CN112749345A - K nearest neighbor matrix decomposition recommendation method based on neural network - Google Patents
K nearest neighbor matrix decomposition recommendation method based on neural network Download PDFInfo
- Publication number
- CN112749345A CN112749345A CN202110176380.4A CN202110176380A CN112749345A CN 112749345 A CN112749345 A CN 112749345A CN 202110176380 A CN202110176380 A CN 202110176380A CN 112749345 A CN112749345 A CN 112749345A
- Authority
- CN
- China
- Prior art keywords
- matrix
- neural network
- scoring
- user
- items
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000011159 matrix material Substances 0.000 title claims abstract description 98
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 32
- 239000013598 vector Substances 0.000 claims description 37
- 230000006870 function Effects 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 16
- 210000004205 output neuron Anatomy 0.000 claims description 11
- 210000002569 neuron Anatomy 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 210000002364 input neuron Anatomy 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000009471 action Effects 0.000 description 5
- 238000001914 filtration Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram 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
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- G06N3/045—Combinations of 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
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a K neighbor matrix decomposition recommendation method based on a neural network, which comprises the following steps: acquiring a data set from a website, selecting a plurality of users and a plurality of items in the data set, and forming an initial matrix according to the plurality of users and the plurality of items; predicting and filling partial vacancy values of the initial matrix by using an RBF neural network so as to reduce the sparsity of the initial matrix; calculating by using a KNN algorithm for the target user to find K nearest neighbor users in the initial matrix obtained in the step S2, similarly, finding K nearest neighbor items for the target item, and constructing a scoring matrix according to the K nearest neighbor users and the K nearest neighbor items; carrying out matrix decomposition on the scoring matrix, wherein the U matrix represents the characteristics of a target user, the V matrix represents the characteristics of a target project, a new implicit characteristic is extracted, and the interpretability of the decomposition is increased; and multiplying the determinant of the scoring matrix by 1/k to serve as a scoring standard, giving a recommendation threshold value, and judging whether to give a recommendation or not.
Description
Technical Field
The invention relates to the field of recommendation systems, in particular to a K neighbor matrix decomposition recommendation method based on a neural network.
Background
With the continuous increase of online information quantity and the continuous increase of online active users, the users can not accurately grasp the information on the internet, so that the user needs to obtain the information which can not be satisfied, the experience of surfing the internet is poor, and the occurrence of a recommendation system becomes an effective strategy for overcoming information overload. The recommendation system excavates items (such as information, services, articles and the like) which are interested by the user from the mass data through a recommendation algorithm according to the requirements, interests and the like of the user, and recommends the results to the user in a personalized list mode.
Algorithms of the existing recommendation systems are roughly divided into three categories, namely a recommendation system based on content, a recommendation system based on collaborative filtering and a hybrid recommendation system. The difficulty of the content-based recommendation system lies in the acquisition and filtering of user content information, the difficulty of the hybrid recommendation system lies in the combination of the information content of a user and entity information of the user, and the problem of how to express the relation between the information content of the user and the entity information of the user into structured data, so that collaborative filtering becomes the most mainstream recommendation system algorithm at present.
The invention provides a neural network-based K neighbor matrix decomposition algorithm, which aims to solve the problems of matrix sparsity and expansibility in a collaborative filtering algorithm and enable a machine to learn and predict behavior characteristics of a user from an incomplete information matrix when high-quality information and edge information of the user cannot be acquired.
Disclosure of Invention
The invention aims to provide a K neighbor matrix decomposition recommendation method based on a neural network, which is characterized in that according to a public data set provided by a website, the missing matrix is completed by using the currently popular neural network, similar users and similar items are clustered by using a clustering technology, and finally, the items favored by the users are recommended to the corresponding users in a recommendation list mode.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a K neighbor matrix decomposition recommendation method based on a neural network comprises the following steps:
step S1, acquiring a data set from a website, selecting a plurality of users and a plurality of projects in the data set, and forming an initial matrix according to the users and the projects;
step S2, predicting and filling partial vacancy values of the initial matrix by using an RBF neural network so as to reduce the sparsity of the initial matrix;
step S3, for the target user, calculating by using a KNN algorithm to find K nearest neighbor users in the initial matrix obtained in the step S2, similarly, for the target project, also finding K nearest neighbor projects, and constructing a scoring matrix according to the K nearest neighbor users and the K nearest neighbor projects;
s4, carrying out matrix decomposition on the scoring matrix, wherein a U matrix represents the characteristics of a target user, a V matrix represents the characteristics of a target project, and a latent semantic characteristic U, V increases the interpretability of the decomposition;
and step S5, multiplying the determinant of the scoring matrix by 1/k to serve as a scoring standard, giving a recommendation threshold value, and judging whether to give a recommendation or not.
