CN115599996A - Knowledge graph-based factory sample recommendation system - Google Patents
Knowledge graph-based factory sample recommendation system Download PDFInfo
- Publication number
- CN115599996A CN115599996A CN202211251283.8A CN202211251283A CN115599996A CN 115599996 A CN115599996 A CN 115599996A CN 202211251283 A CN202211251283 A CN 202211251283A CN 115599996 A CN115599996 A CN 115599996A
- Authority
- CN
- China
- Prior art keywords
- sample
- entity
- client
- customer
- entities
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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]
- G06Q30/0631—Item recommendations
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a knowledge graph-based factory sample recommendation system, which relates to the technical field of data mining, and comprises: the knowledge graph building module is used for building a knowledge graph according to the client entity, the sample entity, the supply chain entity, the equipment entity, the production entity and the incidence relation among the entities; the intelligent sample recommending module is used for completing sample recommendation after the intelligent sample recommending module is used according to the relationship among the client entity, the sample entity, the supply chain entity, the equipment entity and the production entity; the data transmission module is used for completing the transmission of all data in the factory sample recommendation system; and the system portal module is used for establishing an interface for interaction between the client and the recommendation system. According to the method, the client entity, the sample entity, the supply chain entity, the production entity, the equipment entity and the incidence relation among the entities in the knowledge graph are fused in the sample recommendation process, so that the practicability of data is greatly improved, and the problem of data sparsity is effectively relieved.
Description
Technical Field
The invention relates to the technical field of data mining, in particular to a factory sample recommendation system based on a knowledge graph.
Background
With the development of industrial internet and internet of things, the data scale of the factory shows explosive growth, and particularly with the rise of small-batch and customized production modes, the factory generates a large amount of sample data. The big data of the sample contains huge value, but the problem of information overload is also caused. In order to solve the problem, the recommendation system is introduced into the field of intelligent manufacturing as an information filtering technology widely applied to the internet, and a factory can provide good decision support and personalized service for a client.
Most of recommendation systems in the internet field recommend samples for clients by using expert rules or collaborative filtering algorithms, but expert rules often have incomplete knowledge about clients of different types of factories and cannot fully consider the relationship between the clients and the recommended samples, and although collaborative filtering algorithms can utilize historical data of the clients to understand the intentions of the clients and recommend suitable samples, incomplete or inaccurate content analysis is avoided to a certain extent, the algorithms rely on a large amount of data and can encounter the problems that new samples or new clients need cold start and the like, and neither client-based collaborative filtering nor sample content-based collaborative filtering is faced with the problems that recommended samples are too similar and the diversity of the recommended samples is insufficient.
In an actual industrial application scene, interaction information of a client and a sample only exists if the client is an old client of a factory, and for a huge sample library, the sample types and data are various, and often the client only has interaction information on a very small number of samples, so that the data are very sparse. When a large amount of unknown information is predicted by using such a small amount of observed data, the accuracy rate and the verification accuracy rate have great difference, that is, the risk of model overfitting is increased, and the operation effect of the system is often poor.
If a new client enters the recommendation system, the system only contains basic attribute information of the client, and the behavior data of the client is lacked, so that the future preference of the client cannot be predicted according to the historical behavior of the client. On the other hand, when a new sample enters the sample library, the behavior data of the sample is lost, and the system cannot recommend the sample to a client who may be interested in the sample, so that the problems of client cold start and sample cold start are caused.
When the above problems exist, a factory needs to make repeated proofs to meet the customer requirements, which leads to an increase in the cost of the factory and a decrease in the customer satisfaction, and therefore, there is a need to provide a system for recommending factory samples based on a knowledge graph, so as to solve the problem in the prior art that recommended samples are too single and inaccurate due to data sparseness and cold start, and provide samples with diversity and relevance for customers.
Disclosure of Invention
The invention aims to provide a plant sample recommendation system based on a knowledge graph, which aims to solve the problems that in the prior art, recommended samples are too single and inaccurate due to data sparseness and cold start.
In order to achieve the purpose, the invention provides the following technical scheme:
a knowledge-graph-based plant sample recommendation system, comprising: the knowledge graph building module is used for building a knowledge graph according to the client entity, the sample entity, the supply chain entity, the equipment entity, the production entity and the incidence relation among the entities; the intelligent sample recommending module is used for completing sample recommendation after the intelligent sample recommending module is used according to the relationship among the client entity, the sample entity, the supply chain entity, the equipment entity and the production entity; the data transmission module is used for completing the transmission of all data in the factory sample recommendation system; and the system portal module is used for establishing an interface for interaction between the client and the recommendation system.
On the basis of the technical scheme, the invention also provides the following optional technical scheme:
in one alternative: the incidence relation among the customer entity, the sample entity, the supply chain entity, the production entity and the equipment entity comprises the following steps: an association between a customer entity and a sample entity, an association between a customer entity and a supply chain entity, an association between a sample entity and a production entity, an association between a sample entity and an equipment entity, an association between an equipment entity and a production entity, an association between different customer entities, and an association between different sample entities.
In one alternative: the client entity comprises basic information of the client and client tag data, wherein the client tag data comprises client states and client preferences, and the client states are divided into a business client, a scheme client, a contract client and an attrition client; the client state is divided according to the operation behavior of the client and the transaction record condition; the client preference is obtained from the historical browsing record, the historical price inquiry record, the historical proofing record and the historical transaction record of the client.
In one alternative: the sample entity comprises basic information of the sample and sample label data, wherein the basic information comprises a sample number, a sample name, a sample specification model, a sample attribute and a sample characteristic; the sample label data comprises sample classification, sample type and sample statistical information; wherein the sample type distinguishes between new and old samples, and the sample statistical information includes viewing times, transaction times, collection times, and evaluation scores.
In one alternative: when the sample is recommended, if a client enters the system, the system firstly judges whether the client is an old client or not according to the client portrait, and if the client is an old client, the sample is recommended according to the client entity relationship; if the new customer is the new customer, the relationship data among the customer entities is introduced, the old customer similar to the new customer can be obtained, and a sample more conforming to the preference of the new customer can be obtained according to the basic information of the new customer and the historical information of the old customer similar to the new customer in a correlation analysis mode.
In one alternative: when the sample is recommended, if a sample enters the system, the system firstly judges whether the sample is a new sample, if the sample is the new sample, the new sample is associated with the old sample according to the relation data among the sample entities, and association analysis is carried out according to the attribute data of the new sample and the similar old sample entity relation data.
In one alternative: the supply chain entity comprises supplier information, raw and auxiliary material names, raw and auxiliary material prices, customer information and transaction information; the supply chain entity associates the information of the upstream and downstream of the factory and the raw materials of the sample, and records the transaction behaviors of the sample and the raw materials.
In one alternative: the production entity comprises production information, production process, production procedures, production quality inspection and production batches of the sample; the production process comprises raw material information, standard working hours, stations, personnel, machines and operation guidance.
In one alternative: the device entities include device number/tag number, device name, model, specification, manufacturer, factory code/serial number, manufacturing date, incoming date, acceptance date, export date, and attribute type.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the method, the client entity, the sample entity, the supply chain entity, the production entity, the equipment entity and the incidence relation among the entities in the knowledge graph are fused in the sample recommendation process, so that the practicability of data is greatly improved, and the problem of data sparsity is effectively relieved;
2. according to the invention, the problems of client cold start and sample cold start are effectively solved by establishing effective association between a new client and a similar old client and between a new sample and a similar old sample;
3. by using the knowledge graph technology, the invention fully excavates the relationship between entities such as supply chains, production, equipment and the like and the entities such as the samples and customer entities by utilizing the attributes of the samples and the behavior information of customers on the samples, realizes the accurate recommendation and diversified recommendation of the samples, ensures that the recommended samples can accurately accord with the interest preference of the customers, and simultaneously can arouse the attention of the customers by recommending some new sample combinations, arouses the interest of the customers and increases the viscosity of the customers;
4. according to the invention, the interpretability of the recommendation result is enhanced by associating the historical data and the recommendation result of the client and the data of a supply chain, a production process and the like;
5. according to the invention, a client can remotely, simply and conveniently acquire the information of the interested sample, so that the quality and the satisfaction degree of user interaction are greatly improved, and the use experience and the communication efficiency of the client are improved.
Drawings
FIG. 1 is a block diagram of a plant sample recommendation system.
FIG. 2 is a flow chart of plant sample knowledge map creation.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. The examples are given solely for the purpose of illustration and are not intended to limit the scope of the invention. Any obvious modifications or variations can be made without departing from the spirit or scope of the present invention.
In one embodiment, a knowledge-graph-based plant sample recommendation system mainly comprises four modules:
the knowledge graph building module is mainly used for building a knowledge graph according to a client entity, a sample entity, a supply chain entity, an equipment entity, a production entity and the incidence relation among the entities;
the intelligent sample recommending module is mainly used for completing sample recommendation after the intelligent sample recommending module is used according to the relationship among the client entity, the sample entity, the supply chain entity, the equipment entity and the production entity;
the data transmission module is mainly used for completing the transmission of all data in the factory sample recommendation system;
and the system portal module mainly functions to provide an interface for interaction between the client and the recommendation system.
According to the method, the practicability of the data is greatly improved and the problem of data sparsity is effectively relieved through the client entity, the sample entity, the supply chain entity, the production entity, the equipment entity and the incidence relation among the entities in the knowledge graph; the problems of client cold start and sample cold start are effectively solved; by using a knowledge graph technology, the accurate and diversified recommendation of the sample is realized by fully utilizing the associated information among the entities, so that the sample recommended by the invention can accurately accord with the design preference of a customer, and meanwhile, the attention of the customer can be brought by recommending some new sample combinations, the interest of the customer is stimulated, and the viscosity of the customer is increased; the interpretability of the recommendation result is enhanced by associating the historical data and the recommendation result of the client and the data of a supply chain, a production process and the like; in addition, the client can remotely, simply and conveniently acquire the information of the interested sample, and the use experience of the client is greatly improved.
The process of establishing the knowledge graph is shown in fig. 2, and the dynamic relationship characteristics among the entities of the sample, equipment, production and supply chain are combined with the basic information of the customer and the structured data information of browsing, order recording and the like of the customer history, so that the intention of the customer is analyzed, and the potential requirements and the intention of the customer are mined.
The established knowledge graph can be used for excavating rich relations such as association relations among customers, sample preferences of customers, similarity among samples, production processes, supply chains and the like, and the relation data can effectively solve the problem of cold start of the customers and the samples.
When accurate recommendation is carried out, if a client enters the system, the system judges whether the client is an old client or not according to a client figure, if the client is not an old client, characteristic data such as historical operation behaviors of the client and the like are seriously lost, only basic information of the client can be relied on during model analysis, a recommendation result is often seriously distorted, a knowledge graph is introduced into relationship data among client entities, the old client similar to a new client can be obtained, and a sample more conforming to the preference of the client is obtained according to the basic information of the new client and the historical information of the old client similar to the new client through correlation analysis.
Similarly, if a sample enters the system, the system firstly judges whether the sample is a new sample, if the sample is the new sample, only the attribute data of the sample exists, and the data of the sample, such as browsing, production, supply chain and the like, are completely absent. And the relation data among the sample entities in the knowledge graph can also relate the new sample with the old sample, and carry out correlation analysis according to the attribute data of the new sample and the data of the customer transaction behavior and the like of the old sample similar to the attribute data, so that the exposure rate of the new sample is improved, and the recommendation effectiveness is greatly increased.
In conclusion, the association relationship among the entities is obtained according to the client entity, the sample entity, the supply chain entity, the production entity and the equipment entity when the knowledge graph is constructed;
the association relationship among the entities comprises: an association between a customer entity and a sample entity, an association between a customer entity and a supply chain entity, an association between a sample entity and a production entity, an association between a sample entity and an equipment entity, an association between an equipment entity and a production entity, an association between different customer entities, and an association between different sample entities.
The contents of the client entity, the sample entity, the supply chain entity, the production entity and the equipment entity are specifically as follows:
the client entity comprises basic information of the client and client tag data, wherein the client tag data comprises client state, client preference and the like; the client state is divided into a business client, a scheme client, a contract client and an attrition client, and the client state is divided according to the operating behavior of the client, transaction records and other conditions; the client preference is obtained from the client's historical browsing records, historical price enquiries, historical proofing records, historical transaction records, etc. The invention can also establish the client label data according to the supply chain preference, the production process and the like of the client so as to enrich the portrait of the client and perfect the relationship between client entities.
The sample entity comprises basic information of the sample and sample label data, wherein the basic information comprises a sample number, a sample name, a sample specification model, a sample attribute, a sample characteristic and the like. The sample label data comprises sample classification, sample type, sample statistical information and the like; the sample type distinguishes new samples from old samples, and the sample statistical information comprises viewing times, transaction times, collection times, evaluation scores and the like.
The supply chain entity comprises supplier information, raw and auxiliary material names, raw and auxiliary material prices, customer information, transaction information and the like. The supply chain entity associates the information of the raw materials and auxiliary materials of the samples and the upstream and downstream of the factory, and records the transaction behaviors of the samples and the raw materials, including order numbers, customer names, sample numbers, sample names, sample numbers and the like.
The production entity comprises production information, production process, production procedures, production quality inspection, production batch and the like of the sample. The production process is an important factor for determining the characteristics and the attributes of the sample and comprises different production process combinations. The production process comprises raw material information, standard working hours, stations, personnel, machines, operation guidance and the like. The production batch is used as a main basis for tracking, and the process, the working procedure, the raw material batch, the machine production time, the raw material batch and the like are related. Production is used as a core department of a factory, and production entities mainly establish an association relationship between the production entities and client entities.
The equipment entity comprises equipment number/label number, equipment name, model, specification, manufacturer, factory code/serial number, manufacture date, admission date, acceptance date, guarantee date, attribute type and the like. The device is used as an important ring in production, and mainly establishes the association relationship between a production entity and a sample entity.
The above description is only for the specific embodiments of the present disclosure, but the scope of the present disclosure is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present disclosure, and all the changes or substitutions should be covered within the scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (9)
1. A knowledge-graph-based plant sample recommendation system, comprising: the knowledge graph building module is used for building a knowledge graph according to the client entity, the sample entity, the supply chain entity, the equipment entity, the production entity and the incidence relation among the entities; the intelligent sample recommending module is used for completing sample recommendation after the intelligent sample recommending module is used according to the relationship among the client entity, the sample entity, the supply chain entity, the equipment entity and the production entity; the data transmission module is used for completing the transmission of all data in the factory sample recommendation system; and the system portal module is used for establishing an interface for interaction between the client and the recommendation system.
2. The knowledge-graph-based plant sample recommendation system of claim 1, wherein the associations between customer entities, sample entities, supply chain entities, production entities, equipment entities comprise: an association between a customer entity and a sample entity, an association between a customer entity and a supply chain entity, an association between a sample entity and a production entity, an association between a sample entity and an equipment entity, an association between an equipment entity and a production entity, an association between different customer entities, and an association between different sample entities.
3. The knowledge-graph-based plant sample recommendation system of claim 1, wherein the customer entities comprise basic information of customers and customer tag data, wherein the customer tag data comprises customer status, customer preferences, the customer status being classified as a merchant customer, a project customer, a contract customer, and an attrition customer; the client state is divided according to the operation behavior of the client and the transaction record condition; the client preference is obtained from the historical browsing record, the historical price inquiry record, the historical proofing record and the historical transaction record of the client.
4. The knowledgegraph-based factory sample recommendation system of claim 1, wherein a sample entity comprises basic information and sample tag data for a sample, the basic information comprising a sample number, a sample name, a sample specification model, a sample attribute, and a sample characteristic; the sample label data comprises sample classification, sample type and sample statistical information; the sample type distinguishes between new samples and old samples, and the sample statistical information comprises viewing times, transaction times, collection times and evaluation scores.
5. The system of claim 4, wherein in case of sample recommendation, if a client enters the system, the system first determines whether the client is a senior client based on the client image, and if the client is a senior client, recommends a sample based on the client entity relationship; if the new customer is the new customer, the relationship data among the customer entities is introduced, the old customer similar to the new customer can be obtained, and a sample more conforming to the preference of the new customer can be obtained according to the basic information of the new customer and the historical information of the old customer similar to the new customer in a correlation analysis mode.
6. The system of claim 4, wherein when a sample is recommended, if a sample enters the system, the system first determines whether the new sample is available, if the new sample is available, associates the new sample with the old sample according to the relationship data between the sample entities, and performs association analysis according to the attribute data of the new sample and the relationship data of the old sample entities similar to the new sample.
7. The knowledge-graph-based plant sample recommendation system of claim 1, wherein supply chain entities comprise supplier information, raw material names, raw material prices, customer information, and transaction information; and the supply chain entity associates the information of the upstream and downstream of the factory and the raw materials of the sample and records the transaction behaviors of the sample and the raw materials.
8. The knowledge-graph-based plant sample recommendation system of claim 1, wherein the production entities comprise production information, production processes, production procedures, production quality checks, and production lots of the samples; the production process comprises raw material information, standard working hours, stations, personnel, machines and operation guidance.
9. The knowledgegraph-based factory sample recommendation system of claim 1, wherein a device entity includes a device number/tag number, a device name, a model, a specification, a manufacturer, a factory code/serial number, a manufacturing date, an admission date, an acceptance date, an export date, and an attribute type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211251283.8A CN115599996A (en) | 2022-10-13 | 2022-10-13 | Knowledge graph-based factory sample recommendation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211251283.8A CN115599996A (en) | 2022-10-13 | 2022-10-13 | Knowledge graph-based factory sample recommendation system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115599996A true CN115599996A (en) | 2023-01-13 |
Family
ID=84847546
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211251283.8A Pending CN115599996A (en) | 2022-10-13 | 2022-10-13 | Knowledge graph-based factory sample recommendation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115599996A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116861077A (en) * | 2023-06-25 | 2023-10-10 | 北京信大融金教育科技有限公司 | Product recommendation method, device, equipment and storage medium based on supply chain system |
-
2022
- 2022-10-13 CN CN202211251283.8A patent/CN115599996A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116861077A (en) * | 2023-06-25 | 2023-10-10 | 北京信大融金教育科技有限公司 | Product recommendation method, device, equipment and storage medium based on supply chain system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wong et al. | Customer online shopping experience data analytics: Integrated customer segmentation and customised services prediction model | |
| Ahlemeyer-Stubbe et al. | A practical guide to data mining for business and industry | |
| US11783432B1 (en) | System for providing access to user interfaces generated using purchase history data | |
| US8571919B2 (en) | System and method for identifying attributes of a population using spend level data | |
| US20110178845A1 (en) | System and method for matching merchants to a population of consumers | |
| US20110178848A1 (en) | System and method for matching consumers based on spend behavior | |
| US20110178847A1 (en) | System and method for identifying a selected demographic's preferences using spend level data | |
| CN112269805B (en) | Data processing method, device, equipment and medium | |
| CN109064265A (en) | Purchase vehicle recommended method and system based on the network platform | |
| US11087392B2 (en) | Systems and methods for analysis of wearable items of a clothing subscription platform | |
| CN112231533A (en) | Data processing method, device, equipment and storage medium | |
| Chang et al. | Personalized travel recommendation: a hybrid method with collaborative filtering and social network analysis | |
| CN116862592B (en) | Automatic push method for SOP private marketing information based on user behavior | |
| CN115330531B (en) | Enterprise risk prediction method based on electricity consumption fluctuation period | |
| Smith et al. | Insight from data analytics with an automotive aftermarket SME | |
| CN115599996A (en) | Knowledge graph-based factory sample recommendation system | |
| US20110178843A1 (en) | System and method for using spend behavior to identify a population of consumers that meet a specified criteria | |
| Rashi et al. | An AI-Based Customer Relationship Management Framework for Business Applications | |
| WO2020008433A2 (en) | Availability ranking system and method | |
| Gong | Employing User Attribute and Item Attribute to Enhance the Collaborative Filtering Recommendation. | |
| CN111833073A (en) | Airline customer segmentation method based on K-Means + + algorithm | |
| CN110347923B (en) | Traceable fast fission type user portrait construction method | |
| CN110738538B (en) | Method and device for identifying similar objects | |
| Liu et al. | Inventory Management of Automobile After-sales Parts Based on Data Mining | |
| CN117557306B (en) | Management system for classifying consumers based on behaviors and characteristics |
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 |