Short video recommendation method and system based on large language model

Abstract

The present invention provides a short video recommendation method, system, storage medium and electronic device based on a large language model, and relates to the field of machine learning. The present invention generates a large language model, an interest description large language model and a breaking circle strategy description large language model by integrating video content, and designs a recommendation system architecture that assists users in breaking the circle. First, accurately identify the multi-dimensional characteristics of the video and the user's interest preferences, and divide different interest communities; secondly, implement a dual-track video recommendation strategy to obtain short video recommendation lists based on interest and auxiliary breaking the circle respectively; then, merge the two short video recommendation lists, and according to the target user's breaking circle willingness and the manager's decision score, adjust the proportion of breaking circle short videos in the recommendation, generate a final short video recommendation list, and explore potential short video content of interest. Further, generate a large language model based on the title, design the title of the breaking circle short video, and stimulate the target user's desire to click and viewing interest.

Description

Technical Field

[0001] The present invention relates to the field of machine learning, and in particular to a short video recommendation method, system, storage medium, and electronic device based on a large language model. Background Art

[0002] In today's digital age, short videos, with their real-time interactivity, wide user participation and instant content dissemination, have become a core medium for the public to express themselves, enjoy entertainment and obtain diverse information.

[0003] With the rapid development of the short video industry and the continuous expansion of its user base, information overload is becoming increasingly serious. The massive amount of short video content and diverse user needs make filtering information that matches user interests a key issue. Existing personalized recommendation algorithms have addressed this issue to a certain extent, improving the user experience. However, this also presents the risk of information homogeneity. While users enjoy the convenience of personalized recommendations, they often become trapped in a rigid information environment, limiting the breadth and depth of their information acquisition.

[0004] Therefore, how to balance the relationship between personalized recommendations and diversified information has become an important topic that requires in-depth thinking and exploration. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a short video recommendation method, system, storage medium and electronic device based on a large language model, which solves the technical problem of how to balance the relationship between personalized recommendations and diversified information.

[0007] (2) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A short video recommendation method based on a large language model, comprising:

[0010] Obtain a collection of short videos and a user set containing the target user;

[0011] A pre-trained BERT-based short video content generation language model is used to generate a description text for each short video. Keywords in the description text are extracted using the TF-IDF algorithm to obtain several tags for each short video. After optimization and merging, the tag set for the short video set is obtained.

[0012] Count the historical short videos of each user's interest within a preset historical time period, combine the tag set to obtain several interest tags of each user, and use the K-means algorithm to perform clustering to determine the interest community to which each user belongs;

[0013] Compare the similarity between the target user's interest tags and the tags of any short video to generate a first short video recommendation list based on the user's personal interests; use a collaborative filtering method based on neighboring users in the interest community to generate a second short video recommendation list based on the interest community; use a Boosting algorithm to fuse the first and second short video recommendation lists to obtain a third short video recommendation list based on interest;

[0014] Locate the communities closest to and farthest from the target user's interest community, and construct corresponding candidate short video recommendation lists; merge the candidate short video recommendation lists of the closest and farthest communities, and allocate the number of recommended short videos based on distance to obtain a fourth short video recommendation list to assist in breaking the circle;

[0015] Based on the target user's willingness to break the circle and the manager's decision score, the proportion of the fourth short video recommendation list is calculated, and combined with the third short video recommendation list to obtain the final short video recommendation list; wherein the pre-trained interest description large language model based on the Transformer deep learning architecture and the breaking circle strategy description large language model are used, combined with the evaluation results given by the manager to obtain the manager's decision score.

[0016] Preferably, a pre-trained title generation large language model based on the Transformer deep learning architecture is used to generate the title of each short video in the short video recommendation list originating from the fourth short video recommendation list.

[0017] Preferably, the method comprises generating a description text of each short video using a pre-trained BERT-based short video content generation large language model, extracting keywords from the description text using a TF-IDF algorithm to obtain a plurality of tags for each short video, and optimizing and merging the tags to obtain a tag set for the short video set; comprising:

[0018] For any short video, a motion detection algorithm is used to decode the continuous video stream into a series of key frames F o , where o represents the index of the key frame;

[0019] Calculate each key frame F o The color histogram defines the color space and color channels, counts the number of pixels of each color value in each color channel, and forms the color histogram feature vector v color,o ;

[0020] The gray-level co-occurrence matrix algorithm is used to extract the texture feature vector v by statistically analyzing the joint probability distribution of pixel pairs in the gray-level image at the preset direction and distance. texture,o ;

[0021] Use the Canny edge detection algorithm to detect the edges in the image, and combine it with the Hough transform to extract the shape features to obtain the shape feature set S o , and converted into shape feature vector v shape,o ;

[0022] Extract the audio features of the short video and convert the dubbing in the short video into a text sequence T audio ;

[0023] Fusion visual feature vector set {v color,o ,v texture,o ,v shape,o} and the converted text sequence T audio , use the pre-trained BERT-based short video content to generate a large language model for encoding, generate a hidden state vector, and generate a descriptive text D through a fully connected layer or classifier;

[0024] Preprocessing the descriptive text D, extracting keywords from the preprocessed descriptive text D using a TF-IDF algorithm, and using the extracted keywords as candidate tags;

[0025] Based on the semantic relationship between the candidate tags, a machine learning algorithm is used to optimize the tags and obtain several tags {l1, l2, ..., l m}, where m represents the total number of tags of the short video;

[0026] Merge the labels of all short videos to obtain the label set L of the short video set total .

[0027] Preferably, the method comprises: collecting statistics of each user's historical short video collection of interest within a preset historical time period, combining the tag set to obtain a number of interest tags for each user, and clustering using a K-means algorithm to determine the interest community to which each user belongs; including:

[0028] The user set is defined as U = {u1,u2,…,u n}, where n represents the total number of users; define user u i The collection of historically interesting short videos within the preset historical time period is where i m Represents user u i The total number of historical short videos of interest; for each short video v j ∈V i , define the tag set of the short video as L total A tag set representing the short video set;

[0029] For user u i , build the corresponding label count dictionary

[0030]

[0031] in, Indicates that label l is in user u i History of Interesting Short Video Collection V i The number of times it appears in ; I represents the indicator function, when 1 when it is, otherwise 0;

[0032] Select the tag whose number of occurrences is greater than the preset threshold as the user u i Merge all users’ interest tags and remove duplicates to obtain the user interest tag set L top ;

[0033] Clustering using the K-means algorithm:

[0034]

[0035] Among them, SSE represents the sum of squares of distances from all users to their corresponding cluster centers; K represents the set number of clusters; C c represents the set of users in the cth cluster; μ c is the center of the cth cluster;

[0036] Taking minimizing SSE as the optimization goal, iteratively update the cluster center and reallocate users to the nearest cluster center to determine user u i Interest group C pi .

[0037] Preferably, the step of comparing the similarity between the target user's interest tag and the tag of any short video to generate a first short video recommendation list based on the user's personal interests includes:

[0038] For target users Define a feature vector If and only if L top When {k} is the user's interest tag, Equal to 1; for short video v j , define its label as If and only if L top When {k} is the tag of the video, Equal to 1; where |L to[ | represents the number of tags in the user’s interest tag set, Ltop {k} represents the kth tag in the user's interest tag set;

[0039] The following formula is used to compare the similarity between the target user's interest tags and the tags of any short video:

[0040]

[0041] Where sim represents the similarity function, · represents the dot product, and ||·|| represents the Euclidean length of the vector;

[0042] By comparing the similarities, we find a collection of videos that are similar to the target user’s interests and sort them according to the similarity scores to generate the first short video recommendation list V based on the user’s personal interests. CBR .

[0043] Preferably, the method of generating a second short video recommendation list based on the interest community to which the target user belongs by using a collaborative filtering method based on neighboring users includes:

[0044] Calculate the similarity between users and determine the target users Neighbor user set of

[0045] Utilize the short videos v that each neighbor user in the interest community to which the target user belongs has watched or rated p The target users are predicted by weighted average method based on the data of Potential interest in unrated short videos; calculated as follows:

[0046]

[0047] in, Indicates target users Short video v p The predicted score of the target user is N(u), which represents the interest community to which the target user belongs. u′_vp Represents neighbor user u′’s response to short video v p The actual rating of

[0048] Sort the short videos based on the predicted scores, and select the short videos with the highest scores as the second short video recommendation list V based on the interest community. UCF .

[0049] Preferably, the process of obtaining the target user's willingness to break the circle includes:

[0050] For target users Interest groups Sort the interesting tags of the cluster center, and define the first several tags with the most occurrences as the mainstream tags of the community; define Function, calculate target users The proportion of videos not related to the mainstream tags of the community that were watched during the preset historical period is used as a measure of the target users. An indicator of personal breaking circle activity;

[0051] Combined with the Jaccard coefficient, define function, as a measure of target users Interest groups Indicators of the tendency of a community to break out of its circle;

[0052] Among them, the target users Interest groups The number of mainstream tags is used as the base number; interest communities will be counted in two adjacent preset historical time periods. The change of tags of watching short videos is taken as the change number; The function is the ratio of the difference between the base number and the variation to the sum of the base number and the variation;

[0053] Define the target user The breaking circle willingness score function:

[0054]

[0055] in, is a value between 0 and 1; α, β represent the corresponding weight coefficients, and α+β=1, α, β∈(0,1).

[0056] Preferably, the process of obtaining the manager decision score includes:

[0057] Using the tags of the third short video recommendation list as input to the interest description large language model to generate a text description of the community interest; and using the tags of the fourth short video recommendation list as input to the breaking circle strategy description large language model to generate a text description of the breaking circle strategy;

[0058] Based on the text description of the community interest and the text description of the breaking circle strategy, combined with the evaluation results given by the manager, the manager's decision score is defined as S decide , the value is between 0 and 1.

[0059] Preferably, based on the target user's willingness to break the circle and the manager's decision score, the proportion of the fourth short video recommendation list is calculated, and combined with the third short video recommendation list to obtain the final short video recommendation list; including:

[0060] Calculate the short video recommendation list V Recommend The number of short videos is:

[0061] NUM(V last )=γ1NUM(V UCF )+γ2NUM(V breakout )

[0062]

[0063] Among them, γ1, γ2 represent the corresponding weight coefficients, and γ1+γ2=1, γ1, γ2∈(0,1); NUM(V UCF ),NUM(V breakout ) represent the third short video recommendation list V UCF And the fourth short video recommendation list V breakout The number of short videos;

[0064] Select V according to the ratio UCF and V breakout The short videos with the highest to lowest ratings are combined to obtain the final recommendation list V Recommend .

[0065] A short video recommendation system based on a large language model, including:

[0066] An acquisition module is used to obtain a short video set and a user set containing target users;

[0067] The recognition module is used to generate a description text for each short video using a pre-trained BERT-based short video content generation large language model, and extract keywords from the description text using the TF-IDF algorithm to obtain multiple tags for each short video, and then optimize and merge them to obtain a tag set for the short video set;

[0068] and for counting the historical short video sets of interest to each user within a preset historical time period, combining the tag set to obtain a number of interest tags for each user, and clustering using the K-means algorithm to determine the interest community to which each user belongs;

[0069] The recommendation module is used to compare the similarity between the target user's interest tags and the tags of any short video to generate a first short video recommendation list based on the user's personal interests; in the interest community to which the target user belongs, a collaborative filtering method based on neighboring users is used to generate a second short video recommendation list based on the interest community; and a Boosting algorithm is used to fuse the first and second short video recommendation lists to obtain a third short video recommendation list based on interest;

[0070] and for respectively locating the communities closest to and farthest from the interest community to which the target user belongs, and constructing corresponding candidate short video recommendation lists; fusing the candidate short video recommendation lists of the nearest and farthest communities, and allocating the number of recommended short videos based on distance, to obtain a fourth short video recommendation list that assists in breaking the circle;

[0071] The optimization module is used to calculate the proportion of the fourth short video recommendation list based on the target user's willingness to break the circle and the manager's decision score, and combine it with the third short video recommendation list to obtain the final short video recommendation list; wherein the pre-trained interest description large language model and breaking the circle strategy description large language model based on the Transformer deep learning architecture are used, combined with the evaluation results given by the manager to obtain the manager's decision score.

[0072] A storage medium stores a computer program for short video recommendation based on a large language model, wherein the computer program enables a computer to execute the short video recommendation method as described above.

[0073] An electronic device, comprising:

[0074] One or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include a program for executing the short video recommendation method as described above.

[0075] (3) Beneficial effects

[0076] The present invention provides a short video recommendation method, system, storage medium, and electronic device based on a large language model. Compared with the existing technology, it has the following advantages:

[0077] The present invention integrates video content to generate a large language model, an interest description large language model, and a breaking circle strategy description large language model, and designs a recommendation system architecture that assists users in breaking the circle. First, accurately identify the multi-dimensional characteristics of the video and the user's interest preferences, and divide the users into different interest communities; second, implement a dual-track video recommendation strategy to obtain short video recommendation lists based on interests and auxiliary breaking circles respectively; then, merge the two short video recommendation lists, and adjust the proportion of breaking circle short videos in the recommendation according to the target user's breaking circle willingness and the manager's decision score, generate the final short video recommendation list, and explore potential short video content of interest. The present invention provides high-quality and diversified short video recommendations to target users while realizing personalized recommendations, which helps them break out of the information cocoon and reach a wider world. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0079] Figure 1 This is a diagram of the architecture of a recommendation system for assisting users in breaking through social circles, provided by an embodiment of the present invention;

[0080] Figure 2 A flowchart of a short video recommendation method based on a large language model provided by an embodiment of the present invention;

[0081] Figure 3 A flowchart of another short video recommendation method based on a large language model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0082] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0083] The embodiments of the present application solve the technical problem of how to balance the relationship between personalized recommendations and diversified information by providing a short video recommendation method, system, storage medium and electronic device based on a large language model.

[0084] The technical solution in the embodiments of the present application is to solve the above technical problems, and the overall idea is as follows:

[0085] Large language models (LLMs) are based on the Transformer architecture

[48] , are deeply trained on massive amounts of text data, and rely on a large parameter set to build models. In recent years, with the improvement of computing power and the continuous innovation of algorithms, research on large language models has made significant progress worldwide. Well-known foreign models such as ChatGPT and LLaMA, as well as domestic models such as Wenxin Yiyan ERNIE bot and PanGu, have demonstrated strong performance in practical applications.

[0086] In this context, the embodiment of the present invention uses large language model technology to design a recommendation system architecture that helps users break through the circle (break through the original solidified information circle), such as Figure 1As shown, it is hoped that through technical means, users can explore diversified information; specifically:

[0087] Step 1: Interest Identification

[0088] In interest recognition, the primary task is to accurately identify the multi-dimensional features of a video and user interests. Visual features are extracted from the video and converted into audio features, achieving comprehensive feature integration. Subsequently, a large language model based on BERT is introduced for short video content generation. Leveraging its powerful multimodal data processing capabilities, it deeply analyzes and integrates multiple visual elements and audio text to generate detailed video descriptions, providing solid textual support for the subsequent development of video tags. By optimizing TF-IDF technology and machine learning algorithms, an efficient and deduplicated video tag set is constructed to support diverse video application requirements.

[0089] Based on video feature recognition and user interest analysis, we further construct user profiles. Through data analysis and clustering techniques, we divide users into interest groups and extract unique interest tags for each group. This process not only helps us gain a deeper understanding of user interests and preferences, but also provides strong support for more personalized video recommendations and content distribution.

[0090] Step 2: Video Recommendation

[0091] Based on video tags and user characteristics, a dual-track video recommendation strategy is designed, closely centered around in-depth analysis of these two factors. Firstly, to enhance recommendation accuracy, personalized recommendations are made based on interests. Based on users' existing interests, content-based and community-based collaborative filtering algorithms are used to accurately identify and recommend short videos closely related to their current interests. Secondly, to stimulate user behavior beyond their niche, recommendations for short videos that transcend niches are made.

[0092] Based on the user's interest groups, the system explores the interests of nearby and distant communities, constructs a candidate set of breakout short videos, and distributes videos based on the distance ratio between communities. This allows users to be recommended non-traditional video content that may spark their interest, even though it doesn't fall within their original interest group. Furthermore, based on the video recommendation results, large language models are trained to describe interests and breakout strategies. The former accurately captures the personalized interests of users and communities, while the latter explains the selection logic and characteristics of breakout short videos, providing a scientific basis for subsequent manager decision-making.

[0093] Step 3: Recommendation Optimization

[0094] By combining these two recommendation lists, the proportion of short videos that break through the social circle is adjusted based on the characteristics of user community-based behavior, individual willingness to break through, and the manager's strategic considerations. User willingness to break through the social circle is based on historical behavior and community dynamics, while manager decisions ensure the compliance and effectiveness of the strategy. Leveraging trained large language models describing interests and large language models describing breaking through social circle strategies, a detailed description of the hybrid strategy is generated, assisting managers in making decisions that align with market and user needs. Ultimately, by weighting the high-scoring videos, a personalized and diverse recommendation list is created.

[0095] Furthermore, during the generation phase of the recommendation list, the present invention fully leverages the creative potential and generative capabilities of the large language model. Combining the user's personalized interests and preferences with the unique appeal of the video content, a large language model for title generation is designed to tailor engaging titles for the short videos in the recommendation list that break through the social circle. These titles capture the user's focus and emotional resonance, stimulating their desire to click and watch, thereby increasing the appeal and conversion rate of the recommended content, attracting users to interact with the recommended videos, and helping them break through the social circle.

[0096] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0097] Example 1:

[0098] like Figure 2 As shown, an embodiment of the present invention provides a short video recommendation method based on a large language model, including:

[0099] S1. Obtain a short video set and a user set containing the target user;

[0100] S2. Use a pre-trained BERT-based short video content generation language model to generate a description text for each short video, and use the TF-IDF algorithm to extract keywords from the description text to obtain several tags for each short video. After optimization and merging, obtain a tag set for the short video set;

[0101] S3. Count the historical short videos of interest to each user within a preset historical time period, combine the tag set to obtain several interest tags of each user, and use the K-means algorithm to perform clustering to determine the interest community to which each user belongs;

[0102] S4. Compare the similarity between the target user's interest tag and the tag of any short video to generate a first short video recommendation list based on the user's personal interests; in the interest community to which the target user belongs, use a collaborative filtering method based on neighboring users to generate a second short video recommendation list based on the interest community; use a Boosting algorithm to fuse the first short video recommendation list and the second short video recommendation list to obtain a third short video recommendation list based on interest;

[0103] S5. Locate the communities closest to and farthest from the interest community to which the target user belongs, and construct corresponding candidate short video recommendation lists; merge the candidate short video recommendation lists of the closest and farthest communities, and allocate the number of recommended short videos based on distance to obtain a fourth short video recommendation list to assist in breaking the circle;

[0104] S6. Based on the target user's willingness to break the circle and the manager's decision score, calculate the proportion of the fourth short video recommendation list, and combine it with the third short video recommendation list to obtain the final short video recommendation list; wherein the pre-trained interest description large language model based on the Transformer deep learning architecture and the breaking circle strategy description large language model are used, combined with the evaluation results given by the manager to obtain the manager's decision score.

[0105] The embodiment of the present invention integrates video content to generate a large language model, an interest description large language model, and a breaking circle strategy description large language model, and designs a recommendation system architecture to assist users in breaking the circle. While realizing personalized recommendations, it provides target users with high-quality and diversified short video recommendations, helping them to break out of the information cocoon and access a wider world.

[0106] In an optional embodiment, if Figure 3 As shown, the embodiment of the present invention also includes:

[0107] S7. Use a pre-trained large language model for title generation based on the Transformer deep learning architecture to generate the title of each short video in the short video recommendation list that originates from the fourth short video recommendation list.

[0108] The embodiment of the present invention further integrates a large language model for title generation, and tailors attractive titles specifically for the short videos that break the circle in the recommendation list to stimulate users' desire to click and interest in watching, thereby improving the attractiveness and conversion rate of the recommended content, attracting users to interact with the recommended videos, and helping users break the circle.

[0109] Next, we will combine Figure 1 The following steps are described in detail:

[0110] In step S1, a short video set and a user set including a target user are obtained.

[0111] The embodiment of the present invention focuses on a short video platform, which has a wide user base, including individuals from different regions, ages, and professional backgrounds. In addition, users can interact with the platform by giving likes and ratings while watching short videos.

[0112] This step obtains a short video set consisting of multiple short videos from the platform, and collects short videos containing target users. The user set U={u1,u2,…,u n}, where n represents the total number of users.

[0113] In step S2, a pre-trained BERT-based short video content generation large language model is used to generate a description text for each short video, and the TF-IDF algorithm is used to extract keywords from the description text to obtain several tags for each short video. After optimization and merging, the tag set of the short video set is obtained.

[0114] It should be noted that compared to other media forms, video is unique in that it has both dynamic image and sound features. Extracting visual features from a video means extracting the features of each key frame of the video.

[0115] Accordingly, this step performs video feature recognition in the interest recognition phase, specifically including:

[0116] For any short video, a motion detection algorithm is used to decode the continuous video stream into a series of key frames F o , where o represents the index of the key frame. Calculate each key frame F o The color histogram defines the color space and color channels, counts the number of pixels of each color value in each color channel, and forms the color histogram feature vector v color,o The gray level co-occurrence matrix algorithm is used to extract the texture feature vector v by statistically analyzing the joint probability distribution of pixel pairs in the gray image at the preset direction and distance. texture,o The Canny edge detection algorithm is used to detect the edges in the image, and the shape features are extracted by combining the Hough transform to obtain the shape feature set S o , and converted into shape feature vector v shape,o Extract the audio features of the short video and convert the dubbing in the short video into a text sequence T audio .

[0117] Based on feature extraction, the fusion visual feature vector set {v color,o ,v texture,o ,v shape,o} and the converted text sequence T audio, using a pre-trained BERT-based short video content generation large language model for encoding, generating a hidden state vector, and then generating descriptive text D through a fully connected layer or classifier. The descriptive text D can include key information such as the video's theme, key events, emotional tendencies, and style. The BERT model captures the contextual relationships in the input sequence through a self-attention mechanism to generate a hidden state vector.

[0118] Preprocess the descriptive text D, including word segmentation, stop word removal, stemming or lemmatization. Then, use the TF-IDF algorithm to extract keywords from the preprocessed descriptive text D, and use the extracted keywords as candidate tags. The processing method is as follows:

[0119]

[0120] Where TF(k,D) is the frequency of word t in D, and k represents the index of word t; k TF(t,D) is the total frequency of all words in D (normalization factor); log represents the logarithmic function; |D total | is the total number of all descriptive texts D; {D′∈D total : t∈D′}| is the number of descriptive texts D containing word t.

[0121] Considering the semantic relationship between tags, avoiding the generation of semantically similar or repeated tags, using machine learning algorithms such as support vector machines to optimize tags, and obtain several tags corresponding to the short video {l1,l2,…,l m}, where m represents the total number of tags of the short video.

[0122] Merge the labels of all short videos to obtain the label set L of the short video set total ,This tag set provides rich metadata support for applications such as video retrieval, ,recommendation systems, and content classification.

[0123] In step S3, the historical short video sets of interest to each user within a preset historical time period are counted, and several interest tags of each user are obtained by combining the tag set. K-means algorithm is used for clustering to determine the interest community to which each user belongs.

[0124] This step performs the interest community division in the interest identification phase. That is, based on the video tag set generated in the previous step, users are divided into communities based on their viewing interests. Specifically, it includes:

[0125] First define user u i The collection of historically interesting short videos within the preset historical time period is For example, define user u iThe collection of videos that were watched, or "liked," within two weeks is where i m Represents user u i The total number of historically interested / liked short videos; for each short video v j ∈V i , define the tag set of the short video as L total A tag set representing the short video set.

[0126] Then extract the user interest tags. i , build the corresponding label count dictionary

[0127]

[0128] in, Indicates that label l is in user u i History of Interesting Short Video Collection V i The number of times it appears in ; I represents the indicator function, when 1 if yes, 0 otherwise.

[0129] In order to reduce the impact of the video on the user and only consider the impact of the label on the user, the label with a number of occurrences greater than the preset threshold is selected as the user u i interest tags, such as tags that appear more than 5 times.

[0130] According to the user's interest tags, all users in the user set U are divided into communities. Merge all users' interest tags and remove duplicates to obtain the user interest tag set L top .

[0131] Clustering using the K-means algorithm:

[0132]

[0133] Among them, SSE represents the sum of squares of distances from all users to their corresponding cluster centers; K represents the set number of clusters; C c represents the set of users in the cth cluster; μ c is the center of the cth cluster;

[0134] Taking minimizing SSE as the optimization goal, iteratively update the cluster center and reallocate users to the nearest cluster center to determine user u i Interest group C pi .

[0135] Finally, the clustering results can be output, all users are assigned to K clusters, and user u is defined as i The community is Cpi .

[0136] Furthermore, the labels of the cluster centers can be sorted, for example, the five labels with the highest number of occurrences can be defined as community C. p The mainstream label is (l p1 ,l p2 ,l p3 ,l p4 ,l p5 ), where l p1 ,l p2 ,l p3 ,l p4 ,l p5 ∈L total .

[0137] Based on the precise depiction of user portraits and the clear division of user communities, the embodiment of the present invention designs a dual-track video recommendation strategy. This strategy not only focuses on the core interests of target users, but also screens short video content that suits their preferences through algorithms to ensure the accuracy and satisfaction of recommendations. At the same time, it explores the interest boundaries of target users and selects those short videos that may arouse the curiosity of target users and guide them to cross the existing scope of interest. Such a recommendation mechanism is intended to promote the diversified development of target users' interests, encourage target users to try new things, thereby broaden their horizons, and enjoy a more colorful video viewing experience. For a detailed introduction, please refer to steps S4 to S5:

[0138] In step S4, the similarity between the target user's interest tag and the tag of any short video is compared to generate a first short video recommendation list based on the user's personal interest; in the interest community to which the target user belongs, a collaborative filtering method based on neighboring users is adopted to generate a second short video recommendation list based on the interest community; the Boosting algorithm is adopted to fuse the first short video recommendation list and the second short video recommendation list to obtain a third short video recommendation list based on interest.

[0139] This embodiment of the present invention recommends videos to users based on their interests. Content-based recommendation algorithms are a commonly used recommendation system algorithm that analyzes the content characteristics of short videos that a target user has liked in the past to recommend other short videos with similar interests to the target user. This algorithm primarily relies on the target user's own attributes and does not consider the behavior of other users. In this scenario, the algorithm compares the similarity between the target user's personal interest tags and video tags to recommend videos that better match their interests.

[0140] Accordingly, this step performs interest-based recommendations in the video recommendation phase, specifically including:

[0141] For target users Define a feature vector If and only if L top When {k} is the user's interest tag, Equal to 1; for short video v j , define its label as If and only if L top When {k} is the tag of the video, Equal to 1; where |L top | represents the number of tags in the user’s interest tag set, L top {k} represents the kth tag in the user's interest tag set.

[0142] The following formula is used to compare the similarity between the target user's interest tags and the tags of any short video:

[0143]

[0144] Here, sim represents the similarity function, · represents the dot product, and ||·|| represents the Euclidean length of the vector.

[0145] By comparing the similarities, we find a collection of videos that are similar to the target user’s interests and sort them according to the similarity scores to generate the first short video recommendation list V based on the user’s personal interests. CBR .

[0146] Furthermore, in the context of interest-based community division, the interest similarity of the target user within the interest community to which he belongs is higher than the similarity between different groups. In order to achieve personalized recommendation, a collaborative filtering method based on neighboring users is adopted, including:

[0147] Calculate the similarity between users and determine the target users Neighbor user set of

[0148] Utilize the short videos v that each neighbor user in the interest community to which the target user belongs has watched or rated p The target users are predicted by weighted average method based on the data of Potential interest in unrated short videos; calculated as follows:

[0149]

[0150] in, Indicates target users Short video v p The predicted score of the target user is N(u), which represents the interest community to which the target user belongs. u′ _ vp Represents neighbor user u′’s response to short video v p actual rating.

[0151] Sort the short videos based on the predicted scores, and select the short videos with the highest scores as the second short video recommendation list V based on the interest community. UCF .

[0152] Finally, the Boosting integrated learning idea is used to integrate V CBR and V UCF , the two recommendation lists are regarded as the outputs of two "weak recommenders". Based on the results of performance evaluation, a weight is assigned to each weak recommender. The weight is directly assigned based on the size of the performance index. The better the performance, the greater the weight. The prediction results of these weak recommenders are weighted and combined with simple linear weighting to finally obtain the interest-based recommendation list V normal .

[0153] In step S5, the communities closest to and farthest from the interest community to which the target user belongs are located respectively, and corresponding candidate short video recommendation lists are constructed; the candidate short video recommendation lists of the nearest and farthest communities are integrated, and the number of recommended short videos is allocated based on the distance to obtain a fourth short video recommendation list to assist in breaking the circle.

[0154] For target users In personalized recommendations, make full use of their interest groups The characteristics are designed to not only satisfy their existing interests, but also explore the boundaries of their potential interests, that is, to achieve "breaking the circle".

[0155] Accordingly, this step performs the recommendation of short videos that break through the circle in the video recommendation link. Specifically, a two-step strategy is adopted to construct a set of short videos that break through the circle, including:

[0156] First, identify the cluster centers and Recent communities The distance between the two centers is d min This community's video viewing interests are Close, but different, most likely to generate target users Haven't seen the video but liked it.

[0157] Second, in order to ensure the target users Ability to see the most comprehensive types of videos, explore the possibility of breaking the circle, and further locate the community The farthest community The distance between the two centers is d max , it and The distance between them is the farthest, the interests are the most different, and they represent the target users. New areas that may not have been touched upon may help to break the circle.

[0158] for and Build candidate short video recommendation lists V respectively min and V max The two candidate short video recommendation lists are videos with the highest to lowest ratings in the two communities.

[0159] Based on this, we build the fourth short video recommendation list V to help break the circle. breakout , the number of videos in the two candidate video sets in the new breaking circle short video set meets:

[0160] NUM(V breakout )=NUM(V min )×r+NUM(V max )×s (6)

[0161] Among them, r and s are proportional coefficients, r is the recent community recommendation to the target user The proportion of videos in the recommendation list is s, which is the farthest community will recommend to the target user. The proportion of videos in the recommendation list is r+s=1, r,s∈(0,1). This design aims to ensure that the recommendation list retains content that the target user is familiar with while also introducing enough novel elements to stimulate the target user's curiosity and desire to explore.

[0162] Considering that in k-means clustering, the distance between communities usually reflects the degree of difference in their interests or preferences. A closer distance means that the interests between communities are more similar, while a farther distance indicates a greater difference in interests. Therefore, allocating the number of recommended videos based on distance can make the recommendation list maintain a certain similarity while also introducing novelty and diversity. Therefore, let r:s = d max :d min , It can directly reflect the similarities and differences between communities, making the recommendation strategy easier to understand and implement. According to the obtained ratio, the two selected lists are merged to obtain the fourth short video recommendation list V breakout .

[0163] In step S6, based on the target user's willingness to break the circle and the manager's decision score, the proportion of the fourth short video recommendation list is calculated, and combined with the third short video recommendation list to obtain the final short video recommendation list; wherein the pre-trained interest description large language model based on the Transformer deep learning architecture and the breaking circle strategy description large language model are used, combined with the evaluation results given by the manager to obtain the manager's decision score.