Further, the method comprises a preprocessing step of the initial matrix, the preprocessing step comprising:
deleting users with sparsity of more than 99.5% among the users, wherein the calculation formula of the sparsity a of the users is as follows:
a 1-number of items rated by the user/total number of items;
deleting the items without any score in the plurality of items.
Further, the method further includes a normalization process for the initial matrix, where the normalization process is used to map the score values uniformly to the interval of [0,1], and specifically by using a calculation formula as follows:
wherein y represents an initial item score value, yminLower-bound score value, y, representing the original project datamaxRepresents the upper score value of the raw data and Y represents the normalized data.
Further, the step S2 specifically includes:
s201, normalizing the user vectors with a plurality of common scoring items to be used as the input of the RBF neural network, and determining the number of input layers of the neural network; the RBF neural network has high speed for training large-scale data, selects local optimum instead of global optimum, and prevents overfitting;
s202, randomly selecting a batch of central nodes, adopting an unsupervised gradient descent method, updating the central nodes in a negative feedback manner, and finally determining the number of the central nodes of the hidden layer;
s203, the hidden layer neuron kernel function is a Gaussian function, the input information is subjected to space mapping transformation, and the output layer uses a linear weighting function;
s204, performing inverse normalization processing on the output of the neural network to obtain a prediction score in a range of [0,5 ];
s205, filling the blank values by using the trained neural network, filling the unpredictable blank values by using the mean value calculated in the claim 1, and forming a new scoring matrix with the sparsity reduced to be close to 0.
Further, the center, width and adjustment weight parameters are all adaptively adjusted to the optimal values through learning, and the iterative calculation is as follows:
Wkj(t) the adjustment weight between the kth output neuron and the jth hidden layer neuron at the time of the t iteration calculation;
Cpq(t) is the central component of the pth hidden layer neuron for the qt input neuron at the time of the t iteration;
dij(t) is the sum of center Cpq(t) a corresponding radial width;
eta is a learning factor;
e is an RBF neural network evaluation function:
wherein, the expected output value of the kth output neuron at the ith input sample is the expected output value of the kth output neuron; the net output value of the kth output neuron at the ith input sample is obtained.
Further, the step S3 includes:
dividing the new scoring matrix into user vectors according to rows, dividing the user vectors into item vectors according to columns, solving the similarity between a new target user i vector and the user vectors, solving the similarity between a new target item j and a target user, sequencing the similarity user vectors and the item vectors to respectively solve the previous K values, and forming a new scoring matrix by the K user vectors and the K item vectors; the similarity calculation method is shown as the formula (2):
where xi represents the value of an element in the i (j) vector, yi represents the corresponding value of an element in the user (item) vector, and Sim represents the calculated similarity value.
Further, the step S4 includes the following specific steps:
and SVD (singular value decomposition) is used for carrying out matrix decomposition on the new K-dimensional matrix M, the matrix is decomposed into SR (singular value decomposition), wherein U represents the total implicit feature of the target user, V represents the total implicit feature of the target item, the elements in the sigma are the singular values of the matrix, new KR (K) K-U (K) r (r) V (r) K) is continuously constructed according to the previous r singular values in the sigma, the dimensionality of the implicit matrix is reduced, and the calculation amount is reduced.
Further, the step S5 includes the following specific steps:
multiplying the value of the determinant of the scoring matrix by 1/k to serve as the predicted scoring of a target user I on a target item j, setting a threshold value c to be 4, adding items with predicted scoring being more than or equal to 4 into a recommendation set Si, wherein the Si comprises elements which are scoring items of more than or equal to 4 in all scoring of the I user, sorting the items in the set Si according to the scoring size, and recommending items of Top5 to the user I, wherein the user I belongs to a set I to be { I1, I2, I3, …, I943 }.
Compared with the prior art, the invention has at least one of the following advantages:
when the high-quality content information and the edge information of the user cannot be obtained, the method and the device can provide interesting movie items for different users according to the single score of the user, but not limited to movie recommendations, such as music recommendations, e-book recommendations, news recommendations and the like.
The sparsity of the initial scoring matrix (initial matrix for short) is effectively reduced by utilizing the strong learning capacity of the neural network, and the recommendation result of the final result is greatly influenced.
The new users and the new items are introduced and the grading prediction is carried out, so that the expansibility of the recommendation system is increased, and the recommendation system can recommend the new users and the new items which are added continuously more and more accurately.
Drawings
FIG. 1 is a general flowchart of a recommendation method based on neural network K-nearest neighbor matrix decomposition according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an RBF neural network according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings 1-2 and the detailed description thereof. The advantages and features of the present invention will become more apparent from the following description. It is to be noted that the drawings are in a very simplified form and are all used in a non-precise scale for the purpose of facilitating and distinctly aiding in the description of the embodiments of the present invention. To make the objects, features and advantages of the present invention comprehensible, reference is made to the accompanying drawings. It should be understood that the structures, ratios, sizes, and the like shown in the drawings and described in the specification are only used for matching with the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the implementation conditions of the present invention, so that the present invention has no technical significance, and any structural modification, ratio relationship change or size adjustment should still fall within the scope of the present invention without affecting the efficacy and the achievable purpose of the present invention.
It is to be noted that, in the present invention, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "include," "include," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, neural network-based K-neighbor matrix factorization recommendation method, article, or field device that includes a list of elements includes not only those elements, but also other elements not expressly listed, or also elements inherent to such a process, neural network-based K-neighbor matrix factorization recommendation method, article, or field device. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of additional like elements in the process of comprising the element, the neural network-based K-neighbor matrix factorization recommendation method, the article, or the field device.
Referring to fig. 1-2, a method for decomposing and recommending a K-nearest neighbor matrix based on a neural network according to the present embodiment includes:
step S1, acquiring a data set from a website, selecting 943 users and 11508 items in the data set, and forming an initial matrix according to the users and the items;
step S2, predicting and filling partial vacancy values of the initial matrix by using an RBF neural network so as to reduce the sparsity of the initial matrix;
step S3, for the target user, calculating by using a KNN algorithm to find K nearest neighbor users in the initial matrix obtained in the step S2, similarly, for the target project, also finding K nearest neighbor projects, and constructing a scoring matrix according to the K nearest neighbor users and the K nearest neighbor projects;
s4, carrying out matrix singular value decomposition on the scoring matrix by SVD, wherein a U matrix represents the characteristics of a target user, a V matrix represents the characteristics of a target project, and latent semantic characteristics U, V increase the interpretability of decomposition;
and step S5, multiplying the determinant of the scoring matrix by 1/k to serve as a scoring standard, giving a recommendation threshold value, and judging whether to give a recommendation or not.
In this embodiment, in step S3, the magnitude of the parameter K is determined by specifically using the magnitude of the Mean Square Error (MSE), and the MSE is calculated as follows:
wherein xiThe score is given to the real item or items,
the scoring criteria of step S5;
in this embodiment, the method further includes a preprocessing step for the initial matrix, where the preprocessing step includes: deleting meaningless rows and columns, and deleting users with sparsity of more than 99.5% among the users, wherein the users with too high sparsity have no reference value, and the final result is not influenced by discarding the users, wherein the calculation formula of the sparsity a of the users is as follows:
a 1-number of items rated by the user/total number of items;
deleting the items without any score in the plurality of items.
In this embodiment, the method further includes a normalization process for the initial matrix, where the normalization process is used to map the score values uniformly to the interval of [0,1], and specifically according to a calculation formula:
wherein y represents an initial item score value, yminLower-bound score value, y, representing the original project datamaxRepresents the upper score value of the raw data and Y represents the normalized data.
In this embodiment, the step S2 specifically includes:
s201, normalizing the user vectors with a plurality of common scoring items to be used as the input of the RBF neural network, and determining the number of input layers of the neural network; the RBF neural network has high speed for training large-scale data, selects local optimum instead of global optimum, and prevents overfitting;
s202, randomly selecting a batch of central nodes, adopting an unsupervised gradient descent method, updating the central nodes in a negative feedback manner, and finally determining the number of the central nodes of the hidden layer;
s203, the hidden layer neuron kernel function (action function) is a Gaussian function, and the input information is subjected to space mapping transformation;
s204, performing inverse normalization processing on the output of the neural network to obtain a prediction score in a range of [0,5 ];
s205, filling the blank values by using the trained neural network, filling the unpredictable blank values by using the mean value calculated in the claim 1, and forming a new scoring matrix with the sparsity reduced to be close to 0.
In this embodiment, the center, the width, and the adjustment weight parameters are all adaptively adjusted to the optimal values through learning, and the iterative computation is as follows:
Wkj(t) the adjustment weight between the kth output neuron and the jth hidden layer neuron at the time of the t iteration calculation;
Cpq(t) is the central component of the pth hidden layer neuron for the qt input neuron at the time of the t iteration;
dij(t) is the sum of center Cpq(t) a corresponding width;
eta is a learning factor, the learning factor is circulated, an initial value is set to be 0.001, the circulation is carried out for 100 times, each step is 0.001, and a final eta value is determined;
e is an RBF neural network evaluation function:
wherein, ylkIs the expected output value of the kth output neuron at the ith input sample; o islkThe net output value of the kth output neuron at the ith input sample is obtained.
In this embodiment, in step S203, the hidden layer neuron kernel function (action function) is a gaussian function, and performs spatial mapping transformation on the input information; the inseparability of data in a low-dimensional space can have differentiability in a high-dimensional space, the nature of a Gaussian kernel is that the 'similarity' between samples is measured, and in a space depicting the 'similarity', the samples of the same type are better gathered together, and then the linear divisibility is realized, wherein a Gaussian function is as follows:
where Xc is the kernel function center, σ is the width parameter of the function, the radial acting range of the control function, and yi is the final network output.
In this embodiment, the step S3 includes: dividing the new scoring matrix into user vectors according to rows, dividing the user vectors into item vectors according to columns, solving the similarity between a new target user i vector and the user vectors, solving the similarity between a new target item j and the item vectors, respectively solving the previous K values of the similarity between the user vectors and the item vectors in a sorted manner, and forming a new scoring matrix by the K user vectors and the K item vectors; the similarity calculation method is shown as the formula (2):
where xi represents the value of an element in the i (j) vector, yi represents the corresponding value of an element in the user (item) vector, and Sim represents the calculated similarity value.
In this embodiment, the step S4 includes the following specific steps: and SVD (singular value decomposition) is used for carrying out matrix decomposition on the new K-dimensional matrix M, the matrix is decomposed into SR (singular value decomposition), wherein U represents the total implicit feature of the target user, V represents the total implicit feature of the target item, elements in the sigma are singular values of the matrix, new KR (K) K-U (K) sigma (r) V (r) K) is continuously constructed according to the previous r (r < K) singular values in the sigma, the dimensionality of the implicit matrix is reduced, and the calculation amount is reduced.
In this embodiment, the step S5 includes the following specific steps: multiplying the value of the determinant of the scoring matrix by 1/k to serve as the predicted scoring of a target user I on a target item j, setting a threshold value c to be 4, adding items with predicted scoring being more than or equal to 4 into a recommendation set Si, wherein the Si comprises elements which are scoring items of more than or equal to 4 in all scoring of the I user, sorting the items in the set Si according to the scoring size, and recommending items of Top5 to the user I, wherein the user I belongs to a set I to be { I1, I2, I3, …, I943 }.
The interest movie items can be provided for different users according to the single scores of the users, but are not limited to movie recommendations, such as music recommendations, e-book recommendations, news recommendations and the like.
