate index, stability indicator of each second-level model；
Recall rate index, accurate rate index, the weight of stability indicator are obtained by analytic hierarchy process (AHP)；
It is ranked up by the score size of each second-level model, obtains alternative model queue；
The second-level model of highest scoring in alternative model queue is selected to carry out data mining for initial model；
Obtain the corresponding achievement data of expected operations objective.
In conclusion a kind of data digging system and method for Logistics Park OA operation analysis of the present invention is adopted by data
Collect modules acquiring data, cooperation is ranked up the second-level model under first-level model to form alternative mould using data mining model library
Type queue, then data mining algorithm is executed according to second-level model by data mining execution module, to obtain expected operations objective
Corresponding achievement data, to facilitate garden operator to carry out decision.
Detailed description of the invention
A kind of structure principle chart of the data digging system for Logistics Park OA operation analysis of Fig. 1 present invention；
Fig. 2 is that user demand confirms that component is known in a kind of data digging system for Logistics Park OA operation analysis of the present invention
The process principle figure of the demand of other user；
Fig. 3 is a kind of process principle figure of the data digging method for Logistics Park OA operation analysis of the present invention.
Specific embodiment
To further understand the features of the present invention, technological means and specific purposes achieved, function, below with reference to
Present invention is further described in detail for the drawings and the specific embodiments.
As shown in Figure 1 to Figure 3, a kind of data digging system for Logistics Park OA operation analysis of the present invention includes that data are adopted
Collect module, data preprocessing module, data memory module, data warehouse module, data mining engine module and result and shows mould
Block, the data collection system are used to acquire enterprise name, enterprises registration address, enterprise's year business including loglstics enterprise and receive
Enter, enterprise's year volume, annual garden rent payment promise breaking record, annual garden property payment promise breaking record, annual cargo
Breakage record, year complain record, annual personnel rate, annual fuel consumption tonnage, loglstics enterprise debt, client by the owner of cargo
The first achievement data including the rate of complaints index, first achievement data is according to actual needs in one of the embodiments,
It can also be other achievement datas relevant to loglstics enterprise operation.
The collection process of the data acquisition module is divided into two classes in one of the embodiments: if loglstics enterprise exists
All kinds of logistic information systems, such as order management system, vehicle dispatch system, Warehouse Management System etc. are used in production, number
According to acquisition module data can be extracted from such logistic information systems；If the management data of loglstics enterprise is from the number such as bill
According to source, then data acquisition module is needed through devices collect datas such as high speed scanners, can also be by being based on OCR image recognition skill
Art is simultaneously converted into the identifiable document format of computer and is acquired.
The data preprocessing module after pre-processing to the first achievement data of data collecting module collected for turning
It is changed to the achievement data of reference format；Specifically, the method that the data preprocessing module carries out data prediction includes following
Step:
(1.1), the first achievement data duplicate removal of acquisition and supplement data cleansing: are omitted into achievement data, identification and modification
Abnormal index data complete data cleansing process；
(1.2), data integration: the first achievement data after data cleansing is merged to form third achievement data collection
It closes, completes data integration process；Specifically, it will be merged from the first achievement data of different data sources, so that through number
Third achievement data is integrated into according to the first achievement data after cleaning；
(1.3), data convert: third achievement data being standardized by modes such as data smoothing, normalization processing
Processing, in the form of the achievement data for the data digging system for being transformed to be suitable for Logistics Park OA operation analysis；
(1.4), data degradation: the achievement data after being transformed the data by PCA Principal Component Analysis carries out dimensionality reduction,
Remove and contribute small dimension variable, to form the achievement data of reference format.
The data memory module is the computer cluster constructed using Hadoop technology, and the data memory module is used for
Data cached preprocessing module carries out the achievement data of the reference format after data prediction；The data warehouse module and data
Memory module is electrically connected, the index of the data warehouse module reference format that memory module is sent for storing data
Data.
The data mining engine module includes that business diagnosis module, data mining model library and data mining execute mould
Block, the achievement data stored in data warehouse module is classified as the excavation of loglstics enterprise operations risks by the business diagnosis module to be needed
It asks, value-added service precision marketing in garden excavates demand and garden service ability promotes excavation demand.
The operations risks of the corresponding achievement data prediction loglstics enterprise of demand can be excavated according to loglstics enterprise operations risks
Grade specifically by analyzing achievement datas such as loglstics enterprise debt, customer complaint rates, predicts the loglstics enterprise
Operations risks grade, wherein the operations risks grade of loglstics enterprise is divided into D grades, C grades, B grades and A grades, and D grades, D grades of expression logistics
Stable, the following profitability of enterprise operation is persistently had an optimistic view of, the higher, client in D grades every transport value of goods of loglstics enterprise
The rate of complaints is low, the freight charges returned money account phase is very short；C grades of expression logistics enterprise operations are stable but business increases gently, in C grades of logistics
Every transport value of goods of enterprise is general, customer complaint rate is general, the freight charges returned money account phase is shorter；B grades of expression logistics enterprise operations
The unstable but following profitability is expected to improve, the higher, customer complaint in B grades every transport value of goods of loglstics enterprise
Rate is higher, the freight charges returned money account phase is longer；A grades of expression logistics enterprise operations are unstable, the following profitability is also more pessimistic, are in B
Transport value of goods is lower, customer complaint rate is higher, the freight charges returned money account phase is longer, exists and is in arrears with loan for the loglstics enterprise every of grade
Risk；Factually border operational effect voluntarily determines every transport value of goods of loglstics enterprise, customer complaint rate, fortune to zone logistics root
The judgment mode for taking returned money account phase length, for the loglstics enterprise of different operations risks grades, garden operator can take difference
Strategy, such as be C grades of loglstics enterprise to operations risks grade, if their lease will expire, garden operator can
To renew with this kind of loglstics enterprise, while business experience can be assigned rich by garden for this high unfavorable factor of customer complaint rate
Rich third-party logistics Consulting Company gives professional guidance, helps loglstics enterprise to reduce this index, can reach garden and logistics
The two-win of enterprise.
Loglstics enterprise pair can be obtained by excavating the corresponding achievement data of demand by analysis garden value-added service precision marketing
The value-added service that garden operator provides such as driver recruits in generation, the interest-degree that gas filling card is handled etc., thus help garden operator into
Row precision marketing；Promoting the corresponding achievement data of excavation demand by garden service ability predicts loglstics enterprise to garden operator
The opinion rating of service level helps garden operator to improve service quality.
Loglstics enterprise operations risks are excavated demand by the business diagnosis module, value-added service precision marketing in garden excavates need to
It asks and garden service ability promotes the corresponding achievement data of excavation demand and is mapped in data mining model library, specifically,
The data mining model library can excavate loglstics enterprise operations risks by decision tree, neural network or algorithm of support vector machine
The corresponding achievement data of demand is handled, and the data mining model library can be by correlation rule or collaborative filtering to garden
Value-added service precision marketing in area's excavates the corresponding achievement data of demand and is handled, and the data mining model library can pass through decision
Tree, neural network or algorithm of support vector machine promote the corresponding achievement data of excavation demand to garden service ability and handle.
User demand confirmation component is additionally provided in the business diagnosis module, for identification the demand of user.Specifically,
The method of the user demand confirmation component demand of user for identification includes the following steps:
(2.1), it obtains user and inputs requirement command；In the present invention, user is referred to as garden operator；
(2.2), natural language processing is carried out to the requirement command of user's input；Wherein, natural language processing include participle,
Form the modes such as term vector；
(2.3), preset standard is intended to template term vector, does Semantic Similarity Measurement based on term vector, obtains defeated to user
The corresponding subdivision demand of the requirement command entered；Wherein, standards sought template term vector is excavated corresponding to loglstics enterprise operations risks
Demand, garden value-added service precision marketing excavate the subdivision demand in demand or garden service ability promotion excavation demand；
(2.4), feedback subdivision demand is to user；
(2.5), user confirms the corresponding subdivision demand of requirement command.
It is excavated in one of the embodiments, when the requirement command that can not judge user belongs to loglstics enterprise operations risks
When demand, garden value-added service precision marketing excavate the one type in demand or garden service ability promotion excavation demand, garden
Area's operator is confirmed by communication way under line with user.
It is accurate for loglstics enterprise operations risks excavation demand, garden value-added service in advance in one of the embodiments,
Excavation demand of marketing or garden service ability promote these three total demands of excavation demand, more segment under every kind of demand of formation
Demand.For example, again may include operations risks grade, the something in certain loglstics enterprise this year under loglstics enterprise operations risks excavation demand
Stream appearance of enterprise rent pays the suchlike subdivision demand such as overdue risk class, constructs standard meaning for each subdivision demand
Artwork plate term vector, standards sought template are standardization, the instruction of writtenization statement.
When the input of user instruction is the risk class of this year " in prediction garden certain loglstics enterprise ", user demand confirms
Component carries out natural language processing by the requirement command that inputs to user, specifically, user demand confirm component by with
The requirement command of family input is segmented, and obtains the term vector that user inputs requirement command using TF-IDF algorithm, and user is defeated
The term vector and standards sought template term vector for entering requirement command carry out Semantic Similarity Measurement, take the mark with highest similarity
Standard is intended to template, and then identifies the subdivision demand of user, then requirement command of the business diagnosis module feedback after analyzing and determining
Corresponding achievement data allows user to confirm.
The business diagnosis module records the requirement command problem of each user's input, packet in one of the embodiments,
It includes the problem of accurately identifying and fails the problem of identification judges, constantly progress machine learning is improved to user demand instruction issue
Recognition capability.
The data mining model library includes specific data mining model, and the specific data mining model is by first-level model
It constitutes, the first-level model is made of second-level model, and specifically, the first-level model is to pass through correlation rule, decision tree, nerve
Treated to the achievement data progress stored in data warehouse module for network, cluster, support vector machines or collaborative filtering
Data set, the second-level model are to be calculated by correlation rule, decision tree, neural network, cluster, support vector machines or collaborative filtering
Sorting algorithm in method is to the achievement data stored in data warehouse module the data set that carries out that treated.
In one of the embodiments, the second-level model be by ID3 algorithm in decision Tree algorithms, C4.5 algorithm,
The sorting algorithms such as CART algorithm are to the achievement data that stores in the data warehouse module data set that carries out that treated；Wherein, data
Mining model library be it is open, user can also be added during using data mining model library from third-party two
Grade model.
Second-level model under first-level model is ranked up to form " alternative model queue " by the data mining model library,
In, the second-level model under first-level model is ranked up to the method to form " alternative model queue " and is included the following steps:
(3.1), loglstics enterprise operations risks are obtained and excavate demand, garden value-added service precision marketing excavation demand and garden
Service ability promotes the corresponding achievement data of one type in excavation demand；
(3.2), the corresponding number of each second-level model is obtained after being handled the achievement data of the acquisition in step (3.1)
According to collection；
(3.3), the data set in step (3.2) is divided into training set and test set in proportion, wherein at the beginning of training set
Beginning accounting is 0.8, and the initial accounting of test set is 0.2, is unfolded to train to second-level model on training set, carry out on test set
Verifying, record each second-level model recall rate (Recall) index, accurate rate (Precision) index, stability (PSI,
Population Stability Index) index, recall rate index, accurate rate index, stability indicator value are in [0,1]
It is interior；Specifically, each round reduces 0.1 to training set accounting after second-level model expansion training on training set, until training set accounts for
Than being 0.5, i.e. fourth round training set accounting when training is unfolded to second-level model on training set is 0.5；
(3.4), recall rate index, essence are obtained by analytic hierarchy process (AHP) (AHP, Analytic Hierarchy Process)
True rate index, the weight of stability indicator, define recall rate, accurate rate, stability weight use α, β and γ to represent respectively, then
The score score=α recall+ β precisioin+ γ PSI of each second-level model；
(3.4), it is ranked up, obtains " alternative model queue " by the score size of each second-level model.
Specific step is as follows for the analytic hierarchy process (AHP) in the step (3.4) in one of the embodiments:
(3.4.1), Analytic Hierarchy Process Model is established；Specifically, on the basis of analyzing practical problem, related factor is pressed
It is divided into several levels from top to down according to different attribute, all factors of same layer are subordinated to one layer of factor or to influence
Upper layer factor, while also dominating next layer of factor or being influenced by lower layer factors, the method for establishing Analytic Hierarchy Process Model
Include the following steps:
(3.4.1.1) establishes top, and user demand instructs corresponding subdivision demand for identification；
(3.4.1.2), middle layer is established, the recall rate index, accurate rate index, stability for obtaining second-level model refer to
Mark；
(3.4.1.3), the bottom is established, for the concrete scheme for the user demand instruction that achieves a solution, in the present invention, most
Bottom is the second-level model under first-level model；
(3.4.2), construction pairwise comparison matrix；Since the 2nd layer of Analytic Hierarchy Process Model, on being subordinated to or influencing
The same layer factors of one layer of each factor compare scale building pairwise comparison matrix with Paired Comparisons and 1-9, until most lower
Layer；
When comparing the importance of i-th of element upper one layer of some factor opposite with j-th of element, the phase of usage quantity
To weight mijTo describe；If shared n element participation is compared, then M=(mij)n*nReferred to as pairwise comparison matrix；Compare pairs of
M in matrixijValue can refer to by following scales carry out assignment；mijIn 1-9 and its intermediate value reciprocal；
mij=1, element i are identical with importance of the element j to upper level factor；
mij=3, element i is slightly more important than element j；
mij=5, element i is more important than element j；
mij=7, element i are than element j much more significant；
mij=9, element i is more of crucial importance than element j；
mij=2n, n=1,2,3,4, the importance of element i and element j is between mij=2n-1 and mijBetween=2n+1；
mij=1/n, n=1,2 ..., 9, and if only if mji=n；
In the present embodiment, more valued due to garden operator and be accurately identified the object of different classes of operations risks grade
Enterprise is flowed, therefore recall rate is more important than accurate rate, is finally only stability, to obtain the importance of three indexs successively are as follows: call together
Rate > accurate rate > stability is returned, it is m that recall rate is enabled in pairwise comparison matrix11, accurate rate m22, stability m33；
Construct a third-order matrix Mij(3*3) as follows, wherein m12=3, indicate that recall rate is slightly heavy compared to accurate rate