[0164] This step performs the recommendation optimization phase to mix the normal recommendation list V based on user interests. normalList of videos that will help users break the circle V breakout ; Specifically include:

[0165] First, obtain the target user's willingness to break out of the circle; including:

[0166] For target users Interest groups Sort the interesting tags of the cluster center, and define the first several tags with the most occurrences as the mainstream tags of the community; define Function, calculate target users The proportion of videos not related to the mainstream tags of the community that were watched during the preset historical period is used as a measure of the target users. An indicator of personal breaking circle activity.

[0167] Combined with the Jaccard coefficient, define function, as a measure of target users Interest groups An indicator of the tendency of a community to break out of its circle.

[0168] Among them, the target users Interest groups The number of mainstream tags is used as the base number; interest communities will be counted in two adjacent preset historical time periods. The change of tags of watching short videos is taken as the change number; The function is the ratio of the difference between the base and the variation to the sum of the base and the variation.

[0169] In fact, here Essentially, the function is a coefficient of variation. This coefficient calculates the degree of similarity between each type of variation and no variation, using no variation as a baseline. A higher degree of similarity indicates a smaller variation and a lower willingness among community members to experiment with and embrace diverse content. A lower degree of similarity indicates a greater variation and a higher willingness among community members to experiment with and embrace diverse content.

[0170] The target user is defined as the target user by integrating the target user's personal breaking circle activity and the community breaking circle tendency through weighted summation. The breaking circle willingness score function:

[0171]

[0172] in, is a value between 0 and 1; α, β represent the corresponding weight coefficients, and α+β=1, α, β∈(0,1).

[0173] It can be understood that when evaluating a user's willingness to break out of the circle, the embodiment of the present invention comprehensively considers two dimensions: personal historical behavior and overall community trends.

[0174] Next, obtaining the manager's decision score includes:

[0175] The labels of the third short video recommendation list are used as input to the interest description language model to generate a text description of community interests; and the labels of the fourth short video recommendation list are used as input to the breaking circle strategy description language model to generate a text description of the breaking circle strategy.

[0176] It should be noted that the interest description large language model and the breaking circle strategy description large language model here are both trained using the Transformer deep learning architecture. Since the training process itself is an existing technology, it will not be repeated here.

[0177] Based on the text description of the community interest and the text description of the breaking circle strategy, combined with the evaluation results given by the manager, the manager's decision score is defined as S decide , the value is between 0 and 1, and is intended to intuitively reflect the manager's comprehensive ability and effectiveness in balancing the interests of all parties, grasping market dynamics and guiding platform development.

[0178] For example, the evaluation given by the manager can be based on multiple dimensions, such as the feasibility of the service for various types of users, the demand for the service in the target market, the application demand of the service from the company itself, and whether there is sufficient financial support for the service. Based on the evaluation results, a score of 0-25 is given for each dimension, and finally the total score is normalized to obtain the manager's decision score S decide .

[0179] It is understandable that when formulating a video recommendation strategy, embodiments of the present invention recognize that it is also crucial to receive evaluation results from managers to ensure that the strategy meets social and platform needs and optimizes the recommendation effect.

[0180] Furthermore, based on the target user's willingness to break the circle and the manager's decision score, the proportion of the fourth short video recommendation list is calculated, and combined with the third short video recommendation list to obtain a final short video recommendation list; including:

[0181] Calculate the short video recommendation list V Recommend The number of short videos is:

[0182] NUM(V last ))=γ1NUM(V UCF )+γ2NUM(V breakoout ) (8)

[0183]

[0184] Among them, γ1, γ2 represent the corresponding weight coefficients, and γ1+γ2=1, γ1, γ2∈(0,1); NUM(V UCF ),NUM(V breakout ) represent the third short video recommendation list V UCF And the fourth short video recommendation list V breakout The number of short videos;

[0185] Select V according to the ratio UCF and V breakout The short videos with the highest to lowest ratings are combined to obtain the final recommendation list V Recommend .

[0186] In step S7, a pre-trained large language model for title generation based on a Transformer deep learning architecture is used to generate a title for each short video in the short video recommendation list that originates from the fourth short video recommendation list.

[0187] In order to assist target users in breaking the circle to the greatest extent, this step generates a large language model based on the title for each short video in the short video recommendation list from the fourth short video recommendation list, tailors an attractive title, captures the user's interest focus and emotional resonance points, and stimulates the user's desire to click and watch interest, thereby improving the attractiveness and conversion rate of the recommended content and attracting users to interact with the recommended videos.

[0188] At this point, the embodiment of the present invention completes the entire process of the short video recommendation method based on the large language model.

[0189] Example 2:

[0190] An embodiment of the present invention provides a short video recommendation system based on a large language model, including:

[0191] An acquisition module is used to obtain a short video set and a user set containing target users;

[0192] The recognition module is used to generate a description text for each short video using a pre-trained BERT-based short video content generation large language model, and extract keywords from the description text using the TF-IDF algorithm to obtain multiple tags for each short video, and then optimize and merge them to obtain a tag set for the short video set;

[0193] and for counting the historical short video sets of interest to each user within a preset historical time period, combining the tag set to obtain a number of interest tags for each user, and clustering using the K-means algorithm to determine the interest community to which each user belongs;

[0194] The recommendation module is used to compare the similarity between the target user's interest tags and the tags of any short video to generate a first short video recommendation list based on the user's personal interests; in the interest community to which the target user belongs, a collaborative filtering method based on neighboring users is used to generate a second short video recommendation list based on the interest community; and a Boosting algorithm is used to fuse the first and second short video recommendation lists to obtain a third short video recommendation list based on interest;

[0195] and for respectively locating the communities closest to and farthest from the interest community to which the target user belongs, and constructing corresponding candidate short video recommendation lists; fusing the candidate short video recommendation lists of the nearest and farthest communities, and allocating the number of recommended short videos based on distance, to obtain a fourth short video recommendation list that assists in breaking the circle;

[0196] The optimization module is used to calculate the proportion of the fourth short video recommendation list based on the target user's willingness to break the circle and the manager's decision score, and combine it with the third short video recommendation list to obtain the final short video recommendation list; wherein the pre-trained interest description large language model and breaking the circle strategy description large language model based on the Transformer deep learning architecture are used, combined with the evaluation results given by the manager to obtain the manager's decision score.

