CN116431911A - Business collaborative management method and system based on intelligent matching - Google Patents
Business collaborative management method and system based on intelligent matching Download PDFInfo
- Publication number
- CN116431911A CN116431911A CN202310337967.8A CN202310337967A CN116431911A CN 116431911 A CN116431911 A CN 116431911A CN 202310337967 A CN202310337967 A CN 202310337967A CN 116431911 A CN116431911 A CN 116431911A
- Authority
- CN
- China
- Prior art keywords
- user
- project
- item
- information
- input vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 24
- 239000013598 vector Substances 0.000 claims abstract description 73
- 238000003062 neural network model Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 14
- 238000005259 measurement Methods 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 21
- 230000006399 behavior Effects 0.000 claims description 20
- 238000000034 method Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 18
- 238000010586 diagram Methods 0.000 claims description 14
- 238000010276 construction Methods 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 6
- 238000005295 random walk Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000005065 mining Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 239000000872 buffer Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- 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
- G06F16/903—Querying
- G06F16/9035—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a business collaborative management method and a system based on intelligent matching, comprising the following steps: generating an input vector; inputting the input vector to a deep neural network model, and obtaining user preference feature representation and project feature representation by minimizing a cross entropy loss function; and searching the most similar preset number of items as recommendation results according to the inner product measurement by using the user preference characteristic representation and the item characteristic representation through a nearest neighbor algorithm. The business collaborative management method and system based on intelligent matching uses a deep feature embedded network to learn heterogeneous features, so that potential features between users and projects can be fully mined, potential requirements of information publishers and information consumers can be accurately acquired, and the requirements of the two parties can be matched, so that the success rate of information matching is improved.
Description
Technical Field
The invention relates to the technical field of business collaboration, in particular to a business collaboration management method and system based on intelligent matching.
Background
With the advent of the information age, information management has been increasingly emphasized, and business collaboration between internal enterprises, sub-companies and external enterprises has become increasingly important. Currently, although personalized recommendations have found widespread use in many areas, there are relatively few applications in the collaborative area within an enterprise. For large enterprises, the internal cooperation demands of the enterprises are more frequent, and the types of the cooperation demands are more complex, so that a user can hardly find useful information from massive information, and the efficiency of business cooperation is reduced. Meanwhile, the existing recommendation system has the problem of cold start, so that the performance of a recommendation algorithm is not high, and therefore, how to effectively recommend a newly-warehoused item to a user who likes the item is an urgent problem to be solved by the recommendation algorithm.
Disclosure of Invention
In order to solve the defects of the prior art, the invention provides a business collaborative management method and a business collaborative management system based on intelligent matching, which use a deep feature embedded network to learn heterogeneous features, fully mine potential features between users and projects, accurately acquire potential requirements of an information publisher and an information consumer, and match the requirements of the two parties so as to improve the success rate of information matching.
In order to achieve the above object, the present invention adopts the technical scheme that:
a business collaborative management method based on intelligent matching is characterized by comprising the following steps:
s1, generating an input vector;
s2, inputting an input vector into a deep neural network model, and obtaining user preference feature representation and project feature representation by minimizing a cross entropy loss function;
s3, the user preference feature representation and the item feature representation are carried out through a nearest neighbor algorithm, and the most similar preset number of items are searched to be used as recommendation results according to the inner product measurement.
Further, the step S1 includes:
s11, acquiring browsing record information of a user item;
s12, according to the user project browsing record information, mining the connection relation between the projects, and generating a user behavior construction diagram;
s13, using a random walk algorithm to walk each project node in the user behavior construction diagram to generate a project node sequence;
s14, inputting the project node sequence into a Skip-Gram model, and performing Word2vec training to obtain vector representation of each project node;
s15, splicing vectors of a plurality of project nodes to obtain an initial input vector;
s16, inputting the initial input vector into the self-attention model, and calculating weight coefficients of a plurality of project node vectors to obtain the input vector.
Further, the step S13 includes:
and starting from randomly selecting one project node in the graph, randomly selecting one project node from adjacent project nodes of the project node as the next project node until all project nodes are traversed, and generating a project node sequence.
Further, the user item browsing record information includes: user history browsing information, user participation project behavior information, user operation log information, user basic information and project characteristic information.
Further, the generating a user behavior building diagram according to the browsing record information of the user items and the connection relation between the items comprises:
setting a time window, taking each item as a node, connecting a plurality of item nodes through a directed edge, distributing weights for the occurrence total number of the item connection of each edge based on user behaviors, and the weights are the frequency of item i to item j.
Further, the activation function of the deep neural network model is a ReLU function.
Further, the deep neural network model predicts the probability of a user participating in item i through a Softmax activation function:
wherein U represents a user embedding vector, the user embedding vector is preference information of a user, vi represents an embedding vector of an item i, the embedding vector of the item is a weight from a ReLU layer to a Softmax output layer, and U represents a user set.
The invention also relates to a business collaborative management system based on intelligent matching, which is characterized by comprising the following steps:
an input vector generation module for generating an input vector;
the deep neural network model module is used for inputting the input vector into the deep neural network model, and obtaining user preference characteristic representation and project characteristic representation by minimizing a cross entropy loss function;
and the recommendation result searching module is used for searching the most similar preset number of items as recommendation results according to the inner product measurement by using the user preference characteristic representation and the item characteristic representation through a nearest neighbor algorithm.
The invention also relates to a computer readable storage medium, which is characterized in that the storage medium is stored with a computer program, and the computer program realizes the business collaborative management method based on intelligent matching when being executed by a processor.
The invention also relates to an electronic device, which is characterized by comprising a processor and a memory;
the memory is used for storing the deep neural network model;
the processor is used for executing the business collaborative management method based on intelligent matching by calling the deep neural network model.
The invention also relates to a computer program product comprising a computer program and/or instructions, characterized in that the computer program and/or instructions, when executed by a processor, implement the steps of the business collaborative management method based on intelligent matching.
The beneficial effects of the invention are as follows:
by adopting the business collaborative management method and system based on intelligent matching, the method and system use the deep feature embedded network to learn heterogeneous features, so that potential features between users and projects can be fully mined, potential requirements of information publishers and information consumers can be accurately acquired, and the requirements of the information publishers and the information consumers can be matched, so that the success rate of information matching is improved. The method and the system realize matching and matching of the resource demand party and the resource supply party, solve the problems that the information of the information issuing party and the information consuming party are asymmetric and effective information is difficult to acquire, solve the cold start problem of the recommendation system, improve the recommendation performance and user experience, and effectively improve the accuracy of the matching and matching business collaborative management system. Meanwhile, the service matching system combined with the deep neural network technology can mine hidden and potential characteristics between users and projects in the service matching system, solve the problem of low information butt joint efficiency in the existing matching service system, and can improve the performance of the matching system.
Drawings
Fig. 1 is a schematic flow chart of a business collaborative management method based on intelligent matching.
Fig. 2 is a schematic diagram of a business collaborative management system structure based on intelligent matching.
Detailed Description
For a clearer understanding of the present invention, reference will be made to the following detailed description taken in conjunction with the accompanying drawings and examples.
The first aspect of the present invention relates to a method with a step flow as shown in fig. 1, comprising:
heterogeneous feature information of graph embedded layer generation model
S1, generating an input vector;
s11, acquiring browsing record information of a user item;
the user item browsing record information includes: user history browsing information, user participation project behavior information, user operation log information, user basic information and project characteristic information.
S12, according to the user project browsing record information, mining the connection relation between the projects, and generating a user behavior construction diagram;
based on the assumption that the user has similarity with the browsed items in the past period of time, the hidden connection relationship between the items can be mined from the user item browsing record according to the assumption.
Here a time window is provided, connected by a directed edge, for example: item D and item a are connected because a user accesses items D and a sequentially within the window. By utilizing the collaborative behavior of all users, each edge is assigned a weight based on the total number of occurrences of all user behavior project connections. Specifically, in all user behavior histories, the weight of the edge is equal to the frequency with which item i goes to item j.
S13, using a random walk algorithm to walk each project node in the user behavior construction diagram to generate a project node sequence;
and starting from randomly selecting one project node in the graph, randomly selecting one project node from adjacent project nodes of the project node as the next project node until all project nodes are traversed, and generating a project node sequence.
S14, inputting the project node sequence into a Skip-Gram model, and performing Word2vec training to obtain vector representation of each project node;
the node sequence obtained by random walk is analogically to corpus in natural language, word2vec training is carried out on the corpus by using Skip-Gram, the vector representation of the node in the graph can be obtained, and the vector can capture the structural information of the node.
S15, splicing vectors of a plurality of project nodes to obtain an initial input vector;
according to the Word2vec algorithm, vector representation of project nodes is obtained, and a user history browsing project vector and a participating project vector can be sequentially obtained, and meanwhile, operation logs of users, user basic information and project characteristic information are combined to serve as input layer characteristics of the deep neural network model.
Firstly, selecting relevant characteristics of a user history participation item, and averaging embedded vectors of the item recently participated by the user to obtain comprehensive embedded representation of the user history participation item behavior; then, the item list recently browsed by the user is selected as well, and the comprehensive browsing behavior embedded vector of the user is obtained; finally, splice with other features of the user (basic information and operation log, etc.), wherein here 20 items are selected that the user has recently participated in and browsed.
For some newly-put items, few initial users and few user behaviors exist, so that characteristic information (basic conditions of the items, expected scale of the items, expected income of the items and the like) of the items is introduced in the characteristic representation process, and the newly-put items can be recommended to users who like the newly-put items, thereby solving the problem of cold start.
And splicing the features such as the browsing project vector, the participation project vector, the user operation log, the user basic information, the project characteristic information and the like to finally obtain an initial input vector x.
S16, inputting the initial input vector into the self-attention model, and calculating weight coefficients of a plurality of project node vectors to obtain the input vector.
This does not reflect the actual situation, provided that the different characteristic information contributes equally to the final embedding. The heterogeneous feature information is aggregated here based on learning the weight coefficient of each attribute information from the attention module layer in consideration of different contributions of different kinds of feature information to user behavior. For each initial input vector xi, feature information zi is obtained by the self-attention module, which is weighted and aggregated based on the self-attention mechanism, the zi vector being obtained by weighted summation of the input vector xi over its neighboring vectors, wherein the weights are determined by the relation between the items. The vector zi corresponding to each input vector xi represents both the information of the current i-th user and the relationship between the i-th user and other surrounding users.
The self-attention model calculates importance weights of different characteristic information by adopting a dot product operation mode, and the calculation formula is as follows:
wherein Q, K, V are a query vector matrix, a key vector matrix, and a value vector matrix, respectively. Each row of the three matrices Q, K, V represents a corresponding vector, typically by multiplying the input vector sequence X by the three matrices W, respectively q ，W k ，W v Obtained.
S2, inputting an input vector into a deep neural network model, and obtaining user preference feature representation and project feature representation by minimizing a cross entropy loss function;
the application uses deep learning to model, embedding users and items into the same low-dimensional space. The characteristic representation z obtained by the self-attention module is input into a Deep Neural Network (DNN), the network structure comprises three MLP full-connection layers, wherein an activation function is a ReLU function, the output of the last hidden layer of the network is a characteristic vector of a user, the characteristic vector represents information such as preference of the user, and the like, and the characteristic vector represents an embedded vector of the user. The embedding vector of an item is the weight of the ReLU layer to the Softmax output layer, and its dimension is the same as the embedding vector of the user.
The project recommendation problem is converted into a multi-classification problem, and the output of the output layer is the probability of the user participating in each project. The positive sample is a project in which the user participates; the negative sample is that the user is not engaged in the project.
Predicting the probability of a user to participate in item i by Softmax activation function:
where U and vi represent the embedded vectors of the user and item i, respectively, and U is the user set.
S3, the user preference feature representation and the item feature representation are carried out through a nearest neighbor algorithm, and the most similar preset number of items are searched to be used as recommendation results according to the inner product measurement.
The deep neural network architecture can learn the embedded vectors of the user and the project at one time, and can obtain model parameters by minimizing a Cross-entropy loss function (Cross-entropy loss), so that the characteristic embedding of the user and the project is finally obtained. For users and items, the feature embedding of the output layer is in the same space, and the inner product of the two can represent similarity. The most similar Top N is found as the candidate set according to the inner product metric.
For online services, there are stringent performance requirements and results must be returned within tens of milliseconds. Thus, by preserving user feature embedding and item feature embedding, for each user vector u, for all items v in the set of items i The nearest neighbor algorithm is performed, and the item of the most similar Top N is found according to the inner product measurement to be used as a recommendation result
The invention also relates to a business collaborative management system based on intelligent matching, the structure of which is shown in figure 2, comprising:
an input vector generation module for generating an input vector;
the deep neural network model module is used for inputting the input vector into the deep neural network model, and obtaining user preference characteristic representation and project characteristic representation by minimizing a cross entropy loss function;
and the recommendation result searching module is used for searching the most similar preset number of items as recommendation results according to the inner product measurement by using the user preference characteristic representation and the item characteristic representation through a nearest neighbor algorithm.
By using the system, the above-mentioned operation processing method can be executed and the corresponding technical effects can be achieved.
The embodiments of the present invention also provide a computer-readable storage medium capable of implementing all the steps of the method in the above embodiments, the computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements all the steps of the method in the above embodiments.
The embodiment of the invention also provides an electronic device for executing the method, which is used as an implementation device of the method, and at least comprises a processor and a memory, wherein the memory is particularly used for storing data and related computer programs required by executing the method, such as a deep neural network model and the like, and all the steps of the implementation method are executed by calling the data and the programs in the memory by the processor, so that corresponding technical effects are obtained.
Preferably, the electronic device may comprise a bus architecture, and the bus may comprise any number of interconnected buses and bridges, the buses linking together various circuits, including the one or more processors and memory. The bus may also link together various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., as are well known in the art and, therefore, will not be further described herein. The bus interface provides an interface between the bus and the receiver and transmitter. The receiver and the transmitter may be the same element, i.e. a transceiver, providing a unit for communicating with various other systems over a transmission medium. The processor is responsible for managing the bus and general processing, while the memory may be used to store data used by the processor in performing operations.
Additionally, the electronic device may further include a communication module, an input unit, an audio processor, a display, a power supply, and the like. The processor (or controllers, operational controls) employed may comprise a microprocessor or other processor device and/or logic devices that receives inputs and controls the operation of the various components of the electronic device; the memory may be one or more of a buffer, a flash memory, a hard drive, a removable medium, a volatile memory, a nonvolatile memory, or other suitable means, may store the above-mentioned related data information, may further store a program for executing the related information, and the processor may execute the program stored in the memory to realize information storage or processing, etc.; the input unit is used for providing input to the processor, and can be a key or a touch input device; the power supply is used for providing power for the electronic equipment; the display is used for displaying display objects such as images and characters, and may be, for example, an LCD display. The communication module is a transmitter/receiver that transmits and receives signals via an antenna. The communication module (transmitter/receiver) is coupled to the processor to provide an input signal and to receive an output signal, which may be the same as in the case of a conventional mobile communication terminal. Based on different communication technologies, a plurality of communication modules, such as a cellular network module, a bluetooth module, and/or a wireless local area network module, etc., may be provided in the same electronic device. The communication module (transmitter/receiver) is also coupled to the speaker and microphone via the audio processor to provide audio output via the speaker and to receive audio input from the microphone to implement the usual telecommunications functions. The audio processor may include any suitable buffers, decoders, amplifiers and so forth. In addition, the audio processor is also coupled to the central processor so that sound can be recorded on the host through the microphone and sound stored on the host can be played through the speaker.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
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 flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations 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 a system 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 stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function 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. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present invention should be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.
Claims (11)
1. A business collaborative management method based on intelligent matching is characterized by comprising the following steps:
s1, generating an input vector;
s2, inputting an input vector into a deep neural network model, and obtaining user preference feature representation and project feature representation by minimizing a cross entropy loss function;
s3, the user preference feature representation and the item feature representation are carried out through a nearest neighbor algorithm, and the most similar preset number of items are searched to be used as recommendation results according to the inner product measurement.
2. The method according to claim 1, wherein the step S1 includes:
s11, acquiring browsing record information of a user item;
s12, according to the user project browsing record information, mining the connection relation between the projects, and generating a user behavior construction diagram;
s13, using a random walk algorithm to walk each project node in the user behavior construction diagram to generate a project node sequence;
s14, inputting the project node sequence into a Skip-Gram model, and performing Word2vec training to obtain vector representation of each project node;
s15, splicing vectors of a plurality of project nodes to obtain an initial input vector;
s16, inputting the initial input vector into the self-attention model, and calculating weight coefficients of a plurality of project node vectors to obtain the input vector.
3. The method according to claim 1, wherein the step S13 includes:
and starting from randomly selecting one project node in the graph, randomly selecting one project node from adjacent project nodes of the project node as the next project node until all project nodes are traversed, and generating a project node sequence.
4. The method of claim 1, wherein the user item browsing record information comprises: user history browsing information, user participation project behavior information, user operation log information, user basic information and project characteristic information.
5. The method of claim 1, wherein the generating a user behavior construction graph according to the user item browsing record information and the connection relation between the items comprises:
setting a time window, taking each item as a node, connecting a plurality of item nodes through a directed edge, distributing weights for the occurrence total number of the item connection of each edge based on user behaviors, and the weights are the frequency of item i to item j.
6. The method of claim 1, wherein the activation function of the deep neural network model is a ReLU function.
7. The method of claim 1, wherein the deep neural network model predicts the probability of a user engaging in item i by a Softmax activation function:
wherein U represents a user embedding vector, the user embedding vector is preference information of a user, vi represents an embedding vector of an item i, the embedding vector of the item is a weight from a ReLU layer to a Softmax output layer, and U represents a user set.
8. The business collaborative management system based on intelligent matching is characterized by comprising:
an input vector generation module for generating an input vector;
the deep neural network model module is used for inputting the input vector into the deep neural network model, and obtaining user preference characteristic representation and project characteristic representation by minimizing a cross entropy loss function;
and the recommendation result searching module is used for searching the most similar preset number of items as recommendation results according to the inner product measurement by using the user preference characteristic representation and the item characteristic representation through a nearest neighbor algorithm.
9. A computer readable storage medium, wherein a computer program is stored on the storage medium, and when executed by a processor, the computer program implements the business collaborative management method based on intelligent matchmaking according to any one of claims 1 to 7.
10. An electronic device comprising a processor and a memory;
the memory is used for storing the deep neural network model;
the processor is configured to execute the business collaborative management method based on intelligent matching according to any one of claims 1 to 7 by calling a deep neural network model.
11. A computer program product comprising computer programs and/or instructions which, when executed by a processor, implement the steps of the business collaborative management method based on intelligent matchmaking according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310337967.8A CN116431911A (en) | 2023-03-31 | 2023-03-31 | Business collaborative management method and system based on intelligent matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310337967.8A CN116431911A (en) | 2023-03-31 | 2023-03-31 | Business collaborative management method and system based on intelligent matching |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116431911A true CN116431911A (en) | 2023-07-14 |
Family
ID=87086587
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310337967.8A Pending CN116431911A (en) | 2023-03-31 | 2023-03-31 | Business collaborative management method and system based on intelligent matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116431911A (en) |
-
2023
- 2023-03-31 CN CN202310337967.8A patent/CN116431911A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220365939A1 (en) | Methods and systems for client side search ranking improvements | |
| CN111708876B (en) | Method and device for generating information | |
| US7664671B2 (en) | Methods and systems for profile-based forecasting with dynamic profile selection | |
| WO2014193399A1 (en) | Influence score of a brand | |
| CN111506820A (en) | Recommendation model, method, device, equipment and storage medium | |
| US20180067991A1 (en) | Using Structured Smart Digital Memory to Personalize Digital Agent and Bot Scenarios | |
| US20240005356A1 (en) | System for effective use of data for personalization | |
| CN112948885B (en) | Method, device and system for realizing privacy protection of multiparty collaborative update model | |
| CN115841366B (en) | Method and device for training object recommendation model, electronic equipment and storage medium | |
| CN109451334B (en) | User portrait generation processing method and device and electronic equipment | |
| CN113342418A (en) | Distributed machine learning task unloading method based on block chain | |
| CN111652673B (en) | Intelligent recommendation method, device, server and storage medium | |
| CN116431911A (en) | Business collaborative management method and system based on intelligent matching | |
| Qu et al. | Scoop: Automated social recommendation in enterprise process management | |
| US11922287B2 (en) | Video recommendation with multi-gate mixture of experts soft actor critic | |
| CN114840525A (en) | Work order processing method and device | |
| CN117611059B (en) | Inventory management method and system for dynamic adjustment of storage structure | |
| CN111626805B (en) | Information display method and device | |
| Li et al. | Session-based Recommendation with Temporal Graph Neural Network and Contrastive Learning | |
| WO2023166564A1 (en) | Estimation device | |
| CN117540080A (en) | Method and device for recommending articles in cold start mode for strengthening new user characteristics | |
| CN117493666A (en) | Article cold start recommendation method and article cold start recommendation model training method | |
| CN117972222A (en) | Enterprise information retrieval method and device based on artificial intelligence | |
| Izadkhah et al. | Deep Reinforcement Learning Based Group Recommendation System with Multi-Head Attention Mechanism | |
| CN117745356A (en) | Method, apparatus, device and readable medium for call completing rate prediction |
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 |