While the present invention has been described in detail with reference to the preferred embodiments, it should be understood that the above description should not be taken as limiting the invention. Various modifications and alterations to this invention will become apparent to those skilled in the art upon reading the foregoing description. Accordingly, the scope of the invention should be determined from the following claims.
Claims (8)
1. A K neighbor matrix decomposition recommendation method based on a neural network is characterized by comprising the following steps:
step S1, acquiring a data set from a website, selecting a plurality of users and a plurality of projects in the data set, and forming an initial matrix according to the users and the projects;
step S2, predicting and filling partial vacancy values of the initial matrix by using an RBF neural network so as to reduce the sparsity of the initial matrix;
step S3, for the target user, calculating by using a KNN algorithm to find K nearest neighbor users in the initial matrix obtained in the step S2, similarly, for the target project, also finding K nearest neighbor projects, and constructing a scoring matrix according to the K nearest neighbor users and the K nearest neighbor projects;
step S4, carrying out matrix decomposition on the scoring matrix, wherein a U matrix represents the characteristics of a target user, a V matrix represents the characteristics of a target project, and a latent semantic characteristic U, V is adopted to increase the interpretability of the decomposition;
and step S5, multiplying the determinant of the scoring matrix by 1/k to serve as a scoring standard, giving a recommendation threshold value, and judging whether to give a recommendation or not.
2. The neural network-based K-neighbor matrix decomposition recommendation method of claim 1, further comprising a preprocessing step of the initial matrix, the preprocessing step comprising:
deleting users with sparsity of more than 99.5% among the users, wherein the calculation formula of the sparsity a of the users is as follows:
a 1-number of items rated by the user/total number of items;
deleting the items without any score in the plurality of items.
3. The neural network-based K-nearest neighbor matrix decomposition recommendation method according to claim 1 or 2, further comprising a normalization process for the initial matrix, the normalization process being used for uniformly mapping the score values to the interval of [0,1], specifically by the following calculation formula:
wherein y represents an initial item score value, yminLower-bound score value, y, representing the original project datamaxRepresents the upper score value of the raw data and Y represents the normalized data.
4. The neural network-based K-neighbor matrix decomposition recommendation method according to claim 3, wherein the step S2 specifically includes:
s201, normalizing the user vectors with a plurality of common scoring items to be used as the input of the RBF neural network, and determining the number of input layers of the neural network; the RBF neural network has high speed for training large-scale data, selects local optimum instead of global optimum, and prevents overfitting;
s202, randomly selecting a batch of central nodes, adopting an unsupervised gradient descent method, updating the central nodes in a negative feedback manner, and finally determining the number of the central nodes of the hidden layer;
s203, the hidden layer neuron kernel function is a Gaussian function, the input information is subjected to space mapping transformation, and the output layer uses a linear weighting function;
s204, performing inverse normalization processing on the output of the neural network to obtain a prediction score in a range of [0,5 ];
s205, filling the blank values by using the trained neural network, filling the unpredictable blank values by using the mean value calculated in the claim 1, and forming a new scoring matrix with the sparsity reduced to be close to 0.
5. The neural network-based K-neighbor matrix factorization recommendation method of claim 4,
the center, width and adjusting weight parameters are all self-adaptively adjusted to the optimal values through learning, and the iterative computation is as follows:
Wkj(t) the adjustment weight between the kth output neuron and the jth hidden layer neuron at the time of the t iteration calculation;
Cpq(t) is the central component of the pth hidden layer neuron for the qt input neuron at the time of the t iteration;
dij(t) is the sum of center Cpq(t) a corresponding radial width;
eta is a learning factor;
e is an RBF neural network evaluation function:
wherein, the expected output value of the kth output neuron at the ith input sample is the expected output value of the kth output neuron; the net output value of the kth output neuron at the ith input sample is obtained.
6. The neural network-based K-neighbor matrix decomposition recommendation method of claim 1, wherein the step S3 comprises:
dividing the new scoring matrix into user vectors according to rows, dividing the user vectors into item vectors according to columns, solving the similarity between a new target user i vector and the user vectors, solving the similarity between a new target item j and a target user, sequencing the similarity user vectors and the item vectors to respectively solve the previous K values, and forming a new scoring matrix by the K user vectors and the K item vectors; the similarity calculation method is shown as the formula (2):
where xi represents the value of an element in the i (j) vector, yi represents the corresponding value of an element in the user (item) vector, and Sim represents the calculated similarity value.
7. The neural network-based K-neighbor matrix decomposition recommendation method according to claim 1, wherein the step S4 comprises the following specific steps:
and SVD (singular value decomposition) is used for carrying out matrix decomposition on the new K-dimensional matrix M, the matrix is decomposed into SR (singular value decomposition), wherein U represents the total implicit feature of the target user, V represents the total implicit feature of the target item, the elements in the sigma are the singular values of the matrix, new KR (K) K-U (K) r (r) V (r) K) is continuously constructed according to the previous r singular values in the sigma, the dimensionality of the implicit matrix is reduced, and the calculation amount is reduced.
8. The neural network-based K-neighbor matrix decomposition recommendation method according to claim 1, wherein the step S5 comprises the following specific steps:
multiplying the value of the determinant of the scoring matrix by 1/k to serve as the predicted scoring of a target user I on a target item j, setting a threshold value c to be 4, adding items with predicted scoring being more than or equal to 4 into a recommendation set Si, wherein the Si comprises elements which are scoring items of more than or equal to 4 in all scoring of the I user, sorting the items in the set Si according to the scoring size, and recommending items of Top5 to the user I, wherein the user I belongs to a set I to be { I1, I2, I3, …, I943 }.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110176380.4A CN112749345B (en) | 2021-02-09 | 2021-02-09 | K neighbor matrix decomposition recommendation method based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110176380.4A CN112749345B (en) | 2021-02-09 | 2021-02-09 | K neighbor matrix decomposition recommendation method based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112749345A true CN112749345A (en) | 2021-05-04 |
| CN112749345B CN112749345B (en) | 2023-11-14 |
Family
ID=75651970
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110176380.4A Active CN112749345B (en) | 2021-02-09 | 2021-02-09 | K neighbor matrix decomposition recommendation method based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112749345B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113536140A (en) * | 2021-09-16 | 2021-10-22 | 广东众聚人工智能科技有限公司 | Real-time recommendation method and system based on small sample data |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106535A (en) * | 2013-02-21 | 2013-05-15 | 电子科技大学 | Method for solving collaborative filtering recommendation data sparsity based on neural network |
| CN108038629A (en) * | 2017-12-30 | 2018-05-15 | 北京工业大学 | A kind of optimization method based on collaborative filtering |
| CN108090229A (en) * | 2018-01-10 | 2018-05-29 | 广东工业大学 | A kind of method and apparatus that rating matrix is determined based on convolutional neural networks |
| CN108320187A (en) * | 2018-02-02 | 2018-07-24 | 合肥工业大学 | A kind of recommendation method based on depth social networks |
| CN109902201A (en) * | 2019-03-08 | 2019-06-18 | 天津理工大学 | A kind of recommended method based on CNN and BP neural network |
| US20190251435A1 (en) * | 2018-02-09 | 2019-08-15 | Daniel Shiebler | Matching cross domain user affinity with co-embeddings |
| CN110162706A (en) * | 2019-05-22 | 2019-08-23 | 南京邮电大学 | A kind of personalized recommendation method and system based on interaction data cluster |
| CN111598627A (en) * | 2020-05-26 | 2020-08-28 | 揭阳职业技术学院 | Personalized advertisement pushing method for elevator media terminal |
| CN111882381A (en) * | 2020-06-30 | 2020-11-03 | 桂林电子科技大学 | Travel recommendation method based on collaborative memory network |
-
2021
- 2021-02-09 CN CN202110176380.4A patent/CN112749345B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106535A (en) * | 2013-02-21 | 2013-05-15 | 电子科技大学 | Method for solving collaborative filtering recommendation data sparsity based on neural network |
| CN108038629A (en) * | 2017-12-30 | 2018-05-15 | 北京工业大学 | A kind of optimization method based on collaborative filtering |
| CN108090229A (en) * | 2018-01-10 | 2018-05-29 | 广东工业大学 | A kind of method and apparatus that rating matrix is determined based on convolutional neural networks |
| CN108320187A (en) * | 2018-02-02 | 2018-07-24 | 合肥工业大学 | A kind of recommendation method based on depth social networks |
| US20190251435A1 (en) * | 2018-02-09 | 2019-08-15 | Daniel Shiebler | Matching cross domain user affinity with co-embeddings |
| CN109902201A (en) * | 2019-03-08 | 2019-06-18 | 天津理工大学 | A kind of recommended method based on CNN and BP neural network |
| CN110162706A (en) * | 2019-05-22 | 2019-08-23 | 南京邮电大学 | A kind of personalized recommendation method and system based on interaction data cluster |
| CN111598627A (en) * | 2020-05-26 | 2020-08-28 | 揭阳职业技术学院 | Personalized advertisement pushing method for elevator media terminal |
| CN111882381A (en) * | 2020-06-30 | 2020-11-03 | 桂林电子科技大学 | Travel recommendation method based on collaborative memory network |
Non-Patent Citations (3)
| Title |
|---|
| ZHAOWEI QU 等: "Converged Recommendation System Based on RNN and BP Neural Networks", 《2018 IEEE INTERNATIONAL CONFERENCE ON BIG DATA AND SMART COMPUTING (BIGCOMP)》, pages 308 - 313 * |
| 孙坤: "基于改进协同过滤的个性化推荐方法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》, no. 2, pages 138 - 2237 * |
| 张锋 等: "使用BP神经网络缓解协同过滤推荐算法的稀疏性问题", 《计算机研究与发展》, pages 667 - 672 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113536140A (en) * | 2021-09-16 | 2021-10-22 | 广东众聚人工智能科技有限公司 | Real-time recommendation method and system based on small sample data |
| CN113536140B (en) * | 2021-09-16 | 2022-02-11 | 广东众聚人工智能科技有限公司 | Real-time recommendation method and system based on small sample data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112749345B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3920100A1 (en) | Adaptive pointwise-pairwise learning to rank | |
| CN112967088A (en) | Marketing activity prediction model structure and prediction method based on knowledge distillation | |
| CN111104601B (en) | Antagonistic multi-feedback-level paired personalized ranking method | |
| Navgaran et al. | Evolutionary based matrix factorization method for collaborative filtering systems | |
| Mehmanpazir et al. | Development of an evolutionary fuzzy expert system for estimating future behavior of stock price | |
| CN111583031A (en) | Application scoring card model building method based on ensemble learning | |
| CN111241289B (en) | Text clustering method based on graph theory and SOM network | |
| CN112529638B (en) | Service demand dynamic prediction method and system based on user classification and deep learning | |
| CN113822419B (en) | Self-supervision graph representation learning operation method based on structural information | |
| CN112287166A (en) | Movie recommendation method and system based on improved deep belief network | |
| CN111539444A (en) | Gaussian mixture model method for modified mode recognition and statistical modeling | |
| CN116415177A (en) | Classifier parameter identification method based on extreme learning machine | |
| CN115310675A (en) | Load estimation optimization method based on power grid user data set and neural network | |
| CN112749345B (en) | K neighbor matrix decomposition recommendation method based on neural network | |
| CN114564594A (en) | Knowledge graph user preference entity recall method based on double-tower model | |
| Chen et al. | Poverty/investment slow distribution effect analysis based on Hopfield neural network | |
| Farooq | Genetic algorithm technique in hybrid intelligent systems for pattern recognition | |
| Luo et al. | Addressing imbalance in weakly supervised multi-label learning | |
| Ma et al. | VNAS: Variational Neural Architecture Search | |
| Babeetha et al. | An enhanced kernel weighted collaborative recommended system to alleviate sparsity | |
| Sinaga et al. | Stock trend prediction using SV-kNNC and SOM | |
| CN115292587B (en) | Recommendation method and system based on knowledge distillation and causal reasoning | |
| CN114037931B (en) | Multi-view discriminating method of self-adaptive weight | |
| Liwei | Predictive analysis of machine learning error classification based on bayesian network | |
| Jiang et al. | A CTR prediction approach for advertising based on embedding model and deep learning |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |