CN117114694A - Big data analysis system and method based on CRM - Google Patents
Big data analysis system and method based on CRM Download PDFInfo
- Publication number
- CN117114694A CN117114694A CN202211715862.3A CN202211715862A CN117114694A CN 117114694 A CN117114694 A CN 117114694A CN 202211715862 A CN202211715862 A CN 202211715862A CN 117114694 A CN117114694 A CN 117114694A
- Authority
- CN
- China
- Prior art keywords
- user
- related personnel
- data
- data analysis
- users
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06Q30/01—Customer relationship services
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F16/242—Query formulation
- G06F16/2433—Query languages
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a big data analysis system and a method based on CRM, which belong to the field of customer relation management, wherein the big data analysis system comprises a data acquisition module, a database, a data analysis module and a terminal feedback module, wherein the data acquisition module is used for acquiring basic data information, acquiring user image information through camera equipment, the database is used for storing the acquired data information and analysis results, the data analysis module is used for analyzing shopping states of users and optimally planning paths between related personnel and the users, and the terminal feedback module is used for displaying and reminding the related personnel through terminal equipment according to the analysis results. According to the invention, basic data information and user image information are collected, shopping states of users are analyzed, optimal planning is performed, and meanwhile, display and reminding are performed through terminal equipment, so that the management of the client relationship is enhanced, and the working service efficiency of related personnel is improved.
Description
Technical Field
The invention relates to the field of customer relationship management, in particular to a big data analysis system and method based on CRM.
Background
CRM means customer relationship management, namely, in order to improve core competitiveness, enterprises coordinate interactions between the enterprises and customers in sales, marketing and service by utilizing corresponding information technology and Internet technology, thereby improving management modes and providing innovative personalized customer interaction and service processes for customers. Its final goal is to attract new customers, retain old customers, turn existing customers into faithful customers, and increase the market.
However, when the customers do physical shopping under the line, some customers like to walk by themselves, and the service personnel are not required to follow the customers all the time, but when the customers want to consult, the service personnel can serve other customers, and the customers are required to wait; meanwhile, the situation that the number of clients in a place is larger than the number of service staff can occur, the service staff can not timely serve each client, the waiting time of some clients is long, and therefore the incapacitation of the client is caused, and the loss of the clients is caused.
Therefore, how to timely serve the clients is quite necessary to improve the working efficiency of related personnel and reduce the loss of clients. Thus, there is a need for a CRM-based big data analysis system and method.
Disclosure of Invention
The invention aims to provide a big data analysis system and a method based on CRM, which are used for collecting basic data information, collecting user image information through a camera device, encrypting and storing, analyzing shopping states of users, and therefore optimally planning paths between related personnel and the users, and simultaneously displaying and reminding through terminal equipment, so that the problems in the background technology are solved.
In order to solve the technical problems, the invention provides the following technical scheme: a CRM-based big data analysis system, the big data analysis system comprising: the system comprises a data acquisition module, a database, a data analysis module and a terminal feedback module;
the data acquisition module is connected with the database, the database is connected with the data analysis module, and the data analysis module is connected with the terminal feedback module; the system comprises a data acquisition module, a data analysis module and a terminal feedback module, wherein the data acquisition module is used for acquiring basic data information, acquiring user image information through a camera device, the database is used for storing acquired data information and analysis results, the data analysis module is used for analyzing shopping states of users and optimally planning paths between related personnel and the users, and the terminal feedback module is used for displaying and reminding the related personnel through the terminal device according to the analysis results.
Further, the data acquisition module comprises a basic data input unit and an image acquisition unit, wherein the basic data input unit is used for inputting basic data information of public service places, such as an electronic map, a goods shelf placement position, a product category and the like, the public service places are movable places where people are often gathered and used by the public or serve people, and all public places and facilities thereof, including entertainment places, shops, store shops and the like, used by working, learning, economy, culture, social, entertainment, sports, visiting, medical treatment, health, rest, travel and meeting part of living demands can be provided for the public, and the image acquisition unit is used for acquiring image information of users through camera shooting equipment, such as the existing safety cameras of the public service places and the like.
Further, the database includes a data storage unit, a data encryption unit and a data cleaning unit, the data storage unit stores data through NoSQL, which is broadly referred to as a non-relational database, different from a relational database, which does not guarantee the ACID characteristics of the relational data, and has the following advantages: a. the NoSQL database is easy to expand, and has various types, but one common characteristic is to remove the relational characteristics of the relational database; b. the data has no relation, so the expansion is very easy. Intangible and extensible capabilities are also brought about at the architectural level; c. the NoSQL database has very high read-write performance, particularly under the condition of large data volume, the NoSQL database also has excellent performance, and the NoSQL database has simple structure due to the irrelevance of the NoSQL database; the data encryption unit encrypts data through an RC4 algorithm, so that the data safety can be effectively ensured, the RC4 algorithm is a stream cipher, the stream cipher encryption process uses an initial key to generate a pseudo-random cipher stream, an input element sequence is continuously processed, the cipher stream and the input sequence are usually subjected to exclusive-or operation to generate a corresponding continuous output sequence, the decryption process uses a pseudo-random stream generator algorithm which is the same as the encryption process to generate the same pseudo-random cipher stream to the initial key, and the pseudo-random cipher stream and the ciphertext sequence are subjected to exclusive-or operation to obtain a plaintext sequence; the key length of the RC4 algorithm is variable, byte-oriented operation is used, analysis indicates that the cipher flow period of the RC4 algorithm is possibly more than 10100 completely, and the data cleaning unit is used for automatically cleaning data after a user leaves the public service place, so that the space for storing the data is saved, the redundancy of the data is avoided, and the working efficiency of the system is improved.