[0197] Example 3:

[0198] An embodiment of the present invention provides a storage medium storing a computer program for recommending short videos based on a large language model, wherein the computer program enables a computer to execute the short video recommendation method as described in Example 1.

[0199] Example 4:

[0200] An embodiment of the present invention provides an electronic device, including:

[0201] One or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include a method for executing the short video recommendation method as described in Example 1.

[0202] It is understandable that the short video recommendation system, storage medium and electronic device based on the large language model provided in the embodiments of the present invention correspond to the short video recommendation method based on the large language model provided in the embodiments of the present invention. The explanations, examples and beneficial effects of the relevant contents can refer to the corresponding parts in the short video recommendation method and will not be repeated here.

[0203] In summary, compared with the existing technology, the present invention has the following beneficial effects:

[0204] 1. The embodiment of the present invention integrates video content to generate a large language model, an interest description large language model, and a breaking circle strategy description large language model, and designs a recommendation system architecture to assist users in breaking the circle. While realizing personalized recommendations, it provides target users with high-quality and diversified short video recommendations, helping them to break out of the information cocoon and access a wider world.

[0205] 2. The embodiment of the present invention further integrates a large language model for title generation, and tailors engaging titles specifically for breaking-circle short videos in the recommendation list to stimulate users' desire to click and interest in watching, thereby increasing the attractiveness and conversion rate of recommended content, attracting users to interact with recommended videos, and helping users break through the circle.

[0206] 3. This embodiment of the present invention employs a dual-track video recommendation strategy. This strategy not only focuses on the target user's core interests, using an algorithm to select short videos that match their preferences to ensure recommendation accuracy and satisfaction, but also explores the boundaries of the target user's interests, selecting short videos that are likely to pique their curiosity and guide them beyond their existing interests. This recommendation mechanism is designed to promote the diversification of the target user's interests, encouraging them to try new things, thereby broadening their horizons and enjoying a more diverse video viewing experience.

[0207] 4. When evaluating a user's willingness to break out of their circle, the embodiment of the present invention comprehensively considers two dimensions: personal historical behavior and overall community trends; and when formulating a video recommendation strategy, the embodiment of the present invention recognizes that receiving the evaluation results of managers is also crucial to ensure that the strategy meets social and platform needs and optimizes the recommendation effect.

[0208] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0209] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A short video recommendation method based on a large language model, characterized in that: include: Obtain a collection of short videos and a user set containing the target user; A pre-trained BERT-based short video content generation language model is used to generate a description text for each short video. Keywords in the description text are extracted using the TF-IDF algorithm to obtain several tags for each short video. After optimization and merging, the tag set for the short video set is obtained. Count the historical short videos of each user's interest within a preset historical time period, combine the tag set to obtain several interest tags of each user, and use the K-means algorithm to perform clustering to determine the interest community to which each user belongs; Compare the similarity between the target user's interest tags and the tags of any short video to generate a first short video recommendation list based on the user's personal interests; in the interest community to which the target user belongs, adopt a collaborative filtering method based on neighboring users to generate a second short video recommendation list based on the interest community; Using a Boosting algorithm to fuse the first short video recommendation list and the second short video recommendation list to obtain a third short video recommendation list based on interest; Locate the communities closest to and farthest from the target user's interest group, and build a corresponding list of candidate short videos for recommendation; The candidate short video recommendation lists of the nearest and farthest communities are integrated, and the number of recommended short videos is allocated based on the distance to obtain a fourth short video recommendation list that assists in breaking the circle; Based on the target user's willingness to break the circle and the manager's decision score, the proportion of the fourth short video recommendation list is calculated, and combined with the third short video recommendation list to obtain the final short video recommendation list; wherein the pre-trained interest description large language model based on the Transformer deep learning architecture and the breaking circle strategy description large language model are used, combined with the evaluation results given by the manager, to obtain the manager's decision score; The process of obtaining the target user's willingness to break the circle includes: For target user u i* Interest group C pi* Sort the tags of interest in the cluster center, and define the first several tags with the highest number of occurrences as the mainstream tags of the community; define f serendipity (u i* ) function, calculate the target user u i* The proportion of videos not belonging to the mainstream tag of the community watched during the preset historical period is used as a measure of the target user u i* An indicator of personal breaking circle activity; Combined with the Jaccard coefficient, define g social (C pi* ) function, used as a measure of the target user u i* Interest group C pi* Indicators of the tendency of a community to break out of its circle; Among them, the target user u i* Interest group C pi* The number of mainstream tags is used as the base number; in two adjacent preset historical time periods, interest community C pi* The change of the label of watching short videos is taken as the change number; the g social (C pi* ) function refers to the ratio of the difference between the base number and the variation number to the sum of the base number and the variation number; Define the target user u i* The breaking circle willingness score function: S breakout (u i* )=αf serendipity (u i* )+βg social (C pi* ) Among them, S breakout (u i* ) is a value between 0 and 1; α, β represent the corresponding weight coefficients, and α+β=1, α, β∈(0,1); The process of obtaining the manager's decision score includes: Using the tags of the third short video recommendation list as input to the interest description large language model to generate a text description of the community interest; and using the tags of the fourth short video recommendation list as input to the breaking circle strategy description large language model to generate a text description of the breaking circle strategy; Based on the text description of the community interest and the text description of the breaking circle strategy, combined with the evaluation results given by the manager, the manager's decision score is defined as S decide , the value is between 0 and 1.

2. The short video recommendation method according to claim 1, wherein: Also includes: A pre-trained large language model for title generation based on a Transformer deep learning architecture is used to generate a title for each short video in the short video recommendation list that originates from the fourth short video recommendation list.

3. The short video recommendation method according to claim 1, wherein: The method uses a pre-trained BERT-based short video content generation large language model to generate a description text for each short video, and uses the TF-IDF algorithm to extract keywords from the description text to obtain multiple tags for each short video, and optimizes and merges them to obtain a tag set for the short video set; including: For any short video, a motion detection algorithm is used to decode the continuous video stream into a series of key frames F o , where o represents the index of the key frame; Calculate each key frame F o The color histogram defines the color space and color channels, counts the number of pixels of each color value in each color channel, and forms the color histogram feature vector v color,o ; The gray-level co-occurrence matrix algorithm is used to extract the texture feature vector v by statistically analyzing the joint probability distribution of pixel pairs in the gray-level image at the preset direction and distance. texture,o ; Use the Canny edge detection algorithm to detect the edges in the image, and combine it with the Hough transform to extract the shape features to obtain the shape feature set S o , and converted into shape feature vector v shape,o ; Extract the audio features of the short video and convert the dubbing in the short video into a text sequence T audio ; Fusion visual feature vector set {v color,o ,v texture,o ,v shape,o } and the converted text sequence T audio , use the pre-trained BERT-based short video content to generate a large language model for encoding, generate a hidden state vector, and generate a descriptive text D through a fully connected layer or classifier; Preprocessing the descriptive text D, extracting keywords from the preprocessed descriptive text D using a TF-IDF algorithm, and using the extracted keywords as candidate tags; Based on the semantic relationship between the candidate tags, a machine learning algorithm is used to optimize the tags and obtain several tags {l1, l2, ..., l m }, where m represents the total number of tags of the short video; Merge the labels of all short videos to obtain the label set L of the short video set total .

4. The short video recommendation method according to claim 1, wherein: The method comprises collecting statistics on a set of historically interested short videos of each user within a preset historical time period, combining the tag set to obtain a number of interest tags of each user, and performing clustering using a K-means algorithm to determine the interest community to which each user belongs; include: The user set is defined as U = {u1,u2,…,u n }, where n represents the total number of users; define user u i The collection of historically interesting short videos within the preset historical time period is where i m Represents user u i The total number of historical short videos of interest; for each short video v j ∈V i , define the tag set of the short video as L total A tag set representing the short video set; For user u i , build the corresponding label count dictionary in, Indicates that label l is in user u i History of Interesting Short Video Collection V i The number of times it appears in ; U represents the indicator function, when 1 when it is, otherwise 0; Select the tag whose number of occurrences is greater than the preset threshold as the user u i Merge all users’ interest tags and remove duplicates to obtain the user interest tag set L top ; Clustering using K-means algorithm: Among them, SSE represents the sum of squares of distances from all users to their corresponding cluster centers; K represents the set number of clusters; C c represents the set of users in the cth cluster; μ c is the center of the cth cluster; Taking minimizing SSE as the optimization goal, iteratively update the cluster center and reallocate users to the nearest cluster center to determine user u i Interest group C pi .

5. The short video recommendation method according to claim 4, wherein: The method of comparing the similarity between the target user's interest tag and the tag of any short video to generate a first short video recommendation list based on the user's personal interest includes: For target user u i* , define a feature vector If and only if L top When {k} is the user's interest tag, Equal to 1; for short video v j , define its label as If and only if L top When {k} is the tag of the video, Equal to 1; where |L top | represents the number of tags in the user’s interest tag set, L top {k} represents the kth tag in the user's interest tag set; The following formula is used to compare the similarity between the target user's interest tags and the tags of any short video: Where sim represents the similarity function, · represents the dot product, and ||·|| represents the Euclidean length of the vector; By comparing the similarities, we find a collection of videos that are similar to the target user’s interests and sort them according to the similarity scores to generate the first short video recommendation list V based on the user’s personal interests. CBR ; and / or The method of generating a second short video recommendation list based on the interest community to which the target user belongs by using a collaborative filtering method based on neighboring users includes: Calculate the similarity between users and determine the target user u i* Neighbor user set of Utilize the short videos v that each neighbor user in the interest community to which the target user belongs has watched or rated p The target user u is predicted by weighted average method based on the data of i* Potential interest in unrated short videos; calculated as follows: in, represents the target user u i* Short video v p The predicted score of the target user is N(u), which represents the interest community to which the target user belongs. u′_vp Represents neighbor user u′’s response to short video v p The actual rating of Sort the short videos based on the predicted scores, and select the short videos with the highest scores as the second short video recommendation list V based on the interest community. UCF .

6. The short video recommendation method according to claim 1, wherein: Based on the target user's willingness to break the circle and the manager's decision score, the proportion of the fourth short video recommendation list is calculated, and combined with the third short video recommendation list to obtain a final short video recommendation list; including: Calculate the short video recommendation list V Recommend The number of short videos is: NUM(V last )=γ1NUM(N UCF )+γ2NUM(V breakout ) Among them, γ1, γ2 represent the corresponding weight coefficients, and γ1+γ2=1, γ1, γ2∈(0,1); NUM(V UCF ),NUM(V breakout ) represent the third short video recommendation list V UCF And the fourth short video recommendation list V breakout The number of short videos; Select V according to the ratio UCF and V breakout The short videos with the highest to lowest ratings are combined to obtain the final recommendation list V Recommend .

7. A short video recommendation system based on a large language model, characterized in that: Used to execute the short video recommendation method according to claim 1, comprising: An acquisition module is used to obtain a short video set and a user set containing target users; The recognition module is used to generate a description text for each short video using a pre-trained BERT-based short video content generation large language model, and extract keywords from the description text using the TF-IDF algorithm to obtain multiple tags for each short video, and then optimize and merge them to obtain a tag set for the short video set; and for counting the historical short video sets of interest to each user within a preset historical time period, combining the tag set to obtain a number of interest tags for each user, and clustering using the K-means algorithm to determine the interest community to which each user belongs; The recommendation module is used to compare the similarity between the target user's interest tags and the tags of any short video to generate a first short video recommendation list based on the user's personal interests; in the interest community to which the target user belongs, a collaborative filtering method based on neighboring users is used to generate a second short video recommendation list based on the interest community; and a Boosting algorithm is used to fuse the first and second short video recommendation lists to obtain a third short video recommendation list based on interest; and for respectively locating the communities closest to and farthest from the interest community to which the target user belongs, and constructing corresponding candidate short video recommendation lists; fusing the candidate short video recommendation lists of the nearest and farthest communities, and allocating the number of recommended short videos based on distance, to obtain a fourth short video recommendation list that assists in breaking the circle; The optimization module is used to calculate the proportion of the fourth short video recommendation list based on the target user's willingness to break the circle and the manager's decision score, and combine it with the third short video recommendation list to obtain the final short video recommendation list; wherein the pre-trained interest description large language model and breaking the circle strategy description large language model based on the Transformer deep learning architecture are used, combined with the evaluation results given by the manager to obtain the manager's decision score.

8. A storage medium, characterized in that: The computer program for recommending short videos based on a large language model is stored therein, wherein the computer program enables a computer to execute the short video recommendation method according to any one of claims 1 to 6.

9. An electronic device, characterized in that: include: one or more processors; Memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include a program for executing the short video recommendation method according to any one of claims 1 to 6.

Images