CN113239091A - Intelligent evaluation system for artificial intelligence B2B website users - Google Patents
Intelligent evaluation system for artificial intelligence B2B website users Download PDFInfo
- Publication number
- CN113239091A CN113239091A CN202110533744.XA CN202110533744A CN113239091A CN 113239091 A CN113239091 A CN 113239091A CN 202110533744 A CN202110533744 A CN 202110533744A CN 113239091 A CN113239091 A CN 113239091A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- index
- module
- determining
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 86
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 25
- 238000007418 data mining Methods 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000010276 construction Methods 0.000 claims abstract description 6
- 230000006399 behavior Effects 0.000 claims description 40
- 230000000875 corresponding effect Effects 0.000 claims description 17
- 238000006243 chemical reaction Methods 0.000 claims description 12
- 230000002787 reinforcement Effects 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000005065 mining Methods 0.000 claims description 8
- 238000000034 method Methods 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 6
- 230000002596 correlated effect Effects 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000001360 synchronised effect Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 230000001276 controlling effect Effects 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000013481 data capture Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims 1
- 230000008859 change Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Images
Classifications
-
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
-
- 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0609—Buyer or seller confidence or verification
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Computational Linguistics (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of information, and discloses an intelligent evaluation system of an artificial intelligence B2B website user, which comprises: the system comprises a user information acquisition module, a data mining module, a central control module, a data processing module, a preference extraction module, an evaluation system construction module, an evaluation index determination module, a weight determination module, an evaluation module and an output module. The invention can carry out objective evaluation on the user without adulterating the subjective consciousness of the user; the invention utilizes the personal information, the transaction data and the browsing data of the user to extract the preference, constructs an index system based on the preference, and utilizes the constructed index system containing a plurality of different indexes to comprehensively, comprehensively and accurately evaluate the user, thereby effectively preventing the online transaction risk.
Description
Technical Field
The invention belongs to the technical field of information, and particularly relates to an intelligent evaluation system for an artificial intelligence B2B website user.
Background
At present: in recent years, with the development of communication technology and the remarkable increase of the number of internet users, various applications based on the internet have come to the fore, and great convenience is brought to the daily life of people. On-line shopping is favored by more and more consumers due to the advantages of cross-regional property, interchangeability, all-weather property, and the like. Compared with the selling mode of a physical store, the selling mode of the commodities on the Internet has certain characteristics and advantages, so that a plurality of commodity suppliers touch the Internet in a dispute, and buyers and sellers carry out online transactions through the B2B website, thereby greatly reducing the circulation links, saving the cost and reducing the purchasing difficulty. However, as the number of registered users on the B2B website increases and the number of online transactions increases, the problems of user integrity and transaction risk are highlighted. In order to better evaluate the transaction possibility of online transaction and prevent the transaction risk as much as possible, scientific and objective evaluation of registered users of the website is a very important and meaningful work.
However, the conventional B2B site has few reference indexes for user evaluation, and the evaluation is inaccurate.
Through the above analysis, the problems and defects of the prior art are as follows: the conventional B2B website has few reference indexes for user evaluation and inaccurate evaluation.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an intelligent evaluation system for users of an artificial intelligence B2B website.