Further, the data analysis module comprises a user state analysis unit and a path planning unit, the user state analysis unit is used for analyzing the shopping state of the user according to the collected user image information, knowing the urgency degree of the user for carrying out service by related personnel, the related personnel comprise shopping guides, sales personnel, business handling personnel and the like, the path planning unit is used for planning the optimal path between the related personnel and the user according to the urgency degree of the analyzed user for carrying out service by the related personnel, so that the service efficiency of the related personnel is improved when the number of shopping persons is large, the optimal shopping experience is brought to the user, the management and maintenance of the relation of the user are enhanced, and the continuous development of enterprises is promoted.
Further, the terminal feedback module comprises a screen display unit and an auxiliary prompt unit, wherein the screen display unit displays real-time paths of related personnel through screen display equipment such as mobile phones or computers, and the auxiliary prompt unit is used for carrying out auxiliary prompt on the related personnel in an additional mode such as alarm sound effect, music sound or vibration, so that the related personnel can acquire analysis results timely and serve clients timely, management of client relations is promoted, even if the related personnel cannot see information displayed on the screen under the condition that the clients are more, prompt can be received and serve the clients timely, the clients are prevented from waiting for too long and being restless to cause the loss of the clients, and the service efficiency of the related personnel is improved.
A big data analysis method based on CRM, comprising the steps of:
s1, acquiring basic data information, acquiring user image information through camera equipment, and encrypting and storing;
s2, analyzing the shopping state of the user according to the acquired image information to obtain the urgent degree of the requirement that the user needs related personnel to perform service;
s3, planning an optimal path between the related personnel and the user according to the collected image information and the analyzed degree of urgency of the user for requiring the related personnel to perform service;
and S4, displaying the analysis result to related personnel through a screen display device, and carrying out auxiliary prompt.
Further, in step S2, facial image information of the user is collected through the image capturing device, face tracking is performed on the user through the OpenCV technology, distance is measured through the image ranging technology, and the OpenCV is a cross-platform computer vision and machine learning software library issued based on the apache2.0 license, and can be run on the Linux, windows, android and Mac OS operating systems. The system is lightweight and efficient, is composed of a series of C functions and a small number of C++ classes, provides interfaces of Python, ruby, MATLAB and other languages, and realizes a plurality of general algorithms in the aspects of image processing and computer vision;
the face gesture category of the collected user is as followsWhere n is denoted as the image capturing apparatus number, t is denoted as the time at which the user stands at a certain position and focuses on a certain specific product object, the face pose category ∈10 photographed by the image capturing apparatus n>Is determined, face gesture category +.>Attention target M at t t And face position->Constraint, observe likelihood for face by the following formula +.>And (3) performing calculation:
wherein,expressed as face observation variable, face observation likelihood +.>Expressed as face observation variable +.>By face gesture category->The probability of generation, δ, is expressed as a normalization factor, C γ Image subspace expressed as human face gesture category, subspace refers to partial space with dimension less than or equal to full space, mapping of high-dimensional features to low-dimensional space is realized through projection, and the subspace refers to the part space with dimension less than or equal to full space>Expressed as the distance of the face image to the image subspace, such as reconstruction errors and the like when the image is projected to the sub-control, alpha 2 Represented as variance;
when a customer is looking at a product, it is possible to continue to watch the target, or to transfer to another target, and enhancement is neededTime smoothness, a transition probability matrix P (M) between different attention objects at adjacent moments of a user is calculated by the following formula t |M t-1 ) And (3) performing calculation:
wherein M is t Representing the attention target of the user at time t, M t-1 Representing the attention objective of the user at time t-1, P f Expressed as the probability that adjacent moments are the same target, this value can be set by the relevant personnel themselves, usually P f Has a value of approximately 1, M Total (S) Expressed as a common target quantity of product;
the joint probability distribution P is calculated by the following formula:
wherein T is the total time of data acquisition, N is the total number of the image capturing devices,expressed as +.>Where the attention target is M t The face posture category is +.>The probability of (2) is calculated by a counting statistical method, and a user can watch various products when staying at one position, so that various face gestures exist; numbering the users, and setting the duration time of the joint distribution probability as t Holding device The values are ordered from big to small to form a set r= { t Hold 1 ，t Hold 2 ，…，t Hold m Where m is denoted as user number, t Holding device The larger the value of (C) is, the corresponding user needs related personnel to perform serviceThe higher the urgency of demand.
Further, in step S3, according to the analyzed degree of urgency of the user' S need for related personnel to perform service, although the user is in motion, the user does not move in a large range when paying attention to the commodity, so that the user can be regarded as a stationary state and perform optimal path planning in real time;
the image information of the user is collected in real time, the position of the user is placed in a plane coordinate system which can be set by the related personnel, for example, a plane rectangular coordinate system is established according to the actual topography of the place, the position of the related personnel is used as a starting point, and the position coordinate is (x) i ，y i ) The position of the user is the end point, and the position coordinates are (x j ，y j ) The position coordinates of the adjacent node f of the position of the relevant person are (x k ，y k ) Form the set r= { (x) 1 ，y 1 )，(x 2 ，y 2 )，…，(x m ，y m ) Where m is the number of adjacent nodes, the adjacent nodes represent positions, such as corner positions, where the relevant person passes when traveling to the user position, and the Manhattan distance d between the user position and the adjacent node position is calculated by the following formula 1 (f) And (3) performing calculation:
d 1 (f)＝(x j -x k )+(y j -y k )；
planning a path from the related personnel position to the user position through the following formula:
d(f)＝d 1 (f)+d 2 (f)；
wherein d (f) is the estimated distance from the relevant person position to the user position after passing the adjacent node position, d 2 (f) Expressed as the actual distance from the relevant person's location to the location of the neighboring node, the other neighboring nodes calculate the same, then the optimal path d=d (f) min ；
Taking the user z as a center point, setting the radius as r, constructing Gao Weiqiu area Y, and constructing a high-dimensional space is a physical theory. According to the theory of M proposed in the 90 s, the universe is eleven-dimensional and consists of a vibrating plane; the number of users included in the area Y of Gao Weiqiu is S, and the offset average S is calculated by the following formula:
where z' represents a point contained within region Y of Gao Weiqiu, moving the center point toward the vector direction of the offset mean value, then: z t+1 ＝S t +z t The method comprises the steps of carrying out a first treatment on the surface of the Wherein z is t Represents the center of time t, z t+1 Represents the center of time t+1, S t Representing the offset average value at the time t; the weight ω is calculated by the following formula:
wherein, h (·) is expressed as a negative of the inverse of a kernel function, the kernel function is that the support vector machine maps the input space to the high-dimensional feature space through a certain nonlinear transformation, the kernel function comprises a linear kernel function, a polynomial kernel function, a gaussian kernel function and the like, and θ is expressed as a weight coefficient. The average value shift classification is carried out on the users, the region division is carried out according to each divided type, and related personnel serve the users in the corresponding region, so that the service quality is effectively improved, and the situations that the users adjacent to each other in urgent degree are too far apart to be completely served or the number difference of the users in the region is large, the related personnel are ill-served and the like are avoided; in the area, according to the analyzed demand urgency degree of the users for carrying out service by related personnel, among two users with adjacent demand urgency degrees, the user with high demand urgency degree is a first service grade, the position is a starting point, the user with low demand urgency degree is a second service grade, the position is an ending point, and calculation is carried out, so that a complete optimal path is obtained; when the demand of the user is reduced, the situation that the user possibly has waiting for impatience is indicated, the user is marked, the service level of the user is improved, the position of the related person is taken as a starting point, and the position of the user is taken as an end point to conduct path planning.
Further, in step S4, according to the analysis result, the optimal path is displayed to the related personnel through the screen display device, such as a mobile phone or a computer, and the duration of the joint distribution probability of different users is highlighted through different colors, for example, the high degree of urgency of the user is displayed through red, and meanwhile, auxiliary prompt is performed, for example, vibration or sound is used, so that even if the related personnel cannot view the display device in time, prompt can be received quickly, the impatience caused by the long time of other users is avoided, the management of the client relationship is ensured, and the working service efficiency of the related personnel is improved.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, basic data information such as an electronic map and a goods shelf placement position are collected, user image information is collected through the camera equipment, the shopping state of a user is analyzed according to the collected information, the urgent degree of the need of the user for service by related personnel is obtained, and the ordering is carried out, so that the related personnel can conveniently and quickly know whether the client needs service or not, and the waiting time is prolonged, the incapacitation of the client caused by overlong waiting time of the client is avoided, the loss caused by the loss of the client is avoided, and the good relation between enterprises and the client is maintained. The method has the advantages that the area is divided, the optimal planning is carried out on the paths between related personnel and users in the area, the service quality of the related personnel is improved, the situation that the distance span of the users adjacent to the urgent degree of the demand is large, the related personnel are difficult to timely serve is avoided, meanwhile, the situation that the related personnel arrive at the client side to serve the client rapidly through the terminal equipment for displaying and reminding is facilitated, the work service efficiency of the related personnel is improved, the experience of the client is improved, and the enterprise benefit is guaranteed.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic diagram of the modular composition of a CRM-based big data analysis system of the present invention;
FIG. 2 is a flow chart of the steps of a CRM-based big data analysis method of the present invention;
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-2, the present invention provides the following technical solutions: a CRM-based big data analysis system, the big data analysis system comprising: the system comprises a data acquisition module, a database, a data analysis module and a terminal feedback module;
the data acquisition module is connected with the database, the database is connected with the data analysis module, and the data analysis module is connected with the terminal feedback module;
the system comprises a data acquisition module, an image acquisition unit and an image acquisition unit, wherein the data acquisition module is used for acquiring basic data information of a user through image pickup equipment, the data acquisition module comprises a basic data input unit and the image acquisition unit, the basic data input unit is used for inputting basic data information of public service places, such as an electronic map, a goods shelf placement position, a product category and the like, the public service places are places where people are often gathered, are used by the public or serve people and the public, and can provide public with all public places and facilities used by work, study, economy, culture, social, entertainment, sports, visiting, medical treatment, sanitation, rest, travel and meeting part of life demands, including entertainment places, shops, store stores and the like, and the image acquisition unit is used for acquiring image information of the user through image pickup equipment, such as a safety camera of the public service places.
