CN114117231A - Equipment portrait free creation system and method - Google Patents

Equipment portrait free creation system and method Download PDF

Info

Publication number
CN114117231A
CN114117231A CN202111460716.6A CN202111460716A CN114117231A CN 114117231 A CN114117231 A CN 114117231A CN 202111460716 A CN202111460716 A CN 202111460716A CN 114117231 A CN114117231 A CN 114117231A
Authority
CN
China
Prior art keywords
equipment
model
portrait
library
label
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
Application number
CN202111460716.6A
Other languages
Chinese (zh)
Inventor
江晓
王聿隽
陈泽清
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Henghao Information Technology Co ltd
Original Assignee
Shandong Henghao Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong Henghao Information Technology Co ltd filed Critical Shandong Henghao Information Technology Co ltd
Priority to CN202111460716.6A priority Critical patent/CN114117231A/en
Publication of CN114117231A publication Critical patent/CN114117231A/en
Pending legal-status Critical Current

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
    • G06F16/10—File systems; File servers
    • G06F16/18—File system types
    • G06F16/1805—Append-only file systems, e.g. using logs or journals to store data
    • G06F16/1815—Journaling file systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Computing Systems (AREA)
  • Algebra (AREA)
  • Software Systems (AREA)
  • Processing Or Creating Images (AREA)

Abstract

The invention provides a system and a method for freely creating an equipment portrait, which are characterized in that an equipment functional model is constructed by utilizing the working efficiency, the error rate and the static performance parameters of equipment, a six-dimensional recommendation model is constructed to carry out equipment label, preference scoring of the equipment label, equipment label weight and equipment label weight scoring combination on the equipment in a formulated scene, after a constructed matrix is processed through matrix decomposition, a fitting factor is introduced, the technical problem of an equipment portrait system which can only be applied in a certain specific application scene in the prior art is solved, and the technical effect of the equipment portrait system constructed when a data model and a service application model are uncertain is realized.

