Personalized recommendation system and recommendation method based on library real-time situation

Abstract

The invention belongs to the related technical field of information recommendation, and discloses a personalized recommendation system and recommendation method based on a real-time situation of a library, in the system, a user attribute fusion module performs feature weighted fusion on attribute features; the situational information fusion module performs weighted fusion on the situational information features of the library; the user preference identification module calculates similar feature vectors of candidate book representations and historical book representations and calculates weights of corresponding historical books, all the historical book representations are weighted and then input into a pooling layer, and reading preference features are obtained; and the recommendation probability calculation module splices the features and then sequentially passes through a linear layer, an activation layer and a full connection layer to obtain the recommendation probability of the candidate books. The system not only can capture long-term preferences and personality characteristics of the user, but also can provide recommendation content which better meets requirements for the user in a real-time situation, so that a highly personalized recommendation result is generated.

Description

Technical Field

[0001] The present invention belongs to the technical field related to information recommendation, and more specifically, relates to a personalized recommendation system and a recommendation method based on real-time context of a library. Background Art

[0002] With the rapid development of information technology and the popularization of digital libraries, the number and types of book resources have shown an exponential growth. Libraries not only include various professional books, journals and academic papers, but also continuously update and introduce a variety of digital resources, such as e-books, audio books, video tutorials, etc. The accumulation of such a huge amount of information makes users face tremendous pressure to screen information when looking for specific resources. How to help readers find content that meets their needs in a rich resource library has become an urgent problem to be solved.

[0003] Traditional book recommendation systems mainly rely on collaborative filtering or content matching technology. Collaborative filtering technology recommends resources by finding the reading records of other readers with similar interests to the user, while content matching methods recommend resources based on the similarity between the user's past borrowed content and the content of the book resources. These methods can meet the basic needs of users to a certain extent, but with the diversification of readers' needs and the increasing richness of library resources, traditional recommendation methods are prone to generalization and repetitiveness of recommended content, and cannot automatically respond to users' personalized needs. Summary of the invention

[0004] In view of the above defects or improvement needs of the prior art, the present invention provides a personalized recommendation system and recommendation method based on real-time library context, which aims to dynamically respond to the personalized needs of users and recommend books that meet their personalized needs.

[0005] To achieve the above object, the present invention provides a personalized recommendation system based on real-time context of the library, which comprises:

[0006] A user attribute fusion module is used to construct the attribute features of the current user and input them into the feature weighted fusion network unit to obtain attribute fusion features; the attributes include the user's gender and age;

[0007] A context information fusion module is used to construct context information features of the current library and input them into a feature weighted fusion network unit to obtain context fusion features; the context information includes the geographical location of the current library, the current access time and the current environment;

[0008] The user preference identification module is used to construct the representation of candidate books and the representation of historical books that the current user has consulted, calculate the similar feature vectors of the candidate book representation and the historical book representation, and concatenate the candidate book representation, the historical book representation and the similar feature vectors of the two, and then pass them through the activation layer and the linear layer in turn to obtain the weights of the corresponding historical books. After weighting all the historical book representations, they are input into the pooling layer to obtain the reading preference features corresponding to the candidate books;

[0009] A recommendation probability calculation module is used to concatenate the attribute fusion features, context fusion features, reading preference features and candidate book representations and then pass them through a linear layer, an activation layer and a fully connected layer in sequence to obtain the recommendation probability of the candidate book;

[0010] The feature weighted fusion network unit is used to perform pairwise interaction on the input features to obtain interactive features, and each interactive feature is input into the linear layer and the activation layer in turn to obtain the weight of the corresponding interactive feature, and all interactive features are weighted and summed to achieve feature fusion.

[0011] Optionally, the weight of the historical book and / or the weight of the interactive feature is a normalized weight.

[0012] Optionally, the attributes also include education level, profession and interests and hobbies.

[0013] Optionally, the context information also includes the subject area where the user is located in the library.

[0014] Optionally, the representation of each book includes a vector formed by concatenating the book ID, the title text, and the encoding of the book category.

[0015] Optionally, similar feature vectors of the candidate book representation and the historical book representation are obtained by calculating the outer product of the candidate book representation and the historical book representation.

[0016] Optionally, for feature v i x i and feature v j x j Perform two-by-two interactions to obtain the interactive features f ij The calculation formula is:

[0017] f ij ={(v i ⊙v j )x i x j}

[0018] In the formula, v i is the embedding vector of the i-th attribute, x i is the value of the i-th attribute, v j is the embedding vector of the jth attribute, xj is the value of the jth attribute.

[0019] The present invention also provides a personalized recommendation method based on real-time context of the library, which comprises:

[0020] Acquire input data, wherein the input data includes attributes of the current user, context information of the current library, candidate books, and historical books that the current user has consulted;

[0021] Inputting the input data into the trained personalized recommendation system based on real-time library context as described in any one of the above items to obtain the recommendation probability of each candidate book;

[0022] The candidate books are sorted according to the size of the recommendation probability, and the greater the probability, the higher the ranking.

[0023] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are implemented.

[0024] The present invention also provides a computer program product, comprising a computer program or instructions, which implement the steps of the method described above when executed by a processor.

[0025] In general, compared with the prior art, the above technical solutions conceived by the present invention mainly have the following beneficial effects:

[0026] 1. The present invention proposes a personalized recommendation system based on real-time context of the library, which includes a user attribute fusion module, a context information fusion module, a user preference identification module and a recommendation probability calculation module. The final splicing vector is obtained by splicing attribute fusion features, context fusion features, reading preference features and candidate book representations. The vector not only contains the user's long-term reading preference, but also integrates the specific interests in the current context, thereby more accurately expressing the user's overall preference for candidate books. This ensures that the recommendation system can not only capture the user's long-term preferences and personality characteristics, but also provide users with more demand-oriented recommendation content in real-time contexts, thereby generating highly personalized recommendation results;

[0027] 2. In the present invention, both the user attribute fusion module and the context information fusion module effectively capture the potential relationship between high-dimensional features through feature crossover, thereby making full use of multi-dimensional information, and then automatically learning weights according to the interactive features, giving different weights to different interactive features, thereby distinguishing the relative importance of multi-dimensional attributes, thereby more accurately identifying and distinguishing core features;

[0028] 3. The present invention introduces learnable linear layers and activation layers when obtaining the weights of user and context features. Compared with the direct multiplication of QKV matrices in the traditional Transformer encoder, the number of parameters is only one-tenth under the standard model, and the online reasoning efficiency is greatly improved while maintaining similar reasoning effects. It is in line with the scenario of real-time recommendation based on libraries and can be deployed in offline hosts with higher real-time requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a structural diagram of a personalized recommendation system in one embodiment of the present invention;

[0030] Figure 2 is a data processing flow chart of a feature weighted fusion network unit in one embodiment of the present invention;

[0031] Figure 3 4 is a diagram of the data processing process of a user preference identification module in one embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0033] Example 1

[0034] The present invention provides a personalized recommendation system based on real-time library context, such as Figure 1 The figure shows a structural diagram of a personalized recommendation system in an embodiment of the present invention, which includes a user attribute fusion module, a context information fusion module, a user preference identification module and a recommendation probability calculation module. Each module is introduced in detail below.

[0035] The user attribute fusion module is used to construct the attribute features of the current user and input them into the feature weighted fusion network unit to obtain the attribute fusion features. The attributes include the user's gender and age.

[0036] Specifically, the user attribute fusion module includes a user attribute feature construction unit and a feature weighted fusion network unit connected in sequence. The user attribute feature construction unit is used to construct the current user attribute feature (embedded representation) according to the attribute information of the current user and input it into the subsequent feature weighted fusion network unit for feature fusion. The feature weighted fusion network unit is used to perform two-by-two interactions on the input features to obtain interactive features, and each of the interactive features is respectively input into the linear layer and the activation layer in sequence to obtain the weight of the corresponding interactive feature, and all interactive features are weighted summed to achieve feature fusion.

[0037] like Figure 2 The data processing flow chart of the feature weighted fusion network unit in one embodiment of the present invention is shown, and the feature weighted fusion network unit includes an interaction unit, a weight recognition network and a weighted summation unit. The interaction unit is used to perform two-by-two interactions on the features output by the feature construction unit to obtain interactive features, and the weight network unit includes a linear layer and an activation layer. Each interactive feature output by the interaction unit passes through the linear layer and the activation layer in turn to obtain the weight of the corresponding interactive feature. The summation unit is used to perform weighted summation on each interactive feature to achieve feature fusion.

[0038] Specifically, for feature v i x i and feature vjx j Perform two-by-two interactions to obtain the interactive features f ij The calculation formula is: ij ={(v i ⊙v j )x i x j}, where v i is the embedding vector of the i-th attribute, x i is the value of the i-th attribute, v j is the embedding vector of the jth attribute, x j is the value of the jth attribute.

[0039] In a specific embodiment, in addition to gender and age, attributes may also include education level, major, interests and hobbies, etc., and the above information can be read in the library cloud data warehouse.

[0040] Traditional recommendation systems usually only distinguish user characteristics by item ID, and fail to make full use of users' multi-dimensional attribute information, such as gender, age, education, profession, and hobbies. This method is difficult to accurately portray users' multi-faceted preferences and personalized needs. In fact, the user's attribute characteristics contain rich information (such as gender, age, education, profession, and hobbies, etc.). Simply relying on ID distinction ignores these details and cannot provide a comprehensive user portrait for the recommendation system. To this end, the present invention introduces a user attribute fusion module, which effectively captures the potential relationship between high-dimensional features through feature cross-linking, thereby making full use of multi-dimensional information, and then automatically learning weights according to interactive features, giving different weights to different interactive features, thereby distinguishing the relative importance of multi-dimensional attributes. For example, the interaction between the user's age and hobbies may reflect their preferences better than other feature combinations, while the interactive effect of gender and education is relatively weak. Through this method, the model can more accurately identify and distinguish core features. In addition, the weight recognition network introduces learnable linear layers and activation layers, which only require one-tenth of the parameters under the standard model, and greatly improves the online reasoning efficiency while maintaining similar reasoning effects. It is suitable for scenarios based on real-time library recommendations and can be deployed in offline hosts with higher real-time requirements.

[0041] by Figure 2 Take this as an example. The input data of user attributes is X = {x 1 ,x 2 ,x 3 ,x 4 ,x 5 ,x 6}, each value represents the numerical value of a different attribute. For example, if the attribute is a categorical attribute, such as education level, the values ​​of primary school, junior high school, high school, and university can be 0.2, 0.4, 0.6, and 0.8 respectively; if the attribute is missing, enter 0 directly; for example, if the gender is male and the gender is female, the value is 1 if it belongs to the category, and 0 if it does not belong to the category; if the attribute is a numerical attribute, such as age, the value can be the age value or the normalized age value. The user attribute feature construction unit obtains the embedded representation of the attribute based on the input attribute data ε = {v i x i} i∈x , v i is the trainable embedding representation of the corresponding attribute, x i is the value of the corresponding attribute. After the interaction unit crosses the embedded vectors two by two, the interaction feature f is obtained. ij ={(v i ⊙v j )x i x j}. The interaction feature f ijInput the trainable linear layer and activation layer in sequence to obtain the interaction feature f ij The summation unit performs weighted summation on each interactive feature to achieve feature fusion.

[0042] In one embodiment, the activation layer selects the ReLU activation function, the weight is the normalized weight, and the weight a ij The acquisition process can be expressed as:

[0043] a' ij =hT T ReLU(W(v i ⊙v j )x i x j +b)

[0044]

[0045] In the formula, h T is a learnable weight vector. For example, the output of the ReLU activation layer is a 4*1 vector, so h T It is a 1*4 learnable parameter vector, which is used to transform the dimension into 1*1. The calculated result a' ij is a scalar value. W and b are trainable vectors and biases in the linear layer. is an unordered set of combinations of i and j.

[0046] The context information fusion module is used to construct the context information features of the current library and input the feature weighted fusion network unit to obtain the context fusion features. The context information includes the geographical location of the current library, the current access time and the current environment.

[0047] Specifically, the context information fusion module includes a context information feature construction unit and a feature weighted fusion network unit connected in sequence. The context information feature construction unit is used to construct the context information feature (embedded representation) of the current library according to the context information of the current library and input it into the subsequent feature weighted fusion network unit for feature fusion. The feature weighted fusion network unit is used to perform two-by-two interactions on the input features to obtain interactive features, and each of the interactive features is respectively input into the linear layer and the activation layer in sequence to obtain the weight of the corresponding interactive feature, and all interactive features are weighted summed to achieve feature fusion.

[0048] In a specific embodiment, the context information may include information such as the subject area where the user is located in the library in addition to the current geographical location of the library, the current access time and the current environment.

[0049] Specifically, the feature weighted fusion network unit in the context information fusion module can refer to the introduction of the feature weighted fusion network unit in the user attribute fusion module, which will not be repeated here.

[0050] In the invention, in addition to focusing on the long-term attributes of users, real-time contextual information is also introduced. This contextual information can help the recommendation system generate more personalized recommendation content in the specific context of the user's current location.

[0051] In this embodiment, the context information can be divided into the following three categories.

[0052] The first category is the current library's geographical location and current visit time, including the user's city, library name, season, month, and specific visit time. This type of macro information not only provides a geographical and temporal context for recommendations, but also can influence the user's borrowing preferences from a macro perspective. For example, in winter, users may be more inclined to borrow literature and philosophy books, while during specific holidays, travel or relaxing reading books may be more popular. This type of time and location information is encoded as a macro-context feature of the user, which helps the model better understand the changes in the user's reading needs in different seasons and geographical environments.

[0053] The second category is the current environment, such as weather, temperature, and air humidity. These external situations have a significant impact on users' immediate reading tendencies. On sunny and warm days, users may tend to choose more relaxing reading content, while in colder or quieter environments, users may prefer in-depth reading. These objective factors, as situational characteristics, affect users' real-time reading tendencies.

[0054] The third category is the user's real-time reading and retrieval behavior, including the content of the book currently viewed by the user, the search keywords, and the physical area where the user is located in the library (such as the literature area, science and technology area, etc.). This type of information is crucial in the context of book recommendations because it can instantly reflect the user's current interests and needs. Specifically, real-time reading and retrieval behavior directly reveals the user's current reading tendency and interest focus. For example, when a user is browsing the content of a book in a certain field or entering a specific search keyword, the system can instantly capture these dynamic signals and understand the user's current interest direction. This kind of instant information can not only make up for the shortcomings of modeling users' long-term and short-term interests in traditional recommendation systems, but also quickly adjust the recommendation strategy when user behavior changes. By real-time monitoring and analyzing users' reading and retrieval behaviors, the recommendation system can capture these changes in real time and provide recommendation results that are more in line with users' immediate needs. In the context of library recommendations, real-time behavior information has significantly enhanced its representation ability in the model after being processed with extended dimensions. Taking these real-time features as the main input, the system can accurately model users' short-term interest dynamics. For example, if a user is searching for science and technology books or frequently browsing a certain subject, the recommendation system can respond immediately and prioritize highly relevant resources, thereby improving user satisfaction and the relevance of recommendations. This is especially true for the physical area where the user is in the library. This is the biggest difference between traditional e-commerce recommendations and real-time library recommendations. Traditional e-commerce recommendations cannot utilize this kind of contextual information, but can only make recommendations based on user and interaction history.

[0055] The user preference identification module is used to construct candidate book representations and historical book representations that the current user has consulted, calculate the similar feature vectors of the candidate book representations and the historical book representations, and concatenate the candidate book representations, historical book representations, and the similar feature vectors of the two, and then pass them through the activation layer and the linear layer in turn to obtain the weights of the corresponding historical books. After weighting all historical book representations, they are input into the pooling layer to obtain the reading preference features corresponding to the candidate books.

[0056] like Figure 3 The data processing diagram of the user preference identification module in one embodiment of the present invention is shown. First, the book information (such as book ID, book title and category) in the user interaction history is spliced ​​together after coding mapping to form a historical sample and form a user's reading behavior sequence. Correspondingly, these features of the candidate books are also generated through the same processing flow to compare and associate with the historical data for consistency.

[0057] To determine the probability of the user's interest in the current candidate book, in this embodiment, a dot product operation is performed on each book in the historical sequence and the candidate book, and then concatenation, activation, and linear mapping are carried out. In this way, the matching degree (weight) between each historical book and the candidate book can be calculated item by item, which is used to highlight the correlation between the user and the candidate book. In one embodiment, the similarity feature vector of the candidate book representation and the historical book representation is obtained by calculating the outer product of the candidate book representation and the historical book representation. For example, the book ID, title text, and encoding of the book category are respectively represented as e ID , e text , e cate . Then there is the historical book representation: e k = concat(e ID , e text , e cate ). Let the representation of the candidate book be V U (A). Then the formula for calculating the similarity weight between the candidate book V U (A) and any historical interaction book e k is: Subsequently, all historical book representations are weighted and then input into the pooling layer to obtain the reading preference features corresponding to the candidate book.

[0058] The recommendation probability calculation module is used to concatenate the attribute fusion feature, context fusion feature, reading preference feature, and candidate book representation, and then sequentially pass through a linear layer, an activation layer, and a fully connected layer to obtain the recommendation probability of the candidate book.

[0059] In a specific embodiment, the activation layer in the recommendation probability calculation module selects the PReLU activation function or the dice activation function. Since the quality of the book data is inconsistent, when the feature values input in the data set are relatively sparse, dice can be considered. If the data set quality is high and the missing features are few, then PRelu is adopted.

[0060] In the present invention, by concatenating the attribute fusion feature, context fusion feature, reading preference feature, and candidate book representation, the final concatenated vector is obtained. This vector not only contains the user's long-term reading preference but also integrates the specific interests in the current context, thus more accurately expressing the user's overall preference for the candidate book. In this way, it is ensured that the recommendation system can not only capture the user's long-term preference and personality characteristics but also provide more personalized recommendation content for the user in real-time situations. By further calculating the final output vector, the preference probability of the user for the candidate book is obtained. By comparing the recommendation probabilities of all candidate books, the candidate books can be sorted. The higher the recommendation probability, the higher the ranking.

[0061] In general, the present invention extracts features from user information, fully exploits multidimensional attributes and contextual features, and then fuses user representations, historical interaction records, real-time contextual features, and candidate items to finally output the recommendation probability of the target book, thereby significantly improving the relevance and real-time nature of the recommendation, and generating highly personalized recommendation results.

[0062] Example 2

[0063] The present invention also relates to a personalized recommendation method based on real-time library context, which comprises:

[0064] Acquire input data, wherein the input data includes attributes of the current user, context information of the current library, candidate books, and historical books that the current user has consulted;

[0065] Inputting the input data into a trained personalized recommendation system based on real-time library context to obtain a recommendation probability for each candidate book;

[0066] The candidate books are sorted according to the size of the recommendation probability, and the greater the probability, the higher the ranking.

[0067] The personalized recommendation system is the system introduced in Example 1. After training, the system can quickly output the recommendation probability of each candidate book through the processing of each module after obtaining the input data, thereby realizing the recommendation of books. It can be understood that the recommendation system can be trained by traditional network training methods, and the training samples can be the book selection records of historical users.

[0068] Example 3

[0069] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, and the computer program implements the steps of the above method when executed by a processor.

[0070] Specifically, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, an internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0071] Example 4

[0072] The embodiment of the present invention provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of the method of the above embodiment of the present invention.

[0073] The technical features of the above-mentioned embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. It should be noted that "in one embodiment", "for example", "for example", etc. of the present invention are intended to illustrate the present invention, rather than to limit the present invention.

[0074] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent application. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A personalized recommendation system based on real-time library context, characterized in that: include: User attribute fusion module, used to construct the attribute features of the current user and input them into the feature weighted fusion network unit to obtain attribute fusion features; The attributes include the user's gender and age; A context information fusion module is used to construct context information features of the current library and input them into a feature weighted fusion network unit to obtain context fusion features; the context information includes the geographical location of the current library, the current access time and the current environment; The user preference identification module is used to construct the representation of candidate books and the representation of historical books that the current user has consulted, calculate the similar feature vectors of the candidate book representation and the historical book representation, and concatenate the candidate book representation, the historical book representation and the similar feature vectors of the two, and then pass them through the activation layer and the linear layer in turn to obtain the weights of the corresponding historical books. After weighting all the historical book representations, they are input into the pooling layer to obtain the reading preference features corresponding to the candidate books; A recommendation probability calculation module is used to concatenate the attribute fusion features, context fusion features, reading preference features and candidate book representations and then pass them through a linear layer, an activation layer and a fully connected layer in sequence to obtain the recommendation probability of the candidate book; The feature weighted fusion network unit is used to perform pairwise interaction on the input features to obtain interaction features, and each interaction feature is input into the linear layer and the activation layer in turn to obtain the weight of the corresponding interaction feature, and all interaction features are weighted and summed to achieve feature fusion.

2. The personalized recommendation system based on real-time library context as claimed in claim 1, characterized in that: The weight of the historical book and / or the weight of the interactive feature are normalized weights.

3. The personalized recommendation system based on real-time library context as claimed in claim 1, characterized in that: The attributes also include education, profession, and interests and hobbies.

4. The personalized recommendation system based on real-time library context as claimed in claim 1, characterized in that: The context information also includes the subject area where the user is located in the library.

5. The personalized recommendation system based on real-time library context according to any one of claims 1 to 4, characterized in that: The representation of each book includes a vector concatenated from the book ID, title text, and encoding of the book category.

6. The personalized recommendation system based on real-time library context according to any one of claims 1 to 4, characterized in that: The similar feature vectors of the candidate book representation and the historical book representation are obtained by calculating the outer product of the candidate book representation and the historical book representation.

7. The personalized recommendation system based on real-time library context according to any one of claims 1 to 4, characterized in that: For feature v i x i and feature v j x j Perform two-by-two interactions to obtain the interactive features f ij The calculation formula is: f ij ={(v i ⊙v j )x i x j } In the formula, v i is the embedding vector of the i-th attribute, x i is the value of the i-th attribute, v j is the embedding vector of the jth attribute, x j is the value of the jth attribute.

8. A personalized recommendation method based on real-time library context, characterized in that: include: Acquire input data, wherein the input data includes attributes of the current user, context information of the current library, candidate books, and historical books that the current user has consulted; Inputting the input data into the trained personalized recommendation system based on library real-time context as claimed in any one of claims 1 to 7 to obtain the recommendation probability of each candidate book; The candidate books are sorted according to the size of the recommendation probability, and the greater the probability, the higher the ranking.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to claim 8 are implemented.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to claim 8 are implemented.

Images