The invention is realized in this way, an intelligent evaluation system of artificial intelligence B2B website users, the intelligent evaluation system of artificial intelligence B2B website users includes:
the data mining module is connected with the central control module and used for acquiring corresponding comment information and browsing data of the user by utilizing a big data mining technology;
the collecting the corresponding comment information and the browsing data of the user by using the big data mining technology comprises the following steps:
determining the synchronization frequency of a local object and a remote data source, wherein the remote data source is a professional database on a remote Web;
representing remote data source average variation frequency lambda by using Poisson processiWherein i is 1,2, …, n, n represents the number of remote data sources;
determining the average novelty: from the resulting mean variation frequency λiDetermining each object, namely each data item e in professional database on remote WebiCorresponding synchronization frequency fiIn the presence ofAverage novelty of local database under synchronous resource limitation condition
At the maximum, the number of the first,
determining the updating frequency according to the data timeliness:
the ith data record r maintained by the data capture system at time tiThe novelty of (c) is as follows:
then the average timeliness of the data record set S consisting of N data records is as follows:
the data record set S is averaged over time and measured:
calculating to obtain theoretical synchronization frequency of each object by using a Lagrange multiplier, and then synchronizing object data according to the theoretical synchronization frequency to enable the average novelty of a local database to reach the maximum value;
the synchronizing the object data according to the theoretical synchronization frequency comprises the following steps:
for all (s, a) initialization table entries Q0(s,a)＝0；
Wherein Q represents professional representation of computer machine learning field, i.e. Q is representation form of reinforcement learning, s represents state, a represents action, Q (s, a) represents result state of applying action a to state s; initialized to a value of 0, i.e. notLearning an initialization value; in each scenario, the range to the data source is taken as its activity, resulting in a reward value of Ri：
And updating the Q value in a time period 0-t:
wherein q isjRepresents the resultant state value, R, of the jth data record obtained by reinforcement learning in the time interval 0-tjRepresenting the return value obtained by reinforcement learning of the jth data record in the time interval 0-t;
under the premise of resource limitation, namely the maximum interaction times M with the server is a constant value, so that the novelty
Maximum value, F (F)i,λi) Representing the novelty of the corresponding ith data record, the novelty being derived from the timeliness of the data, i.e., the timeliness represents the frequency of update of the object, i.e., the smallest unit data item, in the record, and the novelty being the overall timeliness, ω, of the aggregate record of data items, i.e., the remote data sourceiIs the importance weight;
the preference extraction module is connected with the central control module and is used for mining preference characteristics of the user based on the collected personal information of the user, the historical transaction information, the mined browsing data and other data;
the evaluation system building module is connected with the central control module and used for building a user evaluation system based on the personal information, the transaction information, the comment information, the browsing data and the preference characteristics of the user;
the evaluation index determining module is connected with the central control module and used for determining the evaluation index of the user based on the constructed user evaluation system;
the weight determining module is connected with the central control module and is used for determining the corresponding weight of each index in the evaluation system;
and the evaluation module is connected with the central control module and is used for evaluating the user based on the evaluation index and the weight.
Further, the intelligent evaluation system for the artificial intelligence B2B website users further comprises:
the user information acquisition module is connected with the central control module and is used for acquiring personal information and historical transaction information of a user;
the central control module is connected with the user information acquisition module, the data mining module, the data processing module, the preference extraction module, the evaluation system construction module, the evaluation index determination module, the weight determination module, the evaluation module and the output module and is used for controlling each module to normally work by utilizing a single chip microcomputer or a controller;
the data processing module is connected with the central control module and is used for carrying out normalization, duplicate removal and other processing on the mined data;
and the output module is connected with the central control module and used for outputting the result of the user evaluation.
Further, the mining of user preference characteristics by the preference extraction module based on the collected user personal information, historical transaction information, and mined browsing data and other data comprises:
(1) acquiring historical transaction behaviors of a user, and acquiring a behavior sequence based on the historical transaction behaviors;
(2) determining preference characteristics of the user to be recommended according to the behavior sequence of the user to be recommended; wherein the positive conversion behavior event is positively correlated with the preference profile and the negative conversion behavior event is negatively correlated with the preference profile;
(3) determining a first occurrence probability of the preference feature according to the historical behavior data and the condition data; and determining a second probability of occurrence from the condition data;
(4) determining a distance between the first and second occurrence probabilities; and determining the preference value of the user for the preference feature according to the first occurrence probability and the distance.
Further, the behavior sequence is obtained by sequencing historical transaction events of the user for each article according to historical transaction time, and the historical operation events include positive conversion behavior events and negative conversion behavior events.
Further, the determining the preference characteristics of the user to be recommended according to the behavior sequence of the user to be recommended includes:
performing feature extraction on the attribute information of the user by adopting a convolutional neural network to obtain a first feature vector of the user; and performing feature extraction on the behavior sequence by adopting a timing diagram neural network to obtain a second feature vector of the user, and fusing the first feature vector and the second feature vector to obtain the preference feature of the user.
Further, the first occurrence probability is a probability that the preference feature occurs in the historical behavior data under the influence of the condition data, and the second occurrence probability is a probability that the preference feature occurs when the user has no preference.
Further, the determining the corresponding weight of each index in the evaluation system includes:
1) determining a multi-level index structure of an evaluation system, wherein the multi-level index structure comprises a plurality of index layers with parent-child association relations, and each first index in a parent index layer has a parent-child relation with a plurality of second indexes in the child index layer associated with the first index;
2) according to pre-collected index association data generated during user evaluation, determining the hierarchical weight of each first index in a parent index layer and determining the hierarchical weight of a second index in a child index layer having a parent-child relationship with the first index; the level weight is used for representing the importance degree of the index in the index level compared with other indexes associated with the same parent index;
3) and determining the global weight of the second index in the sub-index layer according to the hierarchical weight of the second index in the sub-index layer and the hierarchical weight of the parent index associated with the second index, wherein the global weight is used for representing the importance degree of the index in the index level to which the index belongs compared with other indexes.
By combining all the technical schemes, the invention has the advantages and positive effects that: the invention utilizes the personal information, the transaction data and the browsing data of the user to extract the preference, constructs an index system based on the preference, and utilizes the constructed index system containing a plurality of different indexes to comprehensively, comprehensively and accurately evaluate the user, thereby effectively preventing the online transaction risk.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained from the drawings without creative efforts.
FIG. 1 is a schematic structural diagram of an intelligent evaluation system for users of an artificial intelligence B2B website according to an embodiment of the present invention;
in the figure: 1. a user information acquisition module; 2. a data mining module; 3. a central control module; 4. a data processing module; 5. a preference extraction module; 6. an evaluation system construction module; 7. an evaluation index determination module; 8. a weight determination module; 9. an evaluation module; 10. and an output module.
Fig. 2 is a flowchart of an intelligent evaluation method for website users of artificial intelligence B2B according to an embodiment of the present invention.
FIG. 3 is a flowchart of a method for mining user preference characteristics based on collected user personal information, historical transaction information, and mined browsing data and other data provided by the preference extraction module according to an embodiment of the present invention.
Fig. 4 is a flowchart of a method for determining a preference characteristic of the user to be recommended according to the behavior sequence of the user to be recommended, according to an embodiment of the present invention.
Fig. 5 is a flowchart of a method for determining respective weights of each index in an evaluation system according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In order to solve the problems in the prior art, the present invention provides an intelligent evaluation system for users of an artificial intelligence B2B website, which is described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the intelligent evaluation system for the artificial intelligence B2B website user according to the embodiment of the present invention includes:
the user information acquisition module 1 is connected with the central control module 3 and is used for acquiring personal information and historical transaction information of a user;
the data mining module 2 is connected with the central control module 3 and is used for acquiring corresponding comment information and browsing data of a user by utilizing a big data mining technology;
the central control module 3 is connected with the user information acquisition module 1, the data mining module 2, the data processing module 4, the preference extraction module 5, the evaluation system construction module 6, the evaluation index determination module 7, the weight determination module 8, the evaluation module 9 and the output module 10, and is used for controlling each module to normally work by utilizing a single chip microcomputer or a controller;
the data processing module 4 is connected with the central control module 3 and is used for carrying out normalization, duplicate removal and other processing on the mined data;
the preference extraction module 5 is connected with the central control module 3 and is used for mining preference characteristics of the user based on the collected personal information of the user, the historical transaction information, the mined browsing data and other data;
the evaluation system building module 6 is connected with the central control module 3 and used for building a user evaluation system based on the personal information, the transaction information, the comment information, the browsing data and the preference characteristics of the user;
the evaluation index determining module 7 is connected with the central control module 3 and used for determining the evaluation index of the user based on the constructed user evaluation system;
the weight determining module 8 is connected with the central control module 3 and is used for determining the corresponding weight of each index in the evaluation system;
the evaluation module 9 is connected with the central control module 3 and used for evaluating the user based on the evaluation indexes and the weights;
and the output module 10 is connected with the central control module 3 and used for outputting the result of the user evaluation.
As shown in fig. 2, the intelligent evaluation method for the artificial intelligence B2B website user according to the embodiment of the present invention includes:
s101, acquiring personal information and historical transaction information of a user through a user information acquisition module; collecting corresponding comment information and browsing data of a user by a data mining module through a big data mining technology;
s102, the central control module utilizes a single chip microcomputer or a controller to control a data processing module to carry out normalization, duplicate removal and other processing on the mined data; mining preference characteristics of the user based on the collected personal information, historical transaction information, the mined browsing data and other data of the user through a preference extraction module;
s103, constructing a user evaluation system based on the personal information, the transaction information, the comment information, the browsing data and the preference characteristics of the user through an evaluation system construction module; determining the evaluation index of the user based on the constructed user evaluation system through an evaluation index determination module;
s104, determining the corresponding weight of each index in the evaluation system through a weight determination module; evaluating the user through an evaluation module based on the evaluation index and the weight; and outputting the result of the user evaluation through an output module.
The method for acquiring the corresponding comment information and the browsing data of the user by utilizing the big data mining technology comprises the following steps:
determining the synchronization frequency of a local object and a remote data source, wherein the remote data source is a professional database on a remote Web;
representing the average change frequency lambdai of the remote data source by using a poisson process, wherein i is 1,2, …, n, n represents the number of the remote data sources;
determining the average novelty: determining the synchronous frequency fi corresponding to each object, namely each data item ei in the professional database on the remote Web according to the obtained average change frequency lambada i, and enabling the average novelty of the local database to be in accordance with the synchronous resource limitation condition
At the maximum, the number of the first,
determining the updating frequency according to the data timeliness:
the timeliness of the ith data record ri maintained by the data capture system at time t is as follows:
then the average timeliness of the data record set S consisting of N data records is as follows:
the data record set S is averaged over time and measured:
calculating to obtain theoretical synchronization frequency of each object by using a Lagrange multiplier, and then synchronizing object data according to the theoretical synchronization frequency to enable the average novelty of a local database to reach the maximum value;
the synchronizing the object data according to the theoretical synchronization frequency comprises the following steps:
initializing table entry Q0(s, a) ═ 0 for all (s, a);
wherein Q represents professional representation of computer machine learning field, i.e. Q is representation form of reinforcement learning, s represents state, a represents action, Q (s, a) represents result state of applying action a to state s; initializing to 0 value, namely not learning initialization value; in each scenario, taking the range to the data source as its activity, the reported value is Ri:
and updating the Q value in a time period 0-t:
qj represents a result state value obtained by performing reinforcement learning on the jth data record within a time interval of 0-t, and Rj represents a return value obtained by performing reinforcement learning on the jth data record within a time interval of 0-t;
under the premise of resource limitation, namely the maximum interaction times M with the server is a constant value, so that the novelty
The value is maximum, F (fi, λ i) represents the novelty corresponding to the ith data record, the novelty is obtained by the data timeliness, that is, the timeliness represents the update frequency of the object in the record, that is, the minimum unit data item, and the novelty refers to the overall timeliness of the collective record of the data items, that is, the remote data source, and ω i is the importance weight.
As shown in fig. 3, the preference extracting module provided in the embodiment of the present invention, based on the collected personal information of the user, the historical transaction information, the mined browsing data, and other data, mining the preference characteristics of the user, includes:
s201, acquiring historical transaction behaviors of a user, and obtaining a behavior sequence based on the historical transaction behaviors;
s202, determining preference characteristics of the user to be recommended according to the behavior sequence of the user to be recommended; wherein the positive conversion behavior event is positively correlated with the preference profile and the negative conversion behavior event is negatively correlated with the preference profile;
s203, determining a first occurrence probability of the preference feature according to the historical behavior data and the condition data; and determining a second probability of occurrence from the condition data;
s204, determining the distance between the first occurrence probability and the second occurrence probability; and determining the preference value of the user for the preference feature according to the first occurrence probability and the distance.
The behavior sequence provided by the embodiment of the invention is obtained by sequencing the historical transaction events of the user aiming at each article according to the historical transaction time, wherein the historical operation events comprise positive conversion behavior events and negative conversion behavior events.
As shown in fig. 4, the determining, according to the behavior sequence of the user to be recommended, the preference characteristic of the user to be recommended includes:
s301, extracting features of the attribute information of the user by adopting a convolutional neural network to obtain a first feature vector of the user;
s302, extracting features of the behavior sequence by adopting a timing diagram neural network to obtain a second feature vector of the user;
s303, fusing the first feature vector and the second feature vector to obtain the preference feature of the user.
The first occurrence probability provided by the embodiment of the invention is the probability of the preference feature occurring in the historical behavior data under the influence of the condition data, and the second occurrence probability is the probability of the preference feature occurring when the user has no preference.
As shown in fig. 5, determining the corresponding weight of each index in the evaluation system provided in the embodiment of the present invention includes:
s401, determining a multi-level index structure of an evaluation system, wherein the multi-level index structure comprises a plurality of index layers with parent-child association relationship, and each first index in a parent index layer and a plurality of second indexes in the child index layer associated with the first index have parent-child relationship;
s402, according to pre-collected index association data generated during user evaluation, determining the hierarchical weight of each first index of a father index layer and determining the hierarchical weight of a second index in a child index layer having a parent-child relationship with the first index; the level weight is used for representing the importance degree of the index in the index level compared with other indexes associated with the same parent index;
and S403, determining the global weight of the second index in the sub-index layer according to the hierarchical weight of the second index in the sub-index layer and the hierarchical weight of the parent index associated with the second index, wherein the global weight is used for representing the importance degree of the index in the index level to which the index belongs compared with other indexes.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, and any modification, equivalent replacement, and improvement made by those skilled in the art within the technical scope of the present invention disclosed herein, which is within the spirit and principle of the present invention, should be covered by the present invention.
Claims (10)
1. An intelligent evaluation system of artificial intelligence B2B website users, characterized in that the intelligent evaluation system of artificial intelligence B2B website users comprises:
the data mining module is connected with the central control module and used for acquiring corresponding comment information and browsing data of the user by utilizing a big data mining technology;
the collecting the corresponding comment information and the browsing data of the user by using the big data mining technology comprises the following steps:
determining the synchronization frequency of a local object and a remote data source, wherein the remote data source is a professional database on a remote Web;
representing remote data source average variation frequency lambda by using Poisson processiWherein i is 1,2, …, n, n represents the number of remote data sources;
determining the average novelty: from the resulting mean variation frequency λiDetermining each object, namely each data item e in professional database on remote WebiCorresponding synchronization frequency fiMaking the average novelty of the local database meet the synchronous resource limitation
At the maximum, the number of the first,
determining the updating frequency according to the data timeliness:
the ith data record r maintained by the data capture system at time tiThe novelty of (c) is as follows:
then the average timeliness of the data record set S consisting of N data records is as follows:
the data record set S is averaged over time and measured:
calculating to obtain theoretical synchronization frequency of each object by using a Lagrange multiplier, and then synchronizing object data according to the theoretical synchronization frequency to enable the average novelty of a local database to reach the maximum value;
the synchronizing the object data according to the theoretical synchronization frequency comprises the following steps:
for all (s, a) initialization table entries Q0(s,a)＝0；
Wherein Q represents professional representation of computer machine learning field, i.e. Q is representation form of reinforcement learning, s represents state, a represents action, Q (s, a) represents result state of applying action a to state s; initializing to 0 value, namely not learning initialization value; in each scenario, the range to the data source is taken as its activity, resulting in a reward value of Ri：
And updating the Q value in a time period 0-t:
wherein q isjRepresents the resultant state value, R, of the jth data record obtained by reinforcement learning in the time interval 0-tjRepresenting the return value obtained by reinforcement learning of the jth data record in the time interval 0-t;
under the premise of resource limitation, namely the maximum interaction times M with the server is a constant value, so that the novelty
Maximum value, F (F)i,λi) Representing the novelty of the corresponding ith data record, the novelty being derived from the timeliness of the data, i.e., the timeliness represents the frequency of update of the object, i.e., the smallest unit data item, in the record, and the novelty being the overall timeliness, ω, of the aggregate record of data items, i.e., the remote data sourceiIs the importance weight;
the preference extraction module is connected with the central control module and is used for mining preference characteristics of the user based on the collected personal information of the user, the historical transaction information, the mined browsing data and other data;
the evaluation system building module is connected with the central control module and used for building a user evaluation system based on the personal information, the transaction information, the comment information, the browsing data and the preference characteristics of the user;
the evaluation index determining module is connected with the central control module and used for determining the evaluation index of the user based on the constructed user evaluation system;
the weight determining module is connected with the central control module and is used for determining the corresponding weight of each index in the evaluation system;
and the evaluation module is connected with the central control module and is used for evaluating the user based on the evaluation index and the weight.
2. The intelligent ratings system of artificial intelligence B2B website users of claim 1, wherein the intelligent ratings system of artificial intelligence B2B website users further comprises:
the user information acquisition module is connected with the central control module and is used for acquiring personal information and historical transaction information of a user;
the central control module is connected with the user information acquisition module, the data mining module, the data processing module, the preference extraction module, the evaluation system construction module, the evaluation index determination module, the weight determination module, the evaluation module and the output module and is used for controlling each module to normally work by utilizing a single chip microcomputer or a controller;
the data processing module is connected with the central control module and is used for carrying out normalization, duplicate removal and other processing on the mined data;
and the output module is connected with the central control module and used for outputting the result of the user evaluation.
3. The intelligent rating system of artificial intelligence B2B website users as claimed in claim 1, wherein said preference extraction module mining user's preference characteristics based on collected user personal information, historical transaction information, and mined browsing data and other data comprises:
(1) acquiring historical transaction behaviors of a user, and acquiring a behavior sequence based on the historical transaction behaviors;
(2) determining preference characteristics of the user to be recommended according to the behavior sequence of the user to be recommended; wherein the positive conversion behavior event is positively correlated with the preference profile and the negative conversion behavior event is negatively correlated with the preference profile;
(3) determining a first occurrence probability of the preference feature according to the historical behavior data and the condition data; and determining a second probability of occurrence from the condition data;
(4) determining a distance between the first and second occurrence probabilities; and determining the preference value of the user for the preference feature according to the first occurrence probability and the distance.
4. The intelligent evaluation system of the artificial intelligence B2B website user of claim 3, wherein the behavior sequence is obtained by sorting historical transaction events of the user for each item according to historical transaction time, and the historical operation events comprise positive conversion behavior events and negative conversion behavior events.
5. The intelligent evaluation system of the artificial intelligence B2B website user of claim 3, wherein the determining the preference characteristics of the user to be recommended according to the behavior sequence of the user to be recommended comprises:
performing feature extraction on the attribute information of the user by adopting a convolutional neural network to obtain a first feature vector of the user; and performing feature extraction on the behavior sequence by adopting a timing diagram neural network to obtain a second feature vector of the user, and fusing the first feature vector and the second feature vector to obtain the preference feature of the user.
6. The intelligent rating system for users of an artificial intelligence B2B website of claim 3, wherein the first probability of occurrence is a probability that the preferred feature occurs in the historical behavior data under the influence of the condition data, and the second probability of occurrence is a probability that the preferred feature occurs when the user has no preference.
7. The intelligent evaluation system of artificial intelligence B2B website users as claimed in claim 1, wherein said determining respective weights for each of the metrics in the evaluation system comprises:
1) determining a multi-level index structure of an evaluation system, wherein the multi-level index structure comprises a plurality of index layers with parent-child association relations, and each first index in a parent index layer has a parent-child relation with a plurality of second indexes in the child index layer associated with the first index;
2) according to pre-collected index association data generated during user evaluation, determining the hierarchical weight of each first index in a parent index layer and determining the hierarchical weight of a second index in a child index layer having a parent-child relationship with the first index; the level weight is used for representing the importance degree of the index in the index level compared with other indexes associated with the same parent index;
3) and determining the global weight of the second index in the sub-index layer according to the hierarchical weight of the second index in the sub-index layer and the hierarchical weight of the parent index associated with the second index, wherein the global weight is used for representing the importance degree of the index in the index level to which the index belongs compared with other indexes.
8. An information data processing terminal, characterized in that the information data processing terminal is used for implementing an intelligent evaluation system of the artificial intelligence B2B website user according to any one of claims 1-7.
9. A computer program product stored on a computer readable medium, comprising a computer readable program for providing a user input interface for applying the intelligent rating system of the artificial intelligence B2B website user as claimed in any one of claims 1-7 when executed on an electronic device.
10. A computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to apply the intelligent rating system of the artificial intelligence B2B website user as recited in any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110533744.XA CN113239091A (en) | 2021-05-14 | 2021-05-14 | Intelligent evaluation system for artificial intelligence B2B website users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110533744.XA CN113239091A (en) | 2021-05-14 | 2021-05-14 | Intelligent evaluation system for artificial intelligence B2B website users |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113239091A true CN113239091A (en) | 2021-08-10 |
Family
ID=77134591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110533744.XA Pending CN113239091A (en) | 2021-05-14 | 2021-05-14 | Intelligent evaluation system for artificial intelligence B2B website users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113239091A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117370424A (en) * | 2023-12-07 | 2024-01-09 | 深圳市易图资讯股份有限公司 | Mobile application comment data analysis mining method and system for economic information analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103636A (en) * | 2011-01-18 | 2011-06-22 | 南京信息工程大学 | Deep web-oriented incremental information acquisition method |
| CN104636447A (en) * | 2015-01-21 | 2015-05-20 | 上海天呈医流科技股份有限公司 | Intelligent evaluation method and system for medical instrument B2B website users |
| CN107451140A (en) * | 2016-05-30 | 2017-12-08 | 阿里巴巴集团控股有限公司 | A kind of determination method and device of user preference degree |
| CN111784455A (en) * | 2020-06-30 | 2020-10-16 | 腾讯科技（深圳）有限公司 | Article recommendation method and recommendation equipment |
| CN112182317A (en) * | 2020-09-28 | 2021-01-05 | 北京字节跳动网络技术有限公司 | Index weight determination method and device, electronic equipment and medium |
-
2021
- 2021-05-14 CN CN202110533744.XA patent/CN113239091A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103636A (en) * | 2011-01-18 | 2011-06-22 | 南京信息工程大学 | Deep web-oriented incremental information acquisition method |
| CN104636447A (en) * | 2015-01-21 | 2015-05-20 | 上海天呈医流科技股份有限公司 | Intelligent evaluation method and system for medical instrument B2B website users |
| CN107451140A (en) * | 2016-05-30 | 2017-12-08 | 阿里巴巴集团控股有限公司 | A kind of determination method and device of user preference degree |
| CN111784455A (en) * | 2020-06-30 | 2020-10-16 | 腾讯科技（深圳）有限公司 | Article recommendation method and recommendation equipment |
| CN112182317A (en) * | 2020-09-28 | 2021-01-05 | 北京字节跳动网络技术有限公司 | Index weight determination method and device, electronic equipment and medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117370424A (en) * | 2023-12-07 | 2024-01-09 | 深圳市易图资讯股份有限公司 | Mobile application comment data analysis mining method and system for economic information analysis |
| CN117370424B (en) * | 2023-12-07 | 2024-02-13 | 深圳市易图资讯股份有限公司 | Mobile application comment data analysis mining method and system for economic information analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Hesitant fuzzy ELECTRE II approach: a new way to handle multi-criteria decision making problems | |
| CN109684538A (en) | A kind of recommended method and recommender system based on individual subscriber feature | |
| CN103353880B (en) | A kind of utilization distinctiveness ratio cluster and the data digging method for associating | |
| CN103049865A (en) | Method and system for initiatively recommending product information service | |
| Wang et al. | Integrating rough set theory with customer satisfaction to construct a novel approach for mining product design rules | |
| CN111723292A (en) | Recommendation method and system based on graph neural network, electronic device and storage medium | |
| CN112541806B (en) | Recommendation method and device based on heterogeneous information network | |
| CN113744032A (en) | Book recommendation method, related device, equipment and storage medium | |
| CN108122153A (en) | Personalized recommendation method based on cloud computing tupe under e-commerce environment | |
| CN112818230A (en) | Content recommendation method and device, electronic equipment and storage medium | |
| CN115344767A (en) | Supplier evaluation method based on network data | |
| Fan et al. | Optimal selection of design scheme in cloud environment: A novel hybrid approach of multi-criteria decision-making based on F-ANP and F-QFD | |
| CN115080868A (en) | Product pushing method, product pushing device, computer equipment, storage medium and program product | |
| CN113239091A (en) | Intelligent evaluation system for artificial intelligence B2B website users | |
| CN114065063A (en) | Information processing method, information processing apparatus, storage medium, and electronic device | |
| CN117495482A (en) | Secondhand mobile phone sales recommendation method and system based on user portrait | |
| CN115827994A (en) | Data processing method, device, equipment and storage medium | |
| CN115619503A (en) | Article recommendation method and device, storage medium and computer equipment | |
| Boulkrinat et al. | Towards recommender systems based on a fuzzy preference aggregation | |
| CN112801384A (en) | Commodity quality evaluation and prediction method, system, medium and equipment | |
| CN111460300A (en) | Network content pushing method and device and storage medium | |
| Zhu et al. | Research on system of data mining technology based on computer | |
| Alghamedy et al. | Enhance NMF-based recommendation systems with social information imputation | |
| Yang et al. | Consumers’ Purchase Behavior Preference in E-Commerce Platform Based on Data Mining Algorithm | |
| CN116187632B (en) | Behavior prediction model training method, house source matching method, device and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210810 |
|
| RJ01 | Rejection of invention patent application after publication |