The database is used for storing collected data information and analysis results, the database comprises a data storage unit, a data encryption unit and a data cleaning unit, the data storage unit stores data through NoSQL, the NoSQL is widely used as a non-relational database, and is different from a relational database, the ACID characteristics of the relational data are not guaranteed, and the database has the following advantages: a. the NoSQL database is easy to expand, and has various types, but one common characteristic is to remove the relational characteristics of the relational database; b. the data has no relation, so the expansion is very easy. Intangible and extensible capabilities are also brought about at the architectural level; c. the NoSQL database has very high read-write performance, particularly under the condition of large data volume, the NoSQL database also has excellent performance, and the NoSQL database has simple structure due to the irrelevance of the NoSQL database; the data encryption unit encrypts data through an RC4 algorithm, so that the data safety can be effectively ensured, the RC4 algorithm is a stream cipher, the stream cipher encryption process uses an initial key to generate a pseudo-random cipher stream, an input element sequence is continuously processed, the cipher stream and the input sequence are usually subjected to exclusive-or operation to generate a corresponding continuous output sequence, the decryption process uses a pseudo-random stream generator algorithm which is the same as the encryption process to generate the same pseudo-random cipher stream to the initial key, and the pseudo-random cipher stream and the ciphertext sequence are subjected to exclusive-or operation to obtain a plaintext sequence; the key length of the RC4 algorithm is variable, byte-oriented operation is used, analysis indicates that the cipher flow period of the RC4 algorithm is possibly more than 10100 completely, and the data cleaning unit is used for automatically cleaning data after a user leaves the public service place, so that the space for storing the data is saved, the redundancy of the data is avoided, and the working efficiency of the system is improved.
The data analysis module is used for analyzing shopping states of users and optimally planning paths between related personnel and the users, the data analysis module comprises a user state analysis unit and a path planning unit, the user state analysis unit is used for analyzing the shopping states of the users according to collected user image information, knowing the urgency degree of the users for carrying out service by the related personnel, the related personnel comprise shopping guides, sales personnel, business handling personnel and the like, the path planning unit is used for planning the optimal paths between the related personnel and the users according to the urgency degree of the analyzed users for carrying out service by the related personnel, so that when the number of shopping persons is large, the service efficiency of the related personnel is improved, optimal shopping experience is brought to the clients, management and maintenance of the client relationship are enhanced, and continuous development of enterprises is promoted.
The terminal feedback module is used for displaying and reminding related personnel through terminal equipment according to analysis results, the terminal feedback module comprises a screen display unit and an auxiliary prompt unit, the screen display unit is used for displaying real-time paths of the related personnel through the screen display equipment such as a mobile phone or a computer, the auxiliary prompt unit is used for carrying out auxiliary reminding on the related personnel through additional modes such as an alarm sound effect, music sound or vibration, the related personnel can be guaranteed to obtain the analysis results in time, the related personnel can serve clients in time, management of the client relationship is promoted, even if the related personnel cannot see information displayed on the screen under the condition of more clients, the related personnel can receive reminding, serve the clients in time, the clients are prevented from waiting for too long time and being uncomfortable to cause the loss of the clients, and the service efficiency of the related personnel is improved.
A big data analysis method based on CRM, comprising the steps of:
s1, acquiring basic data information, acquiring user image information through camera equipment, and encrypting and storing;
s2, analyzing the shopping state of the user according to the acquired image information to obtain the urgent degree of the requirement that the user needs related personnel to perform service;
face image information of a user is acquired through an image pickup device, face tracking is performed on the user through an OpenCV technology, distance is measured through an image ranging technology, and the OpenCV is a cross-platform computer vision and machine learning software library based on Apache2.0 license issue and can be run on Linux, windows, android and Mac OS operating systems. The system is lightweight and efficient, is composed of a series of C functions and a small number of C++ classes, provides interfaces of Python, ruby, MATLAB and other languages, and realizes a plurality of general algorithms in the aspects of image processing and computer vision;
the face gesture category of the collected user is as followsWhere n is denoted as the image capturing apparatus number, t is denoted as the time at which the user stands at a certain position and focuses on a certain specific product object, the face pose category ∈10 photographed by the image capturing apparatus n>Is determined, face gesture category +.>Attention target M at t t And face position->Constraint, observe likelihood for face by the following formula +.>And (3) performing calculation:
wherein,expressed as face observation variable, face observation likelihood +.>Expressed as face observation variable +.>By face gesture category->The probability of generation, δ, is expressed as a normalization factor, C γ Image subspace expressed as human face gesture category, subspace refers to dimension less than or equal to fullPart of the space is projected to realize the mapping of the high-dimensional features to the low-dimensional space, < ->Expressed as the distance of the face image to the image subspace, such as reconstruction errors and the like when the image is projected to the sub-control, alpha 2 Represented as variance;
when a customer looks at a product, it is possible to continue to watch the object and possibly to shift to another object, so that it is necessary to enhance the smoothness of the time, and the probability matrix P (M t |M t-1 ) And (3) performing calculation:
wherein M is t Representing the attention target of the user at time t, M t-1 Representing the attention objective of the user at time t-1, P f Expressed as the probability that adjacent moments are the same target, this value can be set by the relevant personnel themselves, usually P f Has a value of approximately 1, M Total (S) Expressed as a common target quantity of product;
the joint probability distribution P is calculated by the following formula:
wherein T is the total time of data acquisition, N is the total number of the image capturing devices,expressed as +.>Where the attention target is M t The face posture category is +.>The probability of (2) is calculated by a counting statistical method, and a user can watch various products when staying at one position, so that various face gestures exist; numbering the users, and setting the duration time of the joint distribution probability as t Holding device The numerical values are ordered from big to small to form a set
The higher the need for services by the relevant personnel.
S3, planning an optimal path between the related personnel and the user according to the collected image information and the analyzed degree of urgency of the user for requiring the related personnel to perform service;
according to the analyzed urgent degree of the requirement that the user needs related personnel to perform service, although the user is in a dynamic state, the user cannot move in a large range when paying attention to commodities, so that the user can be regarded as a static state, and the optimal path planning is performed in real time;
the image information of the user is collected in real time, the position of the user is placed in a plane coordinate system which can be set by the related personnel, for example, a plane rectangular coordinate system is established according to the actual topography of the place, the position of the related personnel is used as a starting point, and the position coordinate is (x) i ，y i ) The position of the user is the end point, and the position coordinates are (x j ，y j ) The position coordinates of the adjacent node f of the position of the relevant person are (x k ，y k ) Form the set r= { (x) 1 ，y 1 )，(x 2 ，y 2 )，…，(x m ，y m ) Where m is the number of adjacent nodes, the adjacent nodes represent positions, such as corner positions, where the relevant person passes when traveling to the user position, and the Manhattan distance d between the user position and the adjacent node position is calculated by the following formula 1 (f) And (3) performing calculation:
d 1 (f)＝(x j -x k )+(y j -y k )；
planning a path from the related personnel position to the user position through the following formula:
d(f)＝d 1 (f)+d 2 (f)；
wherein d (f) is the estimated distance from the relevant person position to the user position after passing the adjacent node position, d 2 (f) Expressed as the actual distance from the relevant person's location to the location of the neighboring node, the other neighboring nodes calculate the same, then the optimal path d=d (f) min ；
Taking the user z as a center point, setting the radius as r, constructing Gao Weiqiu area Y, and constructing a high-dimensional space is a physical theory. According to the theory of M proposed in the 90 s, the universe is eleven-dimensional and consists of a vibrating plane; the number of users included in the area Y of Gao Weiqiu is S, and the offset average S is calculated by the following formula:
where z' represents a point contained within region Y of Gao Weiqiu, moving the center point toward the vector direction of the offset mean value, then: z t+1 ＝S t +z t The method comprises the steps of carrying out a first treatment on the surface of the Wherein z is t Represents the center of time t, z t+1 Represents the center of time t+1, S t Representing the offset average value at the time t; the weight ω is calculated by the following formula:
wherein, h (·) is expressed as a negative of the inverse of a kernel function, the kernel function is that the support vector machine maps the input space to the high-dimensional feature space through a certain nonlinear transformation, the kernel function comprises a linear kernel function, a polynomial kernel function, a gaussian kernel function and the like, and θ is expressed as a weight coefficient. The average value shift classification is carried out on the users, the region division is carried out according to each divided type, and related personnel serve the users in the corresponding region, so that the service quality is effectively improved, and the situations that the users adjacent to each other in urgent degree are too far apart to be completely served or the number difference of the users in the region is large, the related personnel are ill-served and the like are avoided; in the area, according to the analyzed demand urgency degree of the users for carrying out service by related personnel, among two users with adjacent demand urgency degrees, the user with high demand urgency degree is a first service grade, the position is a starting point, the user with low demand urgency degree is a second service grade, the position is an ending point, and calculation is carried out, so that a complete optimal path is obtained; when the demand of the user is reduced, the situation that the user possibly has waiting for impatience is indicated, the user is marked, the service level of the user is improved, the position of the related person is taken as a starting point, and the position of the user is taken as an end point to conduct path planning.
And S4, displaying the analysis result to related personnel through a screen display device, and carrying out auxiliary prompt.
According to the analysis result, the optimal path is displayed to related personnel through screen display equipment, such as mobile phones or computers, and the duration of joint distribution probability of different users is highlighted through different colors, such as displaying the urgent degree of the user's demand through red, and auxiliary prompt is performed, such as vibration or sound, so that even if the related personnel fail to view the display equipment in time, prompt can be received quickly, impatience caused by other users due to long time, management of customer relations is guaranteed, and work service efficiency of the related personnel is improved.
Embodiment one:
if the position coordinates of the related person are (1, 1), the position coordinates of the user are (8, 6), if the adjacent node f exists 1 Is (2, 1), f 2 The coordinates of (1, 4), f 3 Is (3, 2), f 4 Coordinates (5, 4), then:
d 1 (f 1 )＝(x j -x k )+(y j -y k )＝11；
d 1 (f 2 )＝(x j -x k )+(y j -y k )＝9；
d 1 (f 3 )＝(x j -x k )+(y j -y k )＝9；
d 1 (f 4 )＝(x j -x k )+(y j -y k )＝5；
if d 2 (f 1 )＝1，d 2 (f 2 )＝4，d 2 (f 3 )＝5，d 2 (f 4 ) =11, then:
d(f 1 )＝d 1 (f 1 )+d 2 (f 1 )＝12；
d(f 2 )＝d 1 (f 2 )+d 2 (f 2 )＝13；
d(f 3 )＝d 1 (f 3 )+d 2 (f 3 )＝14；
d(f 4 )＝d 1 (f 4 )+d 2 (f 4 )＝16；
at this time, d (f 1 )<d(f 2 )<d(f 3 )<d(f 4 ) At this time, a neighboring node f is selected 1 And (5) the path of the model (3) is the optimal path, and displaying is carried out.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: the foregoing description is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art, although the present invention has been described in detail with reference to the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A big data analysis system based on CRM, characterized in that: the big data analysis system includes: the system comprises a data acquisition module, a database, a data analysis module and a terminal feedback module;
the data acquisition module is connected with the database, the database is connected with the data analysis module, and the data analysis module is connected with the terminal feedback module; the system comprises a data acquisition module, a data analysis module and a terminal feedback module, wherein the data acquisition module is used for acquiring basic data information, acquiring user image information through a camera device, the database is used for storing acquired data information and analysis results, the data analysis module is used for analyzing shopping states of users and optimally planning paths between related personnel and the users, and the terminal feedback module is used for displaying and reminding the related personnel through the terminal device according to the analysis results.
2. A CRM-based big data analysis system as set forth in claim 1, wherein: the data acquisition module comprises a basic data input unit and an image acquisition unit, wherein the basic data input unit is used for inputting basic data information of public service places, and the image acquisition unit is used for acquiring image information of users through the camera equipment.
3. A CRM-based big data analysis system as set forth in claim 2, wherein: the database comprises a data storage unit, a data encryption unit and a data cleaning unit, wherein the data storage unit stores data through NoSQL, the data encryption unit encrypts the data through an RC4 algorithm, and the data cleaning unit is used for automatically cleaning the data after a user leaves the public service place.
4. A CRM-based big data analysis system as set forth in claim 3, wherein: the data analysis module comprises a user state analysis unit and a path planning unit, wherein the user state analysis unit is used for analyzing the shopping state of a user according to collected user image information and knowing the urgency degree of the user for needing related personnel to carry out service, and the path planning unit is used for planning the optimal path between related personnel and the user according to the analyzed urgency degree of the user for needing related personnel to carry out service.
5. The CRM-based big data analysis system of claim 4, wherein: the terminal feedback module comprises a screen display unit and an auxiliary prompt unit, wherein the screen display unit displays real-time paths of related personnel through screen display equipment, and the auxiliary prompt unit is used for carrying out auxiliary prompt on the related personnel in an additional mode.
6. A big data analysis method based on CRM is characterized in that: the method comprises the following steps:
s1, acquiring basic data information, acquiring user image information through camera equipment, and encrypting and storing;
s2, analyzing the shopping state of the user according to the acquired image information to obtain the urgent degree of the requirement that the user needs related personnel to perform service;
s3, planning an optimal path between the related personnel and the user according to the collected image information and the analyzed degree of urgency of the user for requiring the related personnel to perform service;
and S4, displaying the analysis result to related personnel through a screen display device, and carrying out auxiliary prompt.
7. The CRM-based big data analysis method of claim 6, wherein: in step S2, facial image information of a user is acquired through an image capturing device, face tracking is performed on the user through an OpenCV technology, and a distance is measured through an image ranging technology;
the face gesture category of the collected user is as followsWhere n is denoted as the image pickup apparatus number, t is denoted as the time at which the face is observed with the likelihood +.>And (3) performing calculation:
wherein,expressed as face observation variable, delta expressed as normalization factor, C γ Image subspace representing a face pose class, +.>Expressed as the distance alpha of the face image to the subspace of the image 2 Represented as variance;
a transition probability matrix P (M) between different attention objects at adjacent moments of the user is calculated by the following formula t |M t-1 ) And (3) performing calculation:
wherein M is t Representing the attention target of the user at time t, M t-1 Representing the attention objective of the user at time t-1, P f Expressed as probability of adjacent time being the same target, M Total (S) Expressed as a common target quantity of product;
the joint probability distribution P is calculated by the following formula:
wherein T is the total time of data acquisition, N is the total number of the image capturing devices,represented as being in a face positionWhere the attention target is M t The face posture category is +.>The probability of (2) is calculated by a counting statistical method; numbering the users, and setting the duration time of the joint distribution probability as t Holding device The values are ordered from big to small to form a set r= { t Hold 1 ，t Hold 2 ，…，t Hold m Where m is denoted as user number, t Holding device The larger the value of (c) is, the higher the corresponding user needs related personnel to perform service.
8. The CRM-based big data analysis method of claim 7, wherein: in step S3, according to the analyzed urgent degree of the service requirement of the user for the related personnel, carrying out optimal path planning in real time;
collecting image information of a user in real time, placing the position of the user in a plane coordinate system, wherein the position of related personnel is taken as a starting point, and the position coordinate is (x i ，y i ) The position of the user z is the end point, and the position coordinates are (x j ，y j ) The position coordinates of the adjacent node f of the position of the relevant person are (x k ，y k ) Form the set r= { (x) 1 ，y 1 )，(x 2 ，y 2 )，…，(x m ，y m ) }, itM is the number of adjacent nodes, and Manhattan distance d between the user position and the adjacent node position is calculated by the following formula 1 (f) And (3) performing calculation:
d 1 (f)＝(x j -x k )+(y j -y k )；
planning a path from the related personnel position to the user position through the following formula:
d(f)＝d 1 (f)+d 2 (f)；
wherein d (f) is the estimated distance from the relevant person position to the user position after passing the adjacent node position, d 2 (f) Expressed as the actual distance from the relevant person's location to the location of the neighboring node, the other neighboring nodes calculate the same, then the optimal path d=d (f) min ；
Constructing Gao Weiqiu area Y by taking user z as a center point and setting the radius as r; the number of users included in the area Y of Gao Weiqiu is S, and the offset average S is calculated by the following formula:
where z' represents a point contained within region Y of Gao Weiqiu, moving the center point toward the vector direction of the offset mean value, then: z t+1 ＝S t +z t The method comprises the steps of carrying out a first treatment on the surface of the Wherein z is t Represents the center of time t, z t+1 Represents the center of time t+1, S t Representing the offset average value at the time t; the weight ω is calculated by the following formula:
wherein, h (·) is expressed as a negative of the inverse of a kernel function, the kernel function is that the support vector machine maps the input space to the high-dimensional feature space through a certain nonlinear transformation, the kernel function comprises a linear kernel function, a polynomial kernel function, a gaussian kernel function and the like, and θ is expressed as a weight coefficient. Performing multiple operations, namely performing mean shift classification on users, and performing region division according to each classified type, wherein related personnel serve the users in the corresponding region; in the area, according to the analyzed demand urgency degree of the users for carrying out service by related personnel, among two users with adjacent demand urgency degrees, the user with high demand urgency degree is a first service grade, the position is a starting point, the user with low demand urgency degree is a second service grade, the position is an ending point, and calculation is carried out, so that a complete optimal path is obtained; when the urgent degree of the user demand is reduced, labeling the user, improving the service level of the user, and planning a path by taking the position of the related person as a starting point and the position of the user as an ending point.
9. The CRM-based big data analysis method of claim 8, wherein: in step S4, according to the result of the analysis, the optimal path is displayed to the related personnel through the screen display device, and the duration of the joint distribution probability of different users is highlighted through different colors, and meanwhile, the auxiliary prompt is performed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211715862.3A CN117114694B (en) | 2022-12-29 | 2022-12-29 | Big data analysis system and method based on CRM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211715862.3A CN117114694B (en) | 2022-12-29 | 2022-12-29 | Big data analysis system and method based on CRM |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117114694A true CN117114694A (en) | 2023-11-24 |
| CN117114694B CN117114694B (en) | 2024-05-07 |
Family
ID=88809870
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211715862.3A Active CN117114694B (en) | 2022-12-29 | 2022-12-29 | Big data analysis system and method based on CRM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117114694B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106921945A (en) * | 2017-02-22 | 2017-07-04 | 上海斐讯数据通信技术有限公司 | The positioning shopping guide method and system of a kind of commodity |
| US20170220943A1 (en) * | 2014-09-30 | 2017-08-03 | Mentorica Technology Pte Ltd | Systems and methods for automated data analysis and customer relationship management |
| CN111476613A (en) * | 2020-05-09 | 2020-07-31 | 苏州中仑网络科技有限公司 | Shopping guide auxiliary method and device based on passenger flow analysis, server and storage medium |
| CN111597999A (en) * | 2020-05-18 | 2020-08-28 | 常州工业职业技术学院 | A 4S store sales service management method and system based on video detection |
| CN112418994A (en) * | 2020-11-26 | 2021-02-26 | 珠海格力电器股份有限公司 | Commodity shopping guide method and device, electronic equipment and storage medium |
| CN113887884A (en) * | 2021-09-13 | 2022-01-04 | 青岛颐中科技有限公司 | Supermarket service system |
-
2022
- 2022-12-29 CN CN202211715862.3A patent/CN117114694B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170220943A1 (en) * | 2014-09-30 | 2017-08-03 | Mentorica Technology Pte Ltd | Systems and methods for automated data analysis and customer relationship management |
| CN106921945A (en) * | 2017-02-22 | 2017-07-04 | 上海斐讯数据通信技术有限公司 | The positioning shopping guide method and system of a kind of commodity |
| CN111476613A (en) * | 2020-05-09 | 2020-07-31 | 苏州中仑网络科技有限公司 | Shopping guide auxiliary method and device based on passenger flow analysis, server and storage medium |
| CN111597999A (en) * | 2020-05-18 | 2020-08-28 | 常州工业职业技术学院 | A 4S store sales service management method and system based on video detection |
| CN112418994A (en) * | 2020-11-26 | 2021-02-26 | 珠海格力电器股份有限公司 | Commodity shopping guide method and device, electronic equipment and storage medium |
| CN113887884A (en) * | 2021-09-13 | 2022-01-04 | 青岛颐中科技有限公司 | Supermarket service system |
Non-Patent Citations (1)
| Title |
|---|
| 董力赓: "一种基于动态贝叶斯网的视觉注意力识别方法", 电子学报, vol. 39, no. 3, pages 140 - 145 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117114694B (en) | 2024-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12026665B2 (en) | Identifying inventory items using multiple confidence levels | |
| TWI773797B (en) | System, method and computer program product for tracking multi-joint subjects in an area of real space | |
| US10133933B1 (en) | Item put and take detection using image recognition | |
| US10127438B1 (en) | Predicting inventory events using semantic diffing | |
| US10055853B1 (en) | Subject identification and tracking using image recognition | |
| JP7228670B2 (en) | Real-time inventory tracking using deep learning | |
| US20040036712A1 (en) | Queue management system and method | |
| CA3056868C (en) | Media content tracking | |
| WO2021192190A1 (en) | Person flow prediction system, person flow prediction method, and program recording medium | |
| Ravnik et al. | Interactive and audience adaptive digital signage using real-time computer vision | |
| WO2021042541A1 (en) | Shopping guide method and apparatus in new retail model, device and storage medium | |
| CN117114694B (en) | Big data analysis system and method based on CRM | |
| JP2020067720A (en) | Human attribute estimation system, information processing apparatus and information processing method using the same | |
| CN110458810A (en) | A kind of more classification and Detection method and devices of skin quality based on Face datection | |
| CN112989988B (en) | Information integration method, device, equipment, readable storage medium and program product | |
| Oztaysi et al. | Indoor location tracking technology evaluation by using spherial fuzzy TOPSIS method | |
| Yellanki | Adaptive Infrastructure for Real-Time Consumer Behavior Monitoring in Retail Environments | |
| Collorone et al. | ANTHROPOS-V: benchmarking the novel task of Crowd Volume Estimation | |
| CN119130606A (en) | Intelligent shopping cart system based on deep reinforcement learning and its implementation method | |
| CN111145288A (en) | Target customer virtual imaging method | |
| WO2024257422A1 (en) | Image analysis device, image analysis method, and image analysis program | |
| CN116663951A (en) | User retention data analysis model construction and analysis method, device and equipment | |
| CN111078904A (en) | Cartoon product display classification method based on intelligent product picture album |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240408 Address after: No. 1602, 16th Floor, Unit 2, No. 22, Business Outer Ring Road, Zhengzhou Area (Zhengdong), Henan Pilot Free Trade Zone, Zhengzhou City, Henan Province, 450046 Applicant after: Henan CITIC Big Data Technology Co.,Ltd. Country or region after: China Address before: Building 1, Building E301-65, No. 1 Hagongda Road, Tangjiawan Town, High tech Zone, Zhuhai City, Guangdong Province, 519000 Applicant before: Zhuhai Deep Blue Network Technology Co.,Ltd. Country or region before: China |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |