CN117217763A - Intelligent recommendation method and system for suppliers based on neural network algorithm - Google Patents
Intelligent recommendation method and system for suppliers based on neural network algorithm Download PDFInfo
- Publication number
- CN117217763A CN117217763A CN202311046202.5A CN202311046202A CN117217763A CN 117217763 A CN117217763 A CN 117217763A CN 202311046202 A CN202311046202 A CN 202311046202A CN 117217763 A CN117217763 A CN 117217763A
- Authority
- CN
- China
- Prior art keywords
- data
- neural network
- user
- recommendation
- suppliers
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 63
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 36
- 238000011156 evaluation Methods 0.000 claims abstract description 25
- 230000006399 behavior Effects 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims abstract description 13
- 238000013210 evaluation model Methods 0.000 claims abstract description 9
- 238000011176 pooling Methods 0.000 claims description 31
- 239000013598 vector Substances 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 19
- 238000013527 convolutional neural network Methods 0.000 claims description 14
- 238000009499 grossing Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000007405 data analysis Methods 0.000 claims description 8
- 238000009826 distribution Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000012731 temporal analysis Methods 0.000 claims description 4
- 238000000700 time series analysis Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012300 Sequence Analysis Methods 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims description 2
- 230000003993 interaction Effects 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000003490 calendering Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Abstract
The invention relates to the technical field of neural network algorithms and big data application, in particular to a supplier intelligent recommendation method based on the neural network algorithm, and the invention discloses a supplier intelligent recommendation method and a supplier intelligent recommendation system based on the neural network algorithm, wherein the recommendation method comprises the steps of analyzing the requirement information of a user based on suppliers and user data; carrying out data processing on the requirement information of the user by utilizing a neural network algorithm; building a recommendation training model to complete intelligent recommendation of the suppliers; automatically evaluating the recommended suppliers based on the evaluation model; according to the invention, through analyzing the demand information and the behaviors of the user, the provider matched with the user can be recommended according to the preference and the history interaction of the user, the personalized recommendation result is provided, the demand of the user can be predicted more accurately through a neural network algorithm, more accurate recommendation is provided, the reliability of the recommendation can be known by the user through evaluating the provider, and the workload of self evaluation of the user is reduced.
Description
Technical Field
The invention relates to the technical field of neural network algorithms and big data application, in particular to an intelligent recommendation method for suppliers based on the neural network algorithm.
Background
At present, most enterprise informatization systems have been built and applied for years, a large amount of business and management data are accumulated, and along with the development of emerging technologies such as big data and artificial intelligence, how to apply the emerging technologies, so that data create value, service business and service management are realized, and the comprehensive improvement of enterprise intelligent management is the next goal of most enterprise informatization.
The original supplier source searching mode of most enterprises is evaluated by suppliers under the line of purchasing manager, and then is manually input into the existing informatization system after the evaluation is completed, and the manual selection is performed. Therefore, by applying big data, artificial intelligence and a neural network algorithm, the data value of the purchasing business of the suppliers can be exerted, the invitation efficiency of the suppliers in the purchasing and source-seeking stage is improved, the automatic and accurate recommendation of the high-quality suppliers can be realized, the purchasing transparency is improved, the probability of the risk suppliers for entering the suppliers is reduced, and the labor and operation cost is reduced.
Disclosure of Invention
The present invention has been made in view of the above-described problems.
Therefore, the invention provides the intelligent recommendation method for the suppliers based on the neural network algorithm, which can provide personalized recommendation results according to the user preference and the suppliers matched with the user through analyzing the user demand information and behaviors, can more accurately predict the user demand and the user preference through the neural network algorithm, provides more accurate recommendation, can help the user to know the recommendation reliability through automatically evaluating the suppliers, and reduces the workload of self evaluation of the user.