Description

Equipment portrait free creation system and method
Technical Field
The invention relates to the technical field of Internet of things and artificial intelligence, in particular to a system and a method for freely creating an equipment portrait.
Background
In the internet of things era of everything interconnection, intelligent application of equipment, such as equipment state prediction and early warning, personalized strategy control, additional services and the like, is required to be performed based on the representation of the equipment, so the quality of the intelligent application is directly determined by the performance of an equipment representation system.
In the prior art, the user portrait system can establish user portrait according to the use habits of system users, and then marketing promotion is carried out according to the user portrait, so that good application effects are obtained. However, the user portrait system is mainly used for discovering the preference and interest of people, people create thousands of faces, the portrait model and the application purpose are relatively fixed, and the portrait model belongs to static portrait and is not suitable for equipment portrait with changeable attribute model. Patent No. 2019107009419 entitled "method and system for generating device image" in Dingjia, Zhang Xiao Qing, Zhang Li Chun, etc. proposes a method and system for generating device image, the main idea is: and performing unstructured on the acquired equipment data, and obtaining the attention point and the related attribute value of the equipment for service by using a preset data model in the system.
However, in the process of implementing the technical solution of the embodiments of the present invention, the inventor of the present invention finds that the above-mentioned technology has at least the following technical problems: the technology solves the problem of fixed mode management of POS machines and the like in a certain specific application scene, such as the financial industry, but cannot be competent for an uncertain equipment representation system of a data model and a business application model.
Disclosure of Invention
The embodiment of the invention provides a system and a method for freely creating an equipment portrait, solves the technical problem that the equipment portrait system can only be applied in a certain specific application scene in the prior art, and achieves the technical effect of constructing the equipment portrait system when a data model and a service application model are uncertain.
The technical scheme of the invention is as follows:
a system for free creation of a representation of a device, comprising:
the device log analyzer 010, the feature extractor 020, the device image model 030, the device library 040, the image library 050, and the application scene library 060;
specifically, the device log analyzer 010 is configured to collect daily log files of device operations, and extract related device data information in the files;
the feature extractor 020 is used for extracting features of related data information;
the device portrait model 030 builds a portrait model through device related data information characteristics and adds the portrait model to the portrait library 050;
the equipment library 040 is used for adaptively de-matching an equipment portrait and a scene according to an equipment type when an equipment is added or freely selecting a corresponding equipment portrait and a corresponding scene by a user;
the image library 050 is used for storing the constructed image model and supplementing the image library 050 by updating the image model constructed by the equipment;
the application scene library 060 is used for storing device application scenes and is continuously expanded according to application scenes of users.
A method for freely creating a device portrait, the method comprising: when one device is added, preprocessing the device, wherein the preprocessing refers to acquiring and analyzing daily data and working data of the adding device to obtain a complete label of the device when the adding device is a brand-new device, extracting device characteristics to construct an image model, adding the constructed image model into an image library 050, checking whether the application scene library 060 contains all possible application scenes, and if not, adding a new application scene into the application scene library 060; if the pre-processing is finished, the equipment self-adaptively selects the corresponding portrait model from the portrait library 050 and the corresponding scene from the scene library or self-adaptively selects the proper portrait model and the scene, and further selects the proper portrait model and the scene to finish portrait of the equipment according to self-adaptation of application effects or self-adjustment of equipment labels by users.
Preferably, the method comprises the steps of:
s1, analyzing the added equipment, judging whether the equipment is of an existing equipment type or brand new equipment, and performing corresponding processing on different equipment after determining;
s2, adding equipment to select the portrait model and the scene which are most matched with the equipment from the portrait library 050 and the scene library in a self-adaptive mode, and achieving equipment portrait.
Preferably, the step S1 specifically includes:
constructing a functional model for the equipment according to the working condition of the equipment, and recording the working efficiency of the equipment as a, wherein a is more than or equal to 0 and less than or equal to 100 percent; the error rate is b, wherein b is more than or equal to 0 and less than or equal to 100 percent, the allowable error rate of the equipment is set, and when the allowable error rate is higher than the allowable error rate, the equipment stops working; the static performance index is c, the performance index is a parameter value capable of representing all static performances of the equipment, and c is a set formed by performance parameter values; defining a parameter weight C according to the actual operation of the device1，C2The functional parameters G of the equipment are calculated, and the functional model of the equipment is constructed by utilizing the working efficiency, the error rate and the static performance parameters of the equipment, so that the performance of the equipment can be better displayed, the state of the equipment can be effectively fed back, and convenience is provided in the practical application of the equipment.