It wants, m13=5, indicate that recall rate is important compared to stability, m23=3, indicate that accurate rate is slightly more important than stability.To obtain three
Rank matrix is as follows:
(3.4.3), recall rate, accurate rate, the weight of stability are calculated and consistency check is done to pairwise comparison matrix M；
If M is completely the same pairwise comparison matrix, Ying You mijmjk=mik；By analysis it is found that completely the same pairs of
Comparator matrix, the characteristic value of maximum absolute value are equal to the dimension of the matrix；The step of examining pairwise comparison matrix M consistency is such as
Under,
Calculate the index of matrix M (rank of n > 1 square matrix) inconsistent degree in contrast with measuring oneWherein λmax
(M) be matrix M maximum eigenvalue, n be order of matrix number；By calculating, λmaxBe 3.0385, feature vector be (0.6370,
0.2583,0.1047), i.e., recall rate, accurate rate, stability weight are respectively 0.637,0.258 and 0.104,Wherein, CI=0 has complete consistency；CI has satisfied consistency close to 0；Conversely,
CI is bigger, inconsistent more serious；
The random consistency ratio of pairwise comparison matrix M is calculated by formulaRI is known as mean random consistency and refers to
Mark, it is only related with order of matrix number, and the standard RI for examining pairwise comparison matrix M consistency can be found from available data；Wherein,
Examine the standard RI judgment method of pairwise comparison matrix M consistency as follows,
When CR < 0.1, determine that pairwise comparison matrix M has acceptable consistency, in other words its inconsistent degree be can be with
Receive；When CR >=0.1, then need to adjust pairwise comparison matrix M, until it reaches acceptable consistency；It tables look-up
The corresponding RI of third-order matrix is 0.58 out, therefore is had:Illustrate the standard of pairwise comparison matrix M consistency
RI passes through inspection.
The data mining execution module be used for from data warehouse module extract logistics enterprise operation risk excavate demand,
Value-added service precision marketing in garden excavates demand or garden service ability promotes the corresponding achievement data of excavation demand, described
Data mining execution module executes data mining algorithm according to second-level model.
The data mining execution module executes data mining algorithm packet according to second-level model in one of the embodiments,
Include following steps:
(4.1), the second-level model for selecting highest scoring in " alternative model queue " is initial model, carries out data mining simultaneously
The management level that result is applied to garden is improved；
(4.2), the corresponding achievement data of expected operations objective is obtained, by the corresponding achievement data of practical operations objective and in advance
The corresponding achievement data of phase operations objective compares, and obtains gap value；Wherein, demand is excavated for loglstics enterprise operations risks
Corresponding achievement data, it is contemplated that operations objective can be set as: actual operation risk increases or reduced degree, such as loglstics enterprise
Operations risks grade；The corresponding achievement data of demand is excavated for garden value-added service precision marketing, it is contemplated that operations objective can be set
Are as follows: value-added service bring gross income；The corresponding achievement data of excavation demand is promoted for garden service ability, it is contemplated that manage
Target can be set as: satisfaction of the garden enterprise to garden operator；Garden operator is in the number for Logistics Park OA operation analysis
According to two parameters of setting in digging system: (a) the validity check period, as unit of month；(b) expected operations objective；
(4.3), improve actual effect according to the management level of garden to be adjusted initial model；Specifically, an effect
After fruit round of visits, when gap value within a preset range, the period of validity check next time execute data mining algorithm when after
It is continuous to use last second-level model；When gap value exceeds preset range, then the period of validity check next time executes data mining
Algorithm uses the second-level model that sequence is taken second place in " alternative model queue " instead；When all second-level models quilt in " alternative model queue "
After, gap value still exceeds preset range, and garden operator is in the data digging system for Logistics Park OA operation analysis
Two parameters are reset, are executed step (4.2).
Data mining execution module is executed the expection obtained after data mining algorithm and manages mesh by the result display module
It marks corresponding achievement data and shows user, user can be showed in the form of line chart, pie chart, histogram etc., also can access
Third party's figure plug-in unit is expected the corresponding achievement data of operations objective according to user preferences from a variety of different dimensions, into
And facilitate user to the corresponding achievement data of expected operations objective carry out deeper into understanding and analysis.
Preferably to illustrate the present invention, illustrate the present invention for Logistics Park OA operation analysis by a specific example
Data digging system the course of work.
If a Logistics Park has some loglstics enterprises for being engaged in highway transportation, each not phase of these loglstics enterprise management levels
Together, garden operator needs to evaluate the management level of these loglstics enterprises, so as to adjust the operation plan of garden operator
Slightly, the management tactics of garden operator includes deciding whether that overdue loglstics enterprise is renewed lease, in garden with the rental period
Whether the rent of loglstics enterprise a series of decisions such as is adjusted.
A kind of data digging method for Logistics Park OA operation analysis, includes the following steps:
It obtains the loglstics enterprise operations risks stored in data warehouse module and excavates demand, garden value-added service precision marketing
Excavation demand and garden service ability promote the corresponding achievement data of excavation demand；
The corresponding data set of each second-level model is obtained after the achievement data of acquisition is handled；
Data set is divided into training set and test set, second-level model is unfolded to train on training set, on test set
It is verified, records recall rate index, the accurate rate index, stability indicator of each second-level model；
Recall rate index, accurate rate index, the weight of stability indicator are obtained by analytic hierarchy process (AHP)；
It is ranked up, obtains " alternative model queue " by the score size of each second-level model；
The second-level model of highest scoring in " alternative model queue " is selected to carry out data mining for initial model, wherein one
Grade model is by decision Tree algorithms to the achievement data that stores in data warehouse module carry out that treated data set, second level mould
Type is by sorting algorithms such as ID3 algorithm, C4.5 algorithm, CART algorithms in decision Tree algorithms to storing in data warehouse module
Achievement data carry out treated data set；
Obtain the corresponding achievement data of expected operations objective, wherein it is expected that the corresponding achievement data of operations objective is logistics
The operations risks grade of enterprise.
In conclusion a kind of data digging system and method for Logistics Park OA operation analysis of the present invention is adopted by data
Collect modules acquiring data, cooperation is ranked up the second-level model under first-level model to form alternative mould using data mining model library
Type queue, then data mining algorithm is executed according to second-level model by data mining execution module, to obtain expected operations objective
Corresponding achievement data, to facilitate garden operator to carry out decision.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
Limitation of the scope of the invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art,
Without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection model of the invention
It encloses.Therefore, protection scope of the present invention should be determined by the appended claims.
Claims (10)
1. a kind of data digging system for Logistics Park OA operation analysis, it is characterised in that: including data acquisition module, data
Preprocessing module, data memory module, data warehouse module, data mining engine module and result display module, the data
Acquisition system is used to refer to the first of data collecting module collected for acquiring the first achievement data, the data preprocessing module
Mark data are converted to the achievement data of reference format after being pre-processed；The data memory module is used for data cached pretreatment
Module carries out the achievement data of the reference format after data prediction；The data warehouse module electrically connects with data memory module
It connects, the achievement data of the data warehouse module reference format that memory module is sent for storing data；The data
Excavating engine modules includes business diagnosis module, data mining model library and data mining execution module, the business diagnosis mould
The achievement data stored in data warehouse module is classified as loglstics enterprise operations risks and excavates demand, garden value-added service essence by block
Quasi- marketing excavation demand and garden service ability promote excavation demand, and the business diagnosis module digs loglstics enterprise operations risks
Pick demand, garden value-added service precision marketing excavate demand and garden service ability promotes the corresponding index number of excavation demand
According to being mapped in data mining model library, the data mining model library includes specific data mining model, the specific data
Mining model is made of first-level model, and the first-level model is made of second-level model, and the data mining model library is by level-one mould
Second-level model under type is ranked up to form alternative model queue, and the data mining execution module is used for from data warehouse module
Middle extraction logistics enterprise operation risk excavates demand, value-added service precision marketing in garden excavates demand or garden service ability is promoted
The corresponding achievement data of excavation demand, the data mining execution module execute data mining algorithm according to second-level model,
Data mining execution module is executed the achievement data obtained after data mining algorithm and shows user by the result display module.
2. a kind of data digging system for Logistics Park OA operation analysis according to claim 1, which is characterized in that institute
State data preprocessing module carry out data prediction method the following steps are included:
(1.1), it is abnormal that the first achievement data duplicate removal of acquisition and supplement data cleansing: are omitted into achievement data, identification and modification
Achievement data completes data cleansing process；
(1.2), data integration: the first achievement data after data cleansing being merged to form third achievement data set,
Complete data integration process；
(1.3), data convert: third achievement data is subjected to standardization processing by modes such as data smoothing, normalization processing,
In the form of the achievement data for the data digging system for being transformed to be suitable for Logistics Park OA operation analysis；
(1.4), data degradation: the achievement data after being transformed the data by PCA Principal Component Analysis carries out dimensionality reduction, removes
Small dimension variable is contributed, to form the achievement data of reference format.
3. a kind of data digging system for Logistics Park OA operation analysis according to claim 1, it is characterised in that: institute
It states data mining model library and demand is excavated to loglstics enterprise operations risks by decision tree, neural network or algorithm of support vector machine
Corresponding achievement data is handled, and the data mining model library rises in value to garden by correlation rule or collaborative filtering
Service precision marketing excavates the corresponding achievement data of demand and is handled, and the data mining model library passes through decision tree, nerve
Network or algorithm of support vector machine promote the corresponding achievement data of excavation demand to garden service ability and handle.
4. a kind of data digging system for Logistics Park OA operation analysis according to claim 1, it is characterised in that: institute
It states and is additionally provided with user demand confirmation component in business diagnosis module, for identification the demand of user.
5. a kind of data digging system for Logistics Park OA operation analysis according to claim 4, it is characterised in that: institute
The method for stating the user demand confirmation component demand of user for identification includes the following steps:
(2.1), it obtains user and inputs requirement command；
(2.2), natural language processing is carried out to the requirement command of user's input；Wherein, natural language processing includes participle, is formed
Term vector mode；
(2.3), preset standard is intended to template term vector, does Semantic Similarity Measurement based on term vector, what acquisition inputted user
The corresponding subdivision demand of requirement command；Wherein, standards sought template term vector correspond to loglstics enterprise operations risks excavate demand,
Value-added service precision marketing in garden excavates the subdivision demand in demand or garden service ability promotion excavation demand；
(2.4), feedback subdivision demand is to user；
(2.5), user confirms the corresponding subdivision demand of requirement command.
6. a kind of data digging system for Logistics Park OA operation analysis according to claim 1, it is characterised in that: institute
Stating first-level model is by correlation rule, decision tree, neural network, cluster, support vector machines or collaborative filtering to data
The achievement data stored in warehouse module carries out that treated data set, the second-level model be by correlation rule, decision tree,
Sorting algorithm in neural network, cluster, support vector machines or collaborative filtering is to the index stored in data warehouse module
Data carry out treated data set.
7. a kind of data digging system for Logistics Park OA operation analysis according to claim 1, which is characterized in that institute
It states and the second-level model under first-level model is ranked up the method to form alternative model queue includes the following steps:
(3.1), loglstics enterprise operations risks are obtained and excavate demand, garden value-added service precision marketing excavation demand and garden service
The corresponding achievement data of one type in capability improving excavation demand；
(3.2), the corresponding data of each second-level model are obtained after being handled the achievement data of the acquisition in step (3.1)
Collection；
(3.3), the data set in step (3.2) is divided into training set and test set, second-level model is unfolded on training set
Training, is verified on test set, records recall rate index, the accurate rate index, stability indicator of each second-level model, call together
Rate index, accurate rate index, stability indicator value are returned in [0,1]；
(3.4), recall rate index, accurate rate index, the weight of stability indicator are obtained by analytic hierarchy process (AHP), definition is recalled
Rate, accurate rate, stability weight use α, β and γ to represent respectively, then the score score=α recall+ β of each second-level model
precisioin+γPSI；
(3.4), it is ranked up by the score size of each second-level model, obtains alternative model queue.
8. a kind of data digging system for Logistics Park OA operation analysis according to claim 7, which is characterized in that institute
Specific step is as follows for the analytic hierarchy process (AHP) for stating in step (3.4),
(3.4.1), Analytic Hierarchy Process Model is established；The method for establishing Analytic Hierarchy Process Model includes the following steps:
(3.4.1.1) establishes top, and user demand instructs corresponding subdivision demand for identification；
(3.4.1.2), middle layer is established, for obtaining recall rate index, the accurate rate index, stability indicator of second-level model；
(3.4.1.3), the bottom is established, the concrete scheme for the user demand instruction that achieves a solution；
(3.4.2), construction pairwise comparison matrix；
(3.4.3), recall rate, accurate rate, the weight of stability are calculated and consistency check is done to pairwise comparison matrix.
9. a kind of data digging system for Logistics Park OA operation analysis according to claim 1, it is characterised in that: institute
Stating data memory module is the computer cluster constructed using Hadoop technology.
10. a kind of data digging method for Logistics Park OA operation analysis, which comprises the following steps:
It obtains the loglstics enterprise operations risks stored in data warehouse module and excavates demand, the excavation of garden value-added service precision marketing
Demand and garden service ability promote the corresponding achievement data of excavation demand；
The corresponding data set of each second-level model is obtained after the achievement data of acquisition is handled；
Data set is divided into training set and test set, second-level model is unfolded to train on training set, is carried out on test set
Verifying, records recall rate index, the accurate rate index, stability indicator of each second-level model；
Recall rate index, accurate rate index, the weight of stability indicator are obtained by analytic hierarchy process (AHP)；
It is ranked up by the score size of each second-level model, obtains alternative model queue；
The second-level model of highest scoring in alternative model queue is selected to carry out data mining for initial model；
Obtain the corresponding achievement data of expected operations objective.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910016603.3A CN109583796A (en) | 2019-01-08 | 2019-01-08 | A kind of data digging system and method for Logistics Park OA operation analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910016603.3A CN109583796A (en) | 2019-01-08 | 2019-01-08 | A kind of data digging system and method for Logistics Park OA operation analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109583796A true CN109583796A (en) | 2019-04-05 |
Family
ID=65915948
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910016603.3A Pending CN109583796A (en) | 2019-01-08 | 2019-01-08 | A kind of data digging system and method for Logistics Park OA operation analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109583796A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110489463A (en) * | 2019-08-26 | 2019-11-22 | 广东电网有限责任公司 | A kind of multiple information acquisition system of garden integrated energy system |
| CN111400495A (en) * | 2020-03-17 | 2020-07-10 | 重庆邮电大学 | Video bullet screen consumption intention identification method based on template characteristics |
| CN111639237A (en) * | 2020-04-07 | 2020-09-08 | 安徽理工大学 | Electric power communication network risk assessment system based on clustering and association rule mining |
| CN111724113A (en) * | 2020-06-23 | 2020-09-29 | 中国铁道科学研究院集团有限公司电子计算技术研究所 | Method and equipment for accepting railway freight order |
| CN111858560A (en) * | 2020-07-24 | 2020-10-30 | 厦门至恒融兴信息技术有限公司 | Financial data automated testing and monitoring system based on data warehouse |
| CN112465393A (en) * | 2020-12-09 | 2021-03-09 | 南威软件股份有限公司 | Enterprise risk early warning method based on correlation analysis FP-Tree algorithm |
| CN113379227A (en) * | 2021-06-08 | 2021-09-10 | 软通智慧信息技术有限公司 | Industrial park data processing method and device, computer equipment and storage medium |
| CN113379344A (en) * | 2021-05-27 | 2021-09-10 | 芜湖锋珉信息科技有限公司 | Intelligent logistics processing method and system based on cloud computing processing |
| CN113672589A (en) * | 2021-04-23 | 2021-11-19 | 国网浙江省电力有限公司金华供电公司 | Wisdom logistics storage garden safety perception system |
| CN113742315A (en) * | 2021-08-17 | 2021-12-03 | 广州工业智能研究院 | Manufacturing big data processing platform and method |
| CN115495789A (en) * | 2022-09-27 | 2022-12-20 | 北京瑞莱智慧科技有限公司 | Privacy calculation method, device, equipment and storage medium |
| TWI793432B (en) * | 2020-08-07 | 2023-02-21 | 國立中央大學 | Document management method and system for engineering project |
| CN115862333A (en) * | 2022-12-07 | 2023-03-28 | 东南大学 | Expressway vehicle-road cooperative scene and function division method considering information flow characteristics |
| CN115964035A (en) * | 2022-12-05 | 2023-04-14 | 广州市海捷计算机科技有限公司 | Data mining model management method and device, electronic equipment and storage medium |
| CN115983572A (en) * | 2022-12-22 | 2023-04-18 | 深圳乐信软件技术有限公司 | Method and device for ordering fund parties, computer equipment and storage medium |
| CN116029571A (en) * | 2023-03-29 | 2023-04-28 | 肯特智能技术（深圳）股份有限公司 | Meta universe-based data processing method and related device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065271A (en) * | 2013-01-23 | 2013-04-24 | 广东电网公司东莞供电局 | Power supply reliability and customer satisfaction quantitative relation model establishment method |
| CN105574593A (en) * | 2015-12-18 | 2016-05-11 | 中南大学 | Track state static-state detection and control system and method based on cloud computing and big data |
| CN106580282A (en) * | 2016-10-25 | 2017-04-26 | 上海斐讯数据通信技术有限公司 | Human body health monitoring device, system and method |
| CN106778846A (en) * | 2016-12-02 | 2017-05-31 | 华北电力大学 | A kind of method for forecasting based on SVMs |
-
2019
- 2019-01-08 CN CN201910016603.3A patent/CN109583796A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065271A (en) * | 2013-01-23 | 2013-04-24 | 广东电网公司东莞供电局 | Power supply reliability and customer satisfaction quantitative relation model establishment method |
| CN105574593A (en) * | 2015-12-18 | 2016-05-11 | 中南大学 | Track state static-state detection and control system and method based on cloud computing and big data |
| CN106580282A (en) * | 2016-10-25 | 2017-04-26 | 上海斐讯数据通信技术有限公司 | Human body health monitoring device, system and method |
| CN106778846A (en) * | 2016-12-02 | 2017-05-31 | 华北电力大学 | A kind of method for forecasting based on SVMs |
Non-Patent Citations (2)
| Title |
|---|
| 曾奕棠 等: "基于云挖掘的物流信息智能分析应用平台研究", 《中国流通经济》 * |
| 李波: "《网络购物商品质量管控及其演进研究》", 31 August 2018 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110489463A (en) * | 2019-08-26 | 2019-11-22 | 广东电网有限责任公司 | A kind of multiple information acquisition system of garden integrated energy system |
| CN111400495A (en) * | 2020-03-17 | 2020-07-10 | 重庆邮电大学 | Video bullet screen consumption intention identification method based on template characteristics |
| CN111639237A (en) * | 2020-04-07 | 2020-09-08 | 安徽理工大学 | Electric power communication network risk assessment system based on clustering and association rule mining |
| CN111639237B (en) * | 2020-04-07 | 2023-04-07 | 安徽理工大学 | Electric power communication network risk assessment system based on clustering and association rule mining |
| CN111724113A (en) * | 2020-06-23 | 2020-09-29 | 中国铁道科学研究院集团有限公司电子计算技术研究所 | Method and equipment for accepting railway freight order |
| CN111724113B (en) * | 2020-06-23 | 2024-05-03 | 中国铁道科学研究院集团有限公司电子计算技术研究所 | Method and apparatus for accepting railway freight orders |
| CN111858560A (en) * | 2020-07-24 | 2020-10-30 | 厦门至恒融兴信息技术有限公司 | Financial data automated testing and monitoring system based on data warehouse |
| CN111858560B (en) * | 2020-07-24 | 2023-11-21 | 厦门至恒融兴信息技术有限公司 | Financial data automatic testing and monitoring system based on data warehouse |
| TWI793432B (en) * | 2020-08-07 | 2023-02-21 | 國立中央大學 | Document management method and system for engineering project |
| CN112465393B (en) * | 2020-12-09 | 2022-07-08 | 南威软件股份有限公司 | Enterprise risk early warning method based on correlation analysis FP-Tree algorithm |
| CN112465393A (en) * | 2020-12-09 | 2021-03-09 | 南威软件股份有限公司 | Enterprise risk early warning method based on correlation analysis FP-Tree algorithm |
| CN113672589A (en) * | 2021-04-23 | 2021-11-19 | 国网浙江省电力有限公司金华供电公司 | Wisdom logistics storage garden safety perception system |
| CN113379344A (en) * | 2021-05-27 | 2021-09-10 | 芜湖锋珉信息科技有限公司 | Intelligent logistics processing method and system based on cloud computing processing |
| CN113379227A (en) * | 2021-06-08 | 2021-09-10 | 软通智慧信息技术有限公司 | Industrial park data processing method and device, computer equipment and storage medium |
| CN113742315A (en) * | 2021-08-17 | 2021-12-03 | 广州工业智能研究院 | Manufacturing big data processing platform and method |
| CN115495789B (en) * | 2022-09-27 | 2023-09-29 | 北京瑞莱智慧科技有限公司 | Privacy calculation method, device, equipment and storage medium |
| CN115495789A (en) * | 2022-09-27 | 2022-12-20 | 北京瑞莱智慧科技有限公司 | Privacy calculation method, device, equipment and storage medium |
| CN115964035A (en) * | 2022-12-05 | 2023-04-14 | 广州市海捷计算机科技有限公司 | Data mining model management method and device, electronic equipment and storage medium |
| CN115862333B (en) * | 2022-12-07 | 2023-11-21 | 东南大学 | Expressway vehicle-road cooperative scene and function division method considering information flow characteristics |
| CN115862333A (en) * | 2022-12-07 | 2023-03-28 | 东南大学 | Expressway vehicle-road cooperative scene and function division method considering information flow characteristics |
| CN115983572A (en) * | 2022-12-22 | 2023-04-18 | 深圳乐信软件技术有限公司 | Method and device for ordering fund parties, computer equipment and storage medium |
| CN115983572B (en) * | 2022-12-22 | 2023-12-12 | 深圳乐信软件技术有限公司 | Method and device for ordering fund parties, computer equipment and storage medium |
| CN116029571A (en) * | 2023-03-29 | 2023-04-28 | 肯特智能技术（深圳）股份有限公司 | Meta universe-based data processing method and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583796A (en) | A kind of data digging system and method for Logistics Park OA operation analysis | |
| CN106548322B (en) | Project examination system | |
| Pramod | Robotic process automation for industry: adoption status, benefits, challenges and research agenda | |
| Bose et al. | Big data, data analytics and artificial intelligence in accounting: An overview | |
| CN104361463A (en) | Small and medium-sized enterprise network financing system and method | |
| Lazarević et al. | A model for business performance improvement: A case of the postal company | |
| CN107609771A (en) | A kind of supplier's value assessment method | |
| JP6049799B2 (en) | Management support program | |
| Amin et al. | Application of optimistic and pessimistic OWA and DEA methods in stock selection | |
| CN112116256A (en) | Data asset management method | |
| Qinqin et al. | The effects of enterprises' attention to digital economy on innovation and cost control: Evidence from A-stock market of China | |
| US11995622B2 (en) | Method of international cash management using machine learning | |
| CN115115257A (en) | Enterprise risk early warning method and system based on relational graph | |
| CN110310012A (en) | Data analysing method, device, equipment and computer readable storage medium | |
| CN114493686A (en) | Operation content generation and pushing method and device | |
| CN114021873A (en) | Data index quantification method and intelligent park enterprise value evaluation system | |
| Stefanski | Into the mire: A closer look at fossil fuel subsidies | |
| CN117273511A (en) | Data analysis method and device | |
| CN109785126A (en) | Business loan dynamic air control cost accounting and floating pricing method and system | |
| Schmidt et al. | What is the role of data in jobs in the United Kingdom, Canada, and the United States?: A natural language processing approach | |
| CN109801162A (en) | A kind of credit rating method that social media data are merged with multi-standard cross-certification | |
| CN115564591A (en) | Financing product determination method and related equipment | |
| Dubey et al. | Contextual relationship among antecedents of truck freight using interpretive structural modelling and its validation using MICMAC analysis | |
| CN115423601A (en) | Method and device for designing online credit product | |
| CN113379211A (en) | Block chain-based logistics information platform default risk management and control system and method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190405 |
|
| WD01 | Invention patent application deemed withdrawn after publication |