Information Recommendation Method, Apparatus, Electronic Device, and Storage Medium

Abstract

The present disclosure discloses an information recommendation method, apparatus, electronic device, and storage medium, relating to the field of computer technologies, and particularly to the field of artificial intelligence technologies. The specific implementation solution is as follows: A large model is used to identify user personality information, generate and display a first recommendation message corresponding to the user personality information; in the case of receiving a confirmation instruction for the first recommendation message, display recommendation information corresponding to the first recommendation message; determine and display extended recommendation information corresponding to the recommendation information according to first interaction information corresponding to the recommendation information. Therefore, the present disclosure can improve the accuracy of information recommendation and at the same time improve the satisfaction of information recommendation.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, particularly to the field of artificial intelligence technology, and specifically to an information recommendation method, apparatus, electronic device, and storage medium. Background Art

[0002] With the development of science and technology, electronic devices can provide users with more and more services, improving the convenience of users' lives. For example, when users face a huge demand for information and difficulties in making choices, information that matches their interests and needs can be provided to them. For example, the items to be recommended can be determined by comparing the user's historical behavior with the characteristics of the items. Summary of the Invention

[0003] This disclosure provides an information recommendation method, apparatus, electronic device, and storage medium, with the main purpose of improving the accuracy of information recommendation while enhancing the satisfaction of information recommendation.

[0004] According to one aspect of this disclosure, an information recommendation method is provided, including:

[0005] Using a large model to identify user personality information, generating and displaying a first recommendation statement corresponding to the user personality information;

[0006] When a confirmation instruction for the first recommendation statement is received, displaying the recommendation information corresponding to the first recommendation statement;

[0007] Determining and displaying extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information.

[0008] According to another aspect of this disclosure, an information recommendation apparatus is provided, including:

[0009] A recommendation statement generation unit, configured to use a large model to identify user personality information, and generate and display a first recommendation statement corresponding to the user personality information;

[0010] An information display unit, configured to display the recommendation information corresponding to the first recommendation statement when a confirmation instruction for the first recommendation statement is received;

[0011] A recommendation information determination unit, configured to determine and display extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information.

[0012] According to another aspect of this disclosure, an electronic device is provided, including:

[0013] At least one processor; and

[0014] A memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method according to any one of the foregoing aspects.

[0016] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to any one of the foregoing aspects.

[0017] According to another aspect of the present disclosure, there is provided a computer program product including a computer program, and when the computer program is executed by a processor, the method according to any one of the foregoing aspects is implemented.

[0018] In one or more embodiments of the present disclosure, a large model is used to identify user personality information, generate and display a first recommendation corresponding to the user personality information; when a confirmation instruction for the first recommendation is received, display recommendation information corresponding to the first recommendation; according to first interaction information corresponding to the recommendation information, determine and display extended recommendation information corresponding to the recommendation information. Therefore, the time for a user to discover interesting information can be reduced through the recommendation, the information display efficiency can be improved, the transparency and interpretability of the recommendation can be improved, the situation where the recommendation changes and makes information recommendation inconvenient can be reduced, the interactivity with the user during the information recommendation process can be improved, and by determining the recommendation through a large model and performing information recommendation and extension of the recommendation information based on the recommendation, the accuracy of the information recommendation can be improved while the satisfaction of the information recommendation can be improved.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0021] Figure 1 is a schematic flowchart of an information recommendation method according to a first embodiment of the present disclosure;

[0022] Figure 2 is a schematic flowchart of an information recommendation method according to a second embodiment of the present disclosure;

[0023] FIG. 3(a) is a first exemplary schematic diagram of an electronic device interface according to an embodiment of the present disclosure;

[0024] FIG. 3(b) is a second exemplary schematic diagram of an electronic device interface according to an embodiment of the present disclosure;

[0025] FIG. 3(c) is a third exemplary schematic diagram of an electronic device interface according to an embodiment of the present disclosure;

[0026] FIG. 3(d) is a fourth exemplary schematic diagram of an electronic device interface according to an embodiment of the present disclosure;

[0027] FIG. 3(e) is a fifth exemplary schematic diagram of an electronic device interface according to an embodiment of the present disclosure;

[0028] Figure 4 is a flowchart of an information recommendation method according to an embodiment of the present disclosure;

[0029] FIG. 5(a) is a schematic structural diagram of an information recommendation device for implementing the information recommendation method according to an embodiment of the present disclosure;

[0030] FIG. 5(b) is a schematic structural diagram of another information recommendation device for implementing the information recommendation method according to an embodiment of the present disclosure;

[0031] Figure 6 is a block diagram of an electronic device for implementing the information recommendation method according to an embodiment of the present disclosure. Detailed Embodiments

[0032] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, descriptions of well-known functions and structures are omitted for clarity and conciseness.

[0033] According to some embodiments, for example, information recommendation can be performed through a content-based recommendation system, a collaborative filtering-based recommendation system, a hybrid model-based recommendation system, and a deep learning-based recommendation system. However, in the process of information recommendation by a content-based recommendation system, a collaborative filtering-based recommendation system, a hybrid model-based recommendation system, and a deep learning-based recommendation system, there will be situations of low user participation and inaccurate recommendation.

[0034] Among them, the content-based recommendation system mainly uses the characteristics of items and the historical behavior of users, and matches users with items by calculating the similarity between the two, so as to provide personalized recommendations for users.

[0035] Among them, the collaborative filtering-based recommendation system uses the similarity between users and other users or items to make recommendations, that is, it predicts the items that users may like based on the historical behavior of users and the behavior of other users.

[0036] Among them, the hybrid model-based recommendation system is a method that combines different types of recommendation algorithms to improve the recommendation quality. It combines multiple algorithms such as collaborative filtering, content filtering, and popularity recommendation, so as to more accurately meet the user's needs.

[0037] Among them, the deep learning-based recommendation system is a method that uses deep neural network models to process the complex relationships between users and items to improve the recommendation accuracy. Its main technical routes include matrix factorization, autoencoders, attention mechanisms, graph neural networks, etc.

[0038] The following will explain the present disclosure in detail with specific embodiments.

[0039] In the first embodiment, as Figure 1 shown, Figure 1 is a schematic flowchart of an information recommendation method according to the first embodiment of the present disclosure. This method can be implemented depending on a computer program and can run on a device for information recommendation. This computer program can be integrated into an application or run as an independent tool application.

[0040] Among them, the information recommendation device can be an electronic device with artificial intelligence interaction capabilities. This electronic device includes but is not limited to: autonomous driving vehicles, wearable devices, handheld devices, personal computers, tablets, in-vehicle devices, smartphones, computing devices, or other processing devices connected to a wireless modem, etc. In different networks, the terminal can be called by different names. For example: user equipment, access terminal, user unit, user station, mobile station, mobile unit, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, cellular phone, cordless phone, personal digital assistant (PDA), fifth-generation mobile communication technology (5G) network, fourth-generation mobile communication technology (4G) network, third-generation mobile communication technology (3G) network, or an electronic device in a future evolved network, etc.

[0041] Specifically, the information recommendation method includes:

[0042] S101. Use a large model to identify user personality information, generate and display a first recommendation corresponding to the user personality information.

[0043] According to some embodiments, the large model may refer to a model in the field of artificial intelligence that can process massive amounts of data and complete various complex tasks. The large model of the present disclosure embodiments can, for example, handle information extraction, information extension, information identification, etc.

[0044] In some embodiments, the user personality information may be information that reflects the user's personality. The user personality information does not specifically refer to a certain fixed information. For example, when the user using the electronic device changes, the user personality information can also change accordingly. For example, when the specific information included in the user personality information changes, the user personality information can also change accordingly.

[0045] In some embodiments, the first recommendation refers to a recommendation corresponding to the user personality information. The "first" in the first recommendation is only used to distinguish it from the rest of the recommendations. It does not specifically refer to a certain fixed recommendation. For example, when the user personality information changes, the first recommendation can also change accordingly. For example, when the generation time point of the first recommendation changes, the first recommendation can also change accordingly.

[0046] According to some embodiments, when executing the information recommendation method, a large model can be used to identify user personality information, generate and display a first recommendation corresponding to the user personality information.

[0047] S102. When a confirmation instruction for the first recommendation is received, display the recommendation information corresponding to the first recommendation.

[0048] According to some embodiments, the confirmation instruction may be, for example, an instruction to confirm the recommendation based on the first recommendation. The confirmation instruction does not specifically refer to a certain fixed instruction. The confirmation instruction includes but is not limited to a voice confirmation instruction, a click confirmation instruction, a timed confirmation instruction, etc. The confirmation instruction may be, for example, a voice confirmation instruction, and the