Preferably, the step S2 specifically includes:
define the tag of a device as Q ═ Q1，q2，...，qk，...，qNWherein qk is the kth attribute of the tag, N represents the tag number, and the tag weight is defined as S ═ Q, TagRating, where Q represents the tag attribute and TagRating represents the weight attribute; model for portrait (Draw) { M }1，...，MmM represents the number of portrait models, and a six-dimensional recommendation model is constructed based on equipment labels and weights thereof
M＝{Q，S，F，E，Draw，y}
Wherein Q represents the label attribute, S represents the label weight, F represents the scoring set of the user to the device label, E represents the scoring set of the user to the device label weight, Y represents a quintuple relation on Q, S, F, E, Draw, that is, the relation is satisfied
Figure BDA0003389107910000031
Y represents a quintuple set formed by the user scoring the device tag and the tag weight.
Preferably, the step S2 specifically includes:
the method comprises the steps of carrying out matrix decomposition on a matrix formed by equipment labels, equipment label weights, equipment label scores and equipment label weight scores to obtain an approximate matrix, calculating score data D of a set formed by the approximate matrix and an image model by introducing a fitting factor ofo to obtain an optimal image model group, and selecting the image model and the scene which are most matched with the image model from an image library 050 and a scene library in a self-adaptive mode to realize equipment image.
The invention has the beneficial effects that:
1. the invention constructs the equipment function model by utilizing the working efficiency, the error rate and the static performance parameters of the equipment, can better show the performance of the equipment, effectively feeds back the state of the equipment and provides convenience for the practical application of the equipment.
2. The invention constructs a six-dimensional recommendation model to perform the combination of the device label, the preference score of the device label, the device label weight and the device label weight score of the device in a formulated scene, solves the problem of device portrait construction under the uncertain condition of a data model and a service application model, embodies the device characteristics more comprehensively and accurately, and prepares for a subsequent recommendation algorithm.
3. In the recommendation algorithm, after the construction matrix is processed through matrix decomposition, the calculation process complexity is reduced by introducing the fitting factor, the calculation value is more accurate, and a more accurate recommendation result is obtained.
4. The technical scheme of the invention can effectively solve the problem of equipment portrait construction under the uncertain conditions of the data model and the business application model.
Drawings
FIG. 1 is a block diagram of an apparatus representation free creation system according to the present invention;
FIG. 2 is a flow chart of the present invention for freely creating a portrait model.
Detailed Description
The embodiment of the invention provides a system and a method for freely creating an equipment portrait, solves the technical problem that an equipment portrait system can only be applied in a certain specific application scene in the prior art, realizes the equipment portrait constructed when a data model and a business application model are uncertain, and achieves the technical effects of 'thousands of machines and thousands of faces' (different portraits of different equipment) and 'thousands of views and thousands of faces' (different portraits of different business application scenes), and the general idea is as follows:
when one device is added, preprocessing the device, wherein the preprocessing refers to acquiring and analyzing daily data and working data of the adding device to obtain a complete label of the device when the adding device is a brand-new device, extracting device characteristics to construct an image model, adding the constructed image model into an image library 050, checking whether the application scene library 060 contains all possible application scenes, and if not, adding a new application scene into the application scene library 060; if the pre-processing is finished, the equipment self-adaptively selects the corresponding portrait model from the portrait library 050 and the corresponding scene from the scene library or self-adaptively selects the proper portrait model and the scene, and further selects the proper portrait model and the scene to finish portrait of the equipment according to self-adaptation of application effects or self-adjustment of equipment labels by users.
The technical scheme of the invention can effectively solve the problem of constructing the equipment portrait under the uncertain conditions of the data model and the business application model, and the system or the method can better show the equipment performance and effectively feed back the equipment state by constructing the equipment function model by utilizing the working efficiency, the error rate and the static performance parameters of the equipment through a series of effect investigation, is convenient for the practical application of the equipment, constructs a six-dimensional recommendation model to carry out the combination of equipment labels, preference scores of the equipment labels, equipment label weights and equipment label weight scores on the equipment under the formulated scene, solves the problem of constructing the equipment portrait under the uncertain conditions of the data model and the business application model, more comprehensively and accurately reflects the equipment characteristics, prepares for the subsequent recommendation algorithm, processes the constructed matrix through matrix decomposition in the recommendation algorithm, by introducing the fitting factors, the complexity of the calculation process is reduced, the calculation value is more accurate, a more accurate recommendation result is obtained, and finally the most fitting equipment portrait can be effectively obtained.
Referring to fig. 1, the device image free creation system according to the present invention includes the following components:
the device log analyzer 010, the feature extractor 020, the device image model 030, the device library 040, the image library 050, and the application scene library 060;
optionally, when a new piece of equipment is newly added, a user may add a multi-attribute tag of an image, or the system performs daily data acquisition and analysis on the new equipment, extracts equipment characteristics, builds an equipment image model 030, adds the built image model to the image library 050, and checks whether the application scene library 060 contains all the possibilities of all possible application scenes, or else, adds a new application scene to the application scene library 060;
optionally, a device is added, and according to the device type, the device portrait and the scene are adaptively de-matched from the device library 040, or the corresponding device portrait and the scene are freely selected by the user;
the device log analyzer 010 is configured to collect daily log files of device operations and extract related device data information in the files;
the feature extractor 020 is used for extracting features of related data information;
the equipment portrait model 030 is constructed through equipment related data information characteristics, and is added to the portrait library 050;
the image library 050 is used for storing the constructed image model and supplementing the image library 050 by updating the image model constructed by the equipment;
the scene library is used for storing the application scene of the equipment and continuously expanding according to the application scene of the user;
the invention relates to an execution method for freely creating an equipment portrait, which specifically comprises the following steps:
s1, analyzing the added equipment, judging whether the equipment is of an existing equipment type or brand new equipment, and performing corresponding processing on different equipment after determining;
optionally, if the adding device is a brand new device type, the following processing is performed:
s11, performing daily label collection on the equipment, namely collecting a physical information label contained in the equipment and a log file during the operation of the equipment, and extracting related equipment data information in the file;
the collected information comprises equipment static information and equipment dynamic information, wherein the equipment static information comprises but is not limited to equipment name, equipment type, equipment operation, equipment maintenance, equipment warranty, equipment operation manual, equipment group information and use site position; the dynamic information of the device is the constantly changing behavior information of the device, including but not limited to the real-time situation information of the device running in different scene environments.
As a specific embodiment, the collected device dynamic information includes, but is not limited to, the working efficiency, error rate, working condition, working parameters of the device under normal operation, and the running status information of the device under abnormal conditions. The collected information table is as follows:
Figure BDA0003389107910000051
when the equipment is introduced, the static information of the equipment can be obtained; under the normal working and running state of the equipment, the state information of the equipment is recorded in a log file, and the real-time equipment dynamic information is obtained by downloading the daily log file of the equipment.
S12, extracting characteristic information of the acquired data as an equipment label;
obtaining static information and dynamic information of the equipment through step S11, extracting effective information as a label for the equipment, analyzing the characteristic information of the equipment to be portrait from multiple dimensions to obtain characteristic analysis results of each dimension, and obtaining label information of each dimension of the equipment to be portrait according to the characteristic analysis results of each dimension; and formulating corresponding label rules;
from the working angle, analyzing the characteristic information of the equipment to be portrait to obtain the working efficiency, the loss rate, the error rate and the like of the equipment in a certain time;
from the perspective of a client, analyzing the characteristic information of the equipment to be portrait to obtain the functions, the use operation, the operation and the maintenance of the equipment and the like;
from the use time perspective, analyzing the characteristic information of the equipment to be portrait to obtain fuzzy users of equipment service, wherein if the equipment is basically always in a running working state, the equipment is possibly used by factory workers who fall over three shifts;
optionally, the feature information of the device to be imaged includes, but is not limited to, associated feature information between devices, feature information when the device operates normally, and feature information when the device operates abnormally;
label information is defined according to the characteristics of the collected information, and a weight is defined to represent an index, namely equipment information reaction; for constructing an image model.
S13, constructing an equipment portrait model 030 according to the label information, and adding the newly constructed equipment portrait model 030 to a portrait library 050; meanwhile, the application scenes of the equipment are summarized, and if a new use scene exists, the new use scene is timely supplemented to the equipment library 040;
equipment label: type, age, place of departure, place of work, etc.;
a behavior tag; working time period, working duration, working efficiency and use frequency;
content labeling: the frequency of use of each of the functions,
common models include, but are not limited to, binding rate type models, activity level models, early warning type models, industry type models, feature analysis models, prediction type models, functional models and classification models;
creating an portrait model to perfect the portrait database 050, and constructing a region and site model according to the collected static and dynamic information of the equipment as follows:
preferably, the device is numbered and described in text according to the geographical location (specific city coordinates) and the location (indoor or outdoor), where the device is located, for example, the location of the area is: SDYT01 in the smoke table chamber, LNDL00 outside the large chamber;
preferably, an activity model is established according to the working time of the equipment, and the working time period and the working time length of the equipment are obtained by downloading daily logs of the equipment to provide data, so that the working time period and the working time length of the workers in the company/factory are reflected;
preferably, a function model is constructed for the equipment according to the working condition of the equipment, the working efficiency of the equipment is recorded as a, wherein a is more than or equal to 0 and less than or equal to 100 percent; the error rate is b, wherein b is more than or equal to 0 and less than or equal to 100 percent, the error rate allowed by the equipment can be set, and when the error rate is more than the allowed error rate, the equipment stops working; the static performance index is c, the performance index is a parameter value capable of representing all static performances of the equipment, and c is a set formed by performance parameter values. Defining the parameter weight C1, C according to the actual operation of the device2Functional parameters G of the computing device:
G＝(C1a-C2b)×||c||
the device comprises a power supply, a power supply and the like, wherein | · | |, represents a norm, whether the device works efficiently is judged through the value of G, a threshold value is set for distinguishing judgment, and when G is not less than or not; when G < sigma, the current device operates inefficiently.
The invention constructs the equipment function model by utilizing the working efficiency, the error rate and the static performance parameters of the equipment, can better show the performance of the equipment, effectively feeds back the state of the equipment and provides convenience for the practical application of the equipment.
And (4) inspecting the application scene of the newly added equipment, and adding scenes which are not in the scene library so as to supplement the scene library.
Optionally, if the added device is an existing device type and a new device type is added to perform the processing operations of the steps S11-S13, the following processing is performed;
s2, adding equipment to select the portrait model and the scene which are most matched with the equipment from the portrait library 050 and the scene library in a self-adaptive mode, and achieving equipment portrait.
S21, automatically selecting an image model from an image library 050 for the added equipment;
through an equipment portrait recommendation algorithm, a proper expert knowledge base rule is recommended for a user, fault diagnosis of current equipment is timely and correspondingly conducted, and a solution is provided.
Labeling devices under given scene conditionsMeaning Q ═ Q1，q2，...，qk，...，qNWherein qk is the kth attribute of the tag, N represents the tag number, and the tag weight is defined as S ═ Q, TagRating, where Q represents the tag attribute and TagRating represents the weight attribute. Model for portrait (Draw) { M }1，...，MmM represents the number of portrait models, and a six-dimensional recommendation model is constructed based on equipment labels and weights thereof
M＝{Q，S，F，E，Draw，Y}
Wherein Q represents the label attribute, S represents the label weight, F represents the scoring set of the user to the device label, E represents the scoring set of the user to the device label weight, Y represents a quintuple relation on Q, S, F, E, Draw, that is, the relation is satisfied
Figure BDA0003389107910000082
Y represents a quintuple set formed by the user scoring the device tag and the tag weight.
The invention constructs a six-dimensional recommendation model to perform the combination of the device label, the preference score of the device label, the device label weight and the device label weight score of the device in a formulated scene, solves the problem of device portrait construction under the uncertain condition of a data model and a service application model, embodies the device characteristics more comprehensively and accurately, and prepares for a subsequent recommendation algorithm.
As a specific example, m ═ q, (q, tagRating), f, e, draw ∈ Y, which indicates that the user scores f for the tag q of the device, and scores e for the tag weight (q, tagRating) of the device, where q indicates the tag of the device and tagRating indicates that the device weights on the tag feature as tagRating.
The process of adaptively portraying a device using a recommendation algorithm is shown in FIG. 2.
The recommendation algorithm specifically comprises the following steps:
and solving the characteristic information of the equipment portrait model 030 from the scoring data D { (q, s, f, e, draw) | Y { (q, s, f, e, draw) ∈ Y }, so as to select the portrait model under the appropriate scene for the equipment. The basic idea is as follows: and searching the image model with the highest conformity according to the preference score of the user on the equipment label and the equipment label, the weight of the equipment label and the weight score of the equipment label, namely calculating the conformity of the equipment characteristics and the image model, and selecting the image model group with higher conformity to be the equipment image model 030. The specific process comprises the following steps:
firstly, defining a tag feature matrix All ═ Q of the equipment; f; s; e, the matrix All is a matrix with dimension of M multiplied by N and composed of a set Q, F, S and E, and the matrix All is subjected to the following matrix decomposition:
AllM×N＝WM×M×∑M×N×VT N×N
wherein the content of the first and second substances,
Figure BDA0003389107910000081
is a decomposition matrix, diagonal element σ1，σ2… is called the decomposition value of the matrix All, decreasing non-negatively, with the other elements being 0. W ═ W1，w2，...，wi，...，wM)，wiIs the left decomposition vector of the matrix All, describing the characteristics of All, corresponding to σi(ii) a And sigmaiA larger value of (A) indicates the characteristic w of AlliThe greater the importance of; in fact wiIs a square matrix ALLALLTThe feature vector of (2). V ═ V (V)1，v2，...，vi，...，vN)，viIs the right decomposition vector of the matrix All, which also describes the characteristics of All, corresponding to σi(ii) a And sigmaiGreater values of (A) indicate the characteristic v of AlliThe greater the importance of; it is true viIs a square matrix ALLALLTThe feature vector of (2).
In actual operation, the decomposition vectors with smaller importance are removed, only r decomposition values are reserved, and the r decomposition values correspond to r left and right decomposition vectors, namely
Figure BDA0003389107910000091
An approximation matrix is then obtained:
Figure BDA0003389107910000092
to approximate matrix
Figure BDA0003389107910000093
And calculating the degree of fit with all portrait models in the portrait library 050, wherein the approximate matrix is a matrix formed by most effective equipment labels, equipment label weights, equipment label scores and equipment label weight scores. The image model is represented by a set Draw, { Mode }1，Mode2，...，ModepP represents the number of image models, by introducing a fitting factor:
Figure BDA0003389107910000094
the fit score is calculated using the following formula:
Figure BDA0003389107910000095
wherein m represents an approximate construction matrix
Figure BDA0003389107910000096
M, n denotes the nth column of the approximate construction matrix,<·>and representing a numerical value obtained by substituting the optimized equipment label, the equipment label weight, the equipment label score and the equipment label weight score into the portrait model Mode.
In the recommendation algorithm, after the construction matrix is processed through matrix decomposition, the calculation process complexity is reduced by introducing the fitting factor, the calculation value is more accurate, and a more accurate recommendation result is obtained.
Finally, putting the calculated fit score into a score data set D, namely D ═ D1，D2，...，DpSetting a threshold value gamma, screening all elements larger than the gamma from the scoring data set D, wherein the corresponding portrait model is the most representative portrait model of the equipment in the current sceneAn image model of the device image.
S22, manually selecting an image model from the image library 050 for the added equipment;
and selecting a proper portrait model and an application scene from the portrait library 050 and the scene library by a user according to actual requirements and equipment characteristics to finish portrait creation of the equipment.
S23, carrying out a detailed process of self-adaptive construction of an equipment portrait for specific equipment;
as a specific embodiment of the present invention, the POS device includes raw data of the POS device, including: the system comprises equipment information, equipment operation, equipment maintenance, client information, group reports and the like, wherein the dynamic information comprises use frequency, use environment, scenes, objects and the like;
all possible portrait models of the POS machine comprise a binding rate model, an activity model, an early warning model, an industry model, a characteristic analysis model, a prediction model, a function model and a classification model;
setting a scene as a civilian clothing store in a certain market, a store owner collects a customer shopping money by using a POS machine, and marking a device tag combination as Q ═ { device information, operation information, device warranty, customer information, group report, use frequency, use time, use object }, a device tag weight S ═ 0.05, 0.1, 0.05, 0.15, 0.05, 0.2, 0.25, 0.2}, a device tag score F ═ 1, 2, 1, 3, 1, 4, 5, 4}, a device weight score E ═ F ═ {1.5, 2, 1, 3, 1, 4, 5, 4}, a calculation process shown in fig. 2, i.e., a step S2, and a score data set D is obtained by using a recommendation model and a recommendation algorithm, i.e., D ═ D1，D2，...，D8And setting a threshold value gamma, screening all elements larger than the gamma from the grading data set D, wherein the corresponding image model (activity model, early warning model, industry model and functional model) is the image model which can represent the image of the equipment most under the current scene.
The technical scheme in the embodiment of the invention at least has the following technical effects or advantages:
1. the invention constructs the equipment function model by utilizing the working efficiency, the error rate and the static performance parameters of the equipment, can better show the performance of the equipment, effectively feeds back the state of the equipment and provides convenience for the practical application of the equipment.
2. The invention constructs a six-dimensional recommendation model to perform the combination of the device label, the preference score of the device label, the device label weight and the device label weight score of the device in a formulated scene, solves the problem of device portrait construction under the uncertain condition of a data model and a service application model, embodies the device characteristics more comprehensively and accurately, and prepares for a subsequent recommendation algorithm.
3. In the recommendation algorithm, after the construction matrix is processed through matrix decomposition, the calculation process complexity is reduced by introducing the fitting factor, the calculation value is more accurate, and a more accurate recommendation result is obtained.
4. The technical scheme of the invention can effectively solve the problem of equipment portrait construction under the uncertain conditions of the data model and the business application model.
Effect investigation:
the technical scheme of the invention can effectively solve the problem of constructing the equipment portrait under the uncertain conditions of the data model and the business application model, and the system or the method can better show the equipment performance and effectively feed back the equipment state by constructing the equipment function model by utilizing the working efficiency, the error rate and the static performance parameters of the equipment through a series of effect investigation, is convenient for the practical application of the equipment, constructs a six-dimensional recommendation model to carry out the combination of equipment labels, preference scores of the equipment labels, equipment label weights and equipment label weight scores on the equipment under the formulated scene, solves the problem of constructing the equipment portrait under the uncertain conditions of the data model and the business application model, more comprehensively and accurately reflects the equipment characteristics, prepares for the subsequent recommendation algorithm, processes the constructed matrix through matrix decomposition in the recommendation algorithm, by introducing the fitting factors, the complexity of the calculation process is reduced, the calculation value is more accurate, a more accurate recommendation result is obtained, and finally the most fitting equipment portrait can be effectively obtained.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (6)

1. A system for free creation of a representation of a device, comprising: the method comprises the following steps:
a device log analyzer (010), a feature extractor (020), a device representation model (030), a device library (040), a representation library (050), an application scenario library (060);
specifically, the equipment log analyzer (010) is used for collecting daily log files of equipment operation and extracting relevant equipment data information in the files;
the feature extractor (020) is used for extracting features of related data information;
the device portrait model (030) builds a portrait model from device-related data information features and adds the portrait model to the portrait library (050);
the equipment library (040) is used for adaptively de-matching equipment portraits and scenes according to equipment types when one piece of equipment is added or freely selecting corresponding equipment portraits and scenes by a user;
the image library (050) is used for storing the constructed image model and supplementing the image library (050) by updating the image model constructed by the equipment;
the application scene library (060) is used for storing the device application scenes and is continuously expanded according to the application scenes of the users.
2. A free creation method of device portrait, the method is characterized in that:
when one device is added, preprocessing the device, wherein the preprocessing refers to acquiring and analyzing daily data and working data of the added device to obtain a complete label of the device when the added device is a brand-new device, extracting device characteristics to construct an image model, adding the constructed image model into the image library (050), checking whether the application scene library (060) contains all possible application scenes, and if not, adding a new application scene into the application scene library (060); if the pre-processing is finished, the equipment selects the corresponding portrait model and the corresponding scene in the portrait base (050) in a self-adaptive mode or selects the proper portrait model and the scene in a self-adaptive mode, the equipment labels are adjusted in a self-adaptive mode or by the user, and the proper portrait model and the scene are further selected to finish equipment portrait.
3. A device representation free creation method as claimed in claim 2, said method comprising the steps of:
s1, analyzing the added equipment, judging whether the equipment is of an existing equipment type or brand new equipment, and performing corresponding processing on different equipment after determining;
s2, adding equipment to select the portrait model and the scene which are most matched with the equipment from the portrait library (050) and the scene library in a self-adaptive mode, and achieving portrait of the equipment.
4. The method for freely creating a device representation as claimed in claim 3, wherein said step S1 specifically comprises:
constructing a functional model for the equipment according to the working condition of the equipment, and recording the working efficiency of the equipment as a, wherein a is more than or equal to 0 and less than or equal to 100 percent; the error rate is b, wherein b is more than or equal to 0 and less than or equal to 100 percent, the allowable error rate of the equipment is set, and when the allowable error rate is higher than the allowable error rate, the equipment stops working; the static performance index is c, the performance index is a parameter value capable of representing all static performances of the equipment, and c is a set formed by performance parameter values; defining a parameter weight C according to the actual operation of the device1，C2The functional parameters G of the equipment are calculated, and the functional model of the equipment is constructed by utilizing the working efficiency, the error rate and the static performance parameters of the equipment, so that the performance of the equipment can be better displayed, the state of the equipment can be effectively fed back, and convenience is provided in the practical application of the equipment.
5. The method for freely creating a device representation as claimed in claim 3, wherein said step S2 specifically comprises:
define the tag of a device as Q ═ Q1,q2,…,qk,…,qNWherein q iskThe k-th attribute of the label, N represents the label number, and the label weight is defined as S ═ Q, TagRating, wherein Q represents the label attribute, and TagRating represents the weight attribute; model for portrait (Draw) { M }1,…,MmM represents the number of portrait models, and a six-dimensional recommendation model is constructed based on equipment labels and weights thereof
M＝{Q,S,F,E,Draw,Y}
Wherein Q represents the label attribute, S represents the label weight, F represents the scoring set of the user to the device label, E represents the scoring set of the user to the device label weight, Y represents a quintuple relation on Q, S, F, E, Draw, that is, the relation is satisfied
Figure FDA0003389107900000021
Y represents a quintuple set formed by the user scoring the device tag and the tag weight.
6. The method for freely creating a device representation as claimed in claim 3, wherein said step S2 specifically comprises:
the method comprises the steps of carrying out matrix decomposition on a matrix formed by utilizing equipment labels, equipment label weights, equipment label scores and equipment label weight scores to obtain an approximate matrix, calculating score data D of a set formed by the approximate matrix and an image model by introducing a fitting factor ofo to obtain an optimal image model group, and selecting the image model and the scene which are most matched with the image model from an image library (050) and a scene library in a self-adaptive mode to realize equipment image.
CN202111460716.6A 2021-12-02 2021-12-02 Equipment portrait free creation system and method Pending CN114117231A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111460716.6A CN114117231A (en) 2021-12-02 2021-12-02 Equipment portrait free creation system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111460716.6A CN114117231A (en) 2021-12-02 2021-12-02 Equipment portrait free creation system and method

Publications (1)

Publication Number Publication Date
CN114117231A true CN114117231A (en) 2022-03-01

Family

ID=80366364

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111460716.6A Pending CN114117231A (en) 2021-12-02 2021-12-02 Equipment portrait free creation system and method

Country Status (1)

Country Link
CN (1) CN114117231A (en)

Similar Documents

Publication Publication Date Title
CN111062757B (en) Information recommendation method and system based on multipath optimizing matching
CN111369016B (en) Image recognition-based on-site operation and maintenance method and system
CN114742477B (en) Enterprise order data processing method, device, equipment and storage medium
CN114494837A (en) Intelligent density identification method and system for fishery resources
CN113222149B (en) Model training method, device, equipment and storage medium
CN113971527A (en) Data risk assessment method and device based on machine learning
CN111931809A (en) Data processing method and device, storage medium and electronic equipment
CN111506504A (en) Software development process measurement-based software security defect prediction method and device
CN113205403A (en) Method and device for calculating enterprise credit level, storage medium and terminal
Ólafsson et al. Intelligent partitioning for feature selection
CN113609393B (en) Digital platform based on data service and data management
CN114282607A (en) Double-sieve model-based dispersion trajectory analysis method and system
CN117709691A (en) Intelligent sub-packaging management method and system based on cloud service
CN116452212B (en) Intelligent customer service commodity knowledge base information management method and system
CN117312678A (en) Intelligent recommendation method, equipment and storage medium for potential clients based on big data
CN112883018A (en) Banking talent information prediction method and device based on big data analysis
CN114117231A (en) Equipment portrait free creation system and method
CN115659044A (en) Recommendation method and system for people and sentry matching, electronic equipment and storage medium
CN114971425A (en) Database information monitoring method, device, equipment and storage medium
CN114185967A (en) Financial data mining method based on cloud computing
CN113421133A (en) Network marketing system based on customer matching
CN113379257A (en) Automatic intelligent distribution method for sales leads
CN112434648A (en) Wall shape change detection method and system
CN111930815A (en) Method and system for constructing enterprise portrait based on industry attribute and business attribute
CN110443305A (en) Self-adaptive features processing method and processing device

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