In order to solve the technical problems, the invention provides the following technical scheme: a provider intelligent recommendation method based on a neural network algorithm comprises the following steps,
analyzing the user's demand information based on the provider and the user data;
carrying out data processing on the requirement information of the user by utilizing a neural network algorithm;
building a recommendation training model to complete intelligent recommendation of the suppliers;
and automatically evaluating the recommended suppliers based on the evaluation model.
As a preferable scheme of the provider intelligent recommendation method based on the neural network algorithm, the invention comprises the following steps: the analysis of the user's demand information based on the provider and user data is obtained by analyzing the provider and user data, including analysis of historical transaction data of the user, time series analysis of user behavior data, and analysis of user preferences based on machine learning.
As a preferable scheme of the provider intelligent recommendation method based on the neural network algorithm, the invention comprises the following steps: the user behavior data is analyzed through a simple exponential smoothing method in the time sequence analysis of the user behavior data, the exponential smoothing method enables the data to be smoother through weighted average of past behavior data, future data change trend is reflected, future behaviors are predicted, and the specific implementation formula is as follows:
wherein,representing the predicted value of time t+1, which is the behavior prediction of the user, y t The representation is the actual observation of time t, < >>The predicted value for time t is represented, and α represents the smoothing coefficient and is 0.78.
As a preferable scheme of the provider intelligent recommendation method based on the neural network algorithm, the invention comprises the following steps: the data processing of the demand information of the user by using the neural network algorithm is to complete the standardization of the data and the duplication removal of the data by using the neural network algorithm, wherein the standardization of the data is to map the data into standard distribution with the mean value of 0 and the standard deviation of 1, so that the data is more uniform in normal distribution, and the specific implementation formula is as follows:
wherein X represents the original data, X' represents the normalized data, mu represents the mean value of the original data X, and sigma represents the standard deviation of the original data.
As a preferable scheme of the provider intelligent recommendation method based on the neural network algorithm, the invention comprises the following steps: the data deduplication is based on a neural network architecture and is realized by measuring similarity measurement among data samples, the neural network architecture is composed of two sub-networks with the same structure, the sub-networks are spaces for processing one data sample, the similarity measurement among the measured data samples is realized by measuring cosine similarity among the original samples, and a specific realization formula is as follows:
wherein A and B respectively represent vector sets of two data samples, a 1 ,a 2 ,...,a n Representing data within vector set A, b 1 ,b 2 ,...,b n Representing data within vector set B;
the implementation rule of the data deduplication is as follows:
when the value of cosine similarity cos (A, B) is 1, the similarity between the two data is completely overlapped, and one data is needed to be removed;
when the value of cosine similarity cos (A, B) is 0, the similarity between the two data is partially overlapped, and one data is needed to be removed;
when the cosine similarity cos (A, B) is-1, it indicates that the similarity between the two data is completely opposite, and the two data should be kept.
As a preferable scheme of the provider intelligent recommendation method based on the neural network algorithm, the invention comprises the following steps: the intelligent recommendation of the suppliers is completed by constructing a recommendation training model through the mutual matching between a convolutional neural network and a loss function, the convolutional neural network realizes the construction of a neural network architecture through the mutual matching among a convolutional layer, a pooling layer and a full-connection layer, and further realizes the processing and analysis of network data to complete the intelligent recommendation of the suppliers;
the convolution layer carries out convolution operation on input data through a convolution kernel, the convolution kernel is used for detecting specific features in the input data, the convolution operation is carried out by generating a plurality of feature maps and inputting the feature maps to the full-connection layer, and a specific implementation formula of the convolution operation is as follows:
wherein A represents a data set of data, K represents a convolution kernel, O represents a generated feature map, m and n represent sequences in the set respectively, and i and j represent data sequences in the generated feature map;
the pooling layer reduces the size of the feature mapping through a maximum pooling operation, and reserves main features, wherein the maximum pooling operation selects a maximum value as output through the maximum pooling operation under the condition of input data A and a range space k of the maximum pooling operation, and a specific implementation formula of the maximum pooling operation is as follows:
max(A,k)＝max i,j (A(i:i+k-1,j:j+k-1))
wherein a represents a data set of data, k represents a range space of a maximum pooling operation, and i and j represent data sequences in the maximum pooling operation space;
the full connection layer maps and flattens the characteristics output by the convolution layer and the pooling layer into one-dimensional vector data and outputs the one-dimensional vector data, and the specific implementation formula is as follows:
f＝Wx+b
where W represents a weight matrix, b represents a bias vector, x represents data in data set a, and i, j represents a data sequence in the data set.
As a preferable scheme of the provider intelligent recommendation method based on the neural network algorithm, the invention comprises the following steps: the intelligent recommendation of the suppliers is completed by constructing a recommendation training model through the mutual matching between a convolutional neural network and a loss function, the convolutional neural network is used for processing and analyzing the original data, and the analysis processing result and the output result of the loss function are compared and authenticated, so that the recommendation of the suppliers is realized;
the loss function is a mean value of square difference between a model predicted value and a true value through a mean square error loss function so as to ensure the accuracy of supplier recommendation, and a specific expression formula of the loss function is as follows:
wherein A represents a predicted value of the model, A' represents a true value, and N represents a total recommended number.
Another object of the present invention is to provide a provider intelligent recommendation system based on a neural network algorithm, which can provide a user with more convenient, efficient and accurate provider selection service by combining the neural network algorithm, personalized recommendation and automatic evaluation technology, thereby improving user experience, reducing time and resource waste, and having important practical application value in provider calendaring and purchasing decisions.
As a preferable scheme of the intelligent recommendation system of the provider based on the neural network algorithm, the invention comprises the following steps: the system comprises a data analysis module, a data processing module, a recommendation model building module and an automatic evaluation module; the data analysis module is used for analyzing the demand information of the user based on the data of the suppliers and the user and extracting key features; the data processing module is used for carrying out data processing on user demand information by utilizing a neural network algorithm so as to further carry out feature extraction; the recommendation model building module is used for building a recommendation training model so as to complete the processing of the feature vectors and intelligent recommendation of suppliers; the automatic evaluation module is used for automatically evaluating the recommended suppliers to determine the matching degree of the recommendation.
A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of a vendor intelligent recommendation method based on a neural network algorithm.
A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of a vendor intelligent recommendation method based on a neural network algorithm.
The invention has the beneficial effects that: according to the invention, through analyzing the demand information and the behaviors of the user, the personalized recommendation result can be provided according to the user preference and the provider matched with the historical interaction recommendation and the user, the user demand and the user preference can be predicted more accurately through the neural network algorithm, more accurate recommendation is provided, the user can be helped to know the recommendation reliability through automatically evaluating the provider, the workload of self evaluation of the user is reduced, the neural network algorithm, the personalized recommendation and the automatic evaluation technology are combined, more convenient, efficient and accurate provider selection service can be provided for the user, the user experience is improved, and the waste of time and resources is reduced, so that the method has important practical application value in the aspects of provider hanging calendar and purchasing decision.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
fig. 1 is a schematic diagram of overall steps of a provider intelligent recommendation method based on a neural network algorithm.
Fig. 2 is a schematic diagram of the overall composition structure of the intelligent recommendation system of the provider based on the neural network algorithm.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways other than those described herein, and persons skilled in the art will readily appreciate that the present invention is not limited to the specific embodiments disclosed below.
Further, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic can be included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
While the embodiments of the present invention have been illustrated and described in detail in the drawings, the cross-sectional view of the device structure is not to scale in the general sense for ease of illustration, and the drawings are merely exemplary and should not be construed as limiting the scope of the invention. In addition, the three-dimensional dimensions of length, width and depth should be included in actual fabrication.
Also in the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper, lower, inner and outer", etc. are based on the orientation or positional relationship shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first, second, or third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected, and coupled" should be construed broadly in this disclosure unless otherwise specifically indicated and defined, such as: can be fixed connection, detachable connection or integral connection; it may also be a mechanical connection, an electrical connection, or a direct connection, or may be indirectly connected through an intermediate medium, or may be a communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
Example 1
Referring to fig. 1, for a first embodiment of the present invention, there is provided a vendor intelligent recommendation method based on a neural network algorithm, comprising the steps of,
s1: the user's demand information is analyzed based on the vendor and user data.
Specifically, the analysis of the user's demand information based on the provider and user data is obtained by analyzing the provider data and the user data, and the analysis of the provider data and the user data includes analysis of historical transaction data of the user, time series analysis of user behavior data and analysis of user's preferences based on machine learning.
Further, the time series analysis of the user behavior data analyzes the user behavior data through a simple exponential smoothing method, weighting processing is performed on past behavior data through a weighted average idea to obtain a smoothed sequence, future behavior prediction is performed by using the smoothed sequence, the exponential smoothing method is that the past behavior data is subjected to weighted average to enable the data to be smoother, further future data change trend is reflected, and future behavior is predicted, wherein the specific implementation formula is as follows:
wherein,representing the predicted value of time t+1, which is the behavior prediction of the user, y t The representation is the actual observation of time t, < >>Time setting representationPredicted value of t, alpha represents a smoothing coefficient, and the value is 0.78;
furthermore, the preference of the user based on the machine learning analysis is realized through a decision tree algorithm, the decision tree algorithm divides the user data into a data space, classifies the data of the data space, analyzes the data based on the characteristics, and further completes the preference analysis of the user, and the specific realization formula is as follows:
where n represents a classification for user preference, p i Representing the probability that the sample belongs to class i, and H represents the entropy of the information of the different classes.
S2: and carrying out data processing on the requirement information of the user by using a neural network algorithm.
Specifically, the data processing of the demand information of the user by using the neural network algorithm is to complete the standardization of the data and the duplication removal of the data by using the neural network algorithm, wherein the standardization of the data is to map the data into standard distribution with a mean value of 0 and a standard deviation of 1, so that the data is more uniform in normal distribution, and the specific implementation formula is as follows:
wherein X represents original data, X' represents data after standardized processing, mu represents the mean value of the original data X, and sigma represents the standard deviation of the original data;
further, the data deduplication is based on a neural network architecture and is realized by measuring similarity measurement between data samples, the neural network architecture is composed of two sub-networks with the same structure, the sub-networks are spaces for processing one data sample, the similarity measurement between the measured data samples is realized by measuring cosine similarity between the data samples, and a specific realization formula is as follows:
wherein A and B respectively represent vector sets of two data samples, a 1 ,a 2 ,...,a n Representing data within vector set A, b 1 ,b 2 ,...,b n Representing data within vector set B;
further, the implementation rule of the data deduplication is as follows:
when the value of cosine similarity cos (A, B) is 1, the similarity between the two data is completely overlapped, and one data is needed to be removed;
when the value of cosine similarity cos (A, B) is 0, the similarity between the two data is partially overlapped, and one data is needed to be removed;
when the cosine similarity cos (A, B) is-1, it indicates that the similarity between the two data is completely opposite, and the two data should be kept.
S3: and building a recommendation training model to complete intelligent recommendation of the suppliers.
Specifically, the recommendation training model is built to complete intelligent recommendation of the suppliers through mutual matching among the convolutional neural network and the loss function, and the convolutional neural network realizes building of a neural network architecture through mutual matching among the convolutional layer, the pooling layer and the full-connection layer, so that processing and analysis of network data are realized, and intelligent recommendation of the suppliers is completed.
Further, the convolution layer performs a convolution operation on the input data through a convolution kernel, the convolution kernel is used for detecting specific features in the input data, the convolution operation is performed by generating a plurality of feature maps and inputting the feature maps to the full-connection layer, and a specific implementation formula of the convolution operation is as follows:
wherein A represents a data set of data, K represents a convolution kernel, O represents a generated feature map, m and n represent sequences in the set, respectively, and i and j represent data sequences in the generated feature map.
Furthermore, the pooling layer reduces the size of the feature map through a maximum pooling operation, and retains the main features, wherein the maximum pooling operation selects the maximum value as output through the maximum pooling operation under the condition of input data a and the range space k of the maximum pooling operation, and a specific implementation formula of the maximum pooling operation is as follows:
max(A,k)＝max i,j (A(i:i+k-1,j:j+k-1))
where a represents a data set of data, k represents a range space of a maximum pooling operation, and i, j represent a data sequence in the maximum pooling operation space.
Specifically, the full connection layer maps and flattens the features output by the convolution layer and the pooling layer into one-dimensional vector data, and outputs the one-dimensional vector data, and a specific implementation formula is as follows:
f＝Wx+b
where W represents a weight matrix, b represents a bias vector, x represents data in the data set a, and i, j represent data sequences in the data set.
Further specifically, the loss function is a mean value of square difference between a model predicted value and a true value calculated through a mean square error loss function so as to ensure the accuracy of supplier recommendation, and a specific expression formula of the loss function is as follows:
wherein A represents a predicted value of the model, A' represents a true value, and N represents a total recommended number.
Specifically, the intelligent recommendation of the suppliers is achieved by constructing a recommendation training model through mutual matching between a convolutional neural network and a loss function, original data are processed and analyzed through the convolutional neural network, and comparison and authentication are carried out on the analysis processing result and the output result of the loss function, so that the recommendation of the suppliers is achieved.
S4: and automatically evaluating the recommended suppliers based on the evaluation model.
Specifically, the automatic evaluation of the recommended suppliers based on the evaluation model is completed through the evaluation model, wherein the evaluation model comprises evaluation of the accuracy of the supplier recommendation, evaluation of the root mean square of the supplier recommendation and evaluation of the average absolute error of the supplier recommendation.
Further, the evaluation of the accuracy of the provider recommendation is that the number of correctly recommended samples is in proportion to the total recommended number of samples, and a specific calculation formula is as follows:
wherein A represents the total recommended number of samples, A n Indicating the number of samples completing the correct recommendation, A c Indicating the accuracy of the vendor recommendation.
Further, the evaluation of the accuracy of the provider recommendation is performed by calculating the proportion of the actual recommended positive category to all the recommended positive categories, and the specific calculation formula is as follows:
wherein N is T Indicating the number of actual recommended positive categories, N F Indicating the number of false recommendations in the recommendation positive category.
Further, the evaluation of the recommended root mean square of the provider is obtained by calculating the root mean square of the recommended data, and the specific calculation formula is as follows:
where N represents the total recommended number of times, a represents the total recommended number of samples, and i represents the data sequence in the recommended data set a.
Specifically, the evaluation of the average absolute error recommended by the provider is obtained by calculating the average absolute error between the recommended data and the real data, and the specific calculation formula is as follows:
wherein N represents the total recommended number, A represents the set of recommended data, i represents the data sequence in the number of correct recommendations A, A represents the number of correct recommendations A n Indicating the number of samples that completed the correct recommendation.
Example 2
Referring to fig. 2, for a second embodiment of the present invention, a provider intelligent recommendation system based on a neural network algorithm is provided, which includes a data analysis module, a data processing module, a recommendation model building module and an automatic evaluation module;
specifically, the data analysis module is used for analyzing the requirement information of the user based on the data of the suppliers and the users and extracting key features; the data processing module is used for carrying out data processing on user demand information by utilizing a neural network algorithm so as to further carry out feature extraction; the recommendation model building module is used for building a recommendation training model so as to complete the processing of the feature vectors and intelligent recommendation of suppliers; the automatic evaluation module is used for automatically evaluating the recommended suppliers to determine the matching degree of the recommendation.
Further, the data analysis module is used for identifying user demand information and preference information through a data mining and analysis technology based on provider and user data; collecting data from a provider database and a user history, the collected data including product information of the provider, price, transaction records of the user, and dot product behavior of the user; and extracting characteristics of the collected data, wherein the extracted characteristics comprise purchase evaluation rate, purchase category and specific field of interest of the user in the transaction record.
Furthermore, the data processing module is used for carrying out data processing on user demand information by utilizing a neural network algorithm, wherein the data processing comprises feature extraction and data preprocessing, and the processed data is used as input data of a recommendation model; the data preprocessing is to perform data cleaning and normalization on the extracted features so that the neural network can process the data.
Specifically, the recommendation model building module is realized by building a neural network model, and the processed data feature vector is used as input data of a recommendation model, so that intelligent recommendation of a provider is completed; the neural network model is realized through a convolutional neural network, the convolutional neural network realizes the construction of a neural network architecture through the mutual coordination among a convolutional layer, a pooling layer and a full-connection layer, and further realizes the processing and analysis of network data, and the intelligent recommendation of suppliers is completed.
Further specifically, the automatic evaluation module automatically evaluates the recommended suppliers based on an evaluation model, so as to determine the recommended matching degree and the recommended quality; the evaluation model includes an evaluation of accuracy of the vendor recommendation, an evaluation of root mean square of the vendor recommendation, and an evaluation of average absolute error of the vendor recommendation.
Further, in order to verify the beneficial effects of the present invention, the beneficial effects of the present invention are highlighted by comparing the experimental data with the table, and the following table is specific:
| application scenario | Accuracy rate of | Accuracy of | Root mean square deviation | Average absolute error |
| The invention is that | 0.93 | 0.91 | 0.11 | 0.07 |
| Conventional recommendation | 0.75 | 0.73 | 0.24 | 0.16 |
By comparing the table data, it is easy to find that the accuracy and precision of the provider intelligent recommendation system based on the neural network algorithm in the invention are higher than those of the conventional technical means, and the root mean square error and average absolute error of the provider recommendation are lower than those of the conventional technical means, so that the intelligent recommendation of the provider is better than that of the conventional technical means.
Still further, the functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-only memory (ROM), a random access memory (RAM, randomAccessMemory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
Furthermore, in an effort to provide a concise description of the exemplary embodiments, all features of an actual implementation may not be described (i.e., those not associated with the best mode presently contemplated for carrying out the invention, or those not associated with practicing the invention).
It should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort might be complex and time consuming, but would nevertheless be a routine undertaking of design, fabrication, and manufacture for those of ordinary skill having the benefit of this disclosure.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.
Claims (10)
1. A provider intelligent recommendation method based on a neural network algorithm is characterized by comprising the following steps of: comprises the steps of,
analyzing the user's demand information based on the provider and the user data;
carrying out data processing on the requirement information of the user by utilizing a neural network algorithm;
building a recommendation training model to complete intelligent recommendation of the suppliers;
and automatically evaluating the recommended suppliers based on the evaluation model.
2. The intelligent recommendation method for suppliers based on neural network algorithm of claim 1, wherein: the analysis of the user's demand information based on the provider and user data is obtained by analyzing the provider and user data, including analysis of historical transaction data of the user, time series analysis of user behavior data, and analysis of user preferences based on machine learning.
3. The intelligent recommendation method for suppliers based on neural network algorithm of claim 2, wherein: the user behavior data is analyzed through a simple exponential smoothing method in the time sequence analysis of the user behavior data, the exponential smoothing method enables the data to be smoother through weighted average of past behavior data, future data change trend is reflected, future behaviors are predicted, and the specific implementation formula is as follows:
wherein,representing the predicted value of time t+1, which is the behavior prediction of the user, y t The representation is the actual observation of time t, < >>The predicted value for time t is represented, and α represents the smoothing coefficient and is 0.78.
4. A neural network algorithm-based vendor intelligent recommendation method as claimed in claim 3, wherein: the data processing of the demand information of the user by using the neural network algorithm is to complete the standardization of the data and the duplication removal of the data by using the neural network algorithm, wherein the standardization of the data is to map the data into standard distribution with the mean value of 0 and the standard deviation of 1, so that the data is more uniform in normal distribution, and the specific implementation formula is as follows:
wherein X represents the original data, X' represents the normalized data, mu represents the mean value of the original data X, and sigma represents the standard deviation of the original data.
5. The intelligent recommendation method for suppliers based on neural network algorithm according to claim 4, wherein: the data deduplication is based on a neural network architecture and is realized by measuring similarity measurement among data samples, the neural network architecture is composed of two sub-networks with the same structure, the sub-networks are spaces for processing one data sample, the similarity measurement among the measured data samples is realized by measuring cosine similarity among the original samples, and a specific realization formula is as follows:
wherein A and B respectively represent vector sets of two data samples, a 1 ,a 2 ,...,a n Representing data within vector set A, b 1 ,b 2 ,...,b n Representing data within vector set B;
the implementation rule of the data deduplication is as follows:
when the value of cosine similarity cos (A, B) is 1, the similarity between the two data is completely overlapped, and one data is needed to be removed;
when the value of cosine similarity cos (A, B) is 0, the similarity between the two data is partially overlapped, and one data is needed to be removed;
when the cosine similarity cos (A, B) is-1, it indicates that the similarity between the two data is completely opposite, and the two data should be kept.
6. The intelligent recommendation method for suppliers based on neural network algorithm according to claim 5, wherein: the intelligent recommendation of the suppliers is completed by constructing a recommendation training model through the mutual matching between a convolutional neural network and a loss function, the convolutional neural network realizes the construction of a neural network architecture through the mutual matching among a convolutional layer, a pooling layer and a full-connection layer, and further realizes the processing and analysis of network data to complete the intelligent recommendation of the suppliers;
the convolution layer carries out convolution operation on input data through a convolution kernel, the convolution kernel is used for detecting specific features in the input data, the convolution operation is carried out by generating a plurality of feature maps and inputting the feature maps to the full-connection layer, and a specific implementation formula of the convolution operation is as follows:
wherein A represents a data set of data, K represents a convolution kernel, O represents a generated feature map, m and n represent sequences in the set respectively, and i and j represent data sequences in the generated feature map;
the pooling layer reduces the size of the feature mapping through a maximum pooling operation, and reserves main features, wherein the maximum pooling operation selects a maximum value as output through the maximum pooling operation under the condition of input data A and a range space k of the maximum pooling operation, and a specific implementation formula of the maximum pooling operation is as follows:
max(A,k)＝max i,j (A(i:i+k-1,j:j+k-1))
wherein a represents a data set of data, k represents a range space of a maximum pooling operation, and i and j represent data sequences in the maximum pooling operation space;
the full connection layer maps and flattens the characteristics output by the convolution layer and the pooling layer into one-dimensional vector data and outputs the one-dimensional vector data, and the specific implementation formula is as follows:
f＝Wx+b
where W represents a weight matrix, b represents a bias vector, x represents data in the data set a, and i, j represent data sequences in the data set.
7. The intelligent recommendation method for suppliers based on neural network algorithm according to claim 6, wherein: the intelligent recommendation of the suppliers is completed by constructing a recommendation training model through the mutual matching between a convolutional neural network and a loss function, the convolutional neural network is used for processing and analyzing the original data, and the analysis processing result and the output result of the loss function are compared and authenticated, so that the recommendation of the suppliers is realized;
the loss function is a mean value of square difference between a model predicted value and a true value through a mean square error loss function so as to ensure the accuracy of supplier recommendation, and a specific expression formula of the loss function is as follows:
wherein A represents a predicted value of the model, A' represents a true value, and N represents a total recommended number.
8. A system adopting the neural network algorithm-based vendor intelligent recommendation method according to any one of claims 1 to 7, comprising a data analysis module, a data processing module, a recommendation model building module and an automatic evaluation module;
the data analysis module is used for analyzing the demand information of the user based on the data of the suppliers and the user and extracting key features;
the data processing module is used for carrying out data processing on user demand information by utilizing a neural network algorithm so as to further carry out feature extraction;
the recommendation model building module is used for building a recommendation training model so as to complete the processing of the feature vectors and intelligent recommendation of suppliers;
the automatic evaluation module is used for automatically evaluating the recommended suppliers to determine the matching degree of the recommendation.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311046202.5A CN117217763A (en) | 2023-08-18 | 2023-08-18 | Intelligent recommendation method and system for suppliers based on neural network algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311046202.5A CN117217763A (en) | 2023-08-18 | 2023-08-18 | Intelligent recommendation method and system for suppliers based on neural network algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117217763A true CN117217763A (en) | 2023-12-12 |
Family
ID=89039794
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311046202.5A Pending CN117217763A (en) | 2023-08-18 | 2023-08-18 | Intelligent recommendation method and system for suppliers based on neural network algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117217763A (en) |
-
2023
- 2023-08-18 CN CN202311046202.5A patent/CN117217763A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108564286B (en) | Artificial intelligent financial wind-control credit assessment method and system based on big data credit investigation | |
| US7930242B2 (en) | Methods and systems for multi-credit reporting agency data modeling | |
| US20210103858A1 (en) | Method and system for model auto-selection using an ensemble of machine learning models | |
| CN112990386B (en) | User value clustering method and device, computer equipment and storage medium | |
| CN110689355A (en) | Client classification method, device, computer equipment and storage medium | |
| CN110930038A (en) | Loan demand identification method, loan demand identification device, loan demand identification terminal and loan demand identification storage medium | |
| CN113901236A (en) | Target identification method and device based on artificial intelligence, electronic equipment and medium | |
| CN115526652A (en) | Client loss early warning method and system based on machine learning | |
| CN117235608B (en) | Risk detection method, risk detection device, electronic equipment and storage medium | |
| Tiruneh et al. | Feature selection for construction organizational competencies impacting performance | |
| CN116501979A (en) | Information recommendation method, information recommendation device, computer equipment and computer readable storage medium | |
| CN110796381A (en) | Method and device for processing evaluation indexes of modeling data, terminal equipment and medium | |
| CN117217763A (en) | Intelligent recommendation method and system for suppliers based on neural network algorithm | |
| CN115310606A (en) | Deep learning model depolarization method and device based on data set sensitive attribute reconstruction | |
| CN115062876A (en) | OCR-based process rapid modeling method, system, equipment and storage medium | |
| CN114841239A (en) | Marketing company financial abnormity analysis method based on machine learning | |
| CN114612239A (en) | Stock public opinion monitoring and wind control system based on algorithm, big data and artificial intelligence | |
| CN113191570A (en) | Fund planning recommendation method, device and equipment based on deep learning | |
| CN112926816B (en) | Vendor evaluation method, device, computer device and storage medium | |
| EP3794523A1 (en) | System and method for processing distorted or inaccurate input data obtained from an environment | |
| CN115237970A (en) | Data prediction method, device, equipment, storage medium and program product | |
| US20230419195A1 (en) | System and Method for Hierarchical Factor-based Forecasting | |
| US20230385664A1 (en) | A computer-implemented method for deriving a data processing and inference pipeline | |
| ABIDAR et al. | Predicting Customer Segment Changes to Enhance Customer Retention: A Case Study for Online Retail using Machine Learning | |
| CN117934035A (en) | Method, device and storage medium for predicting construction cost of building construction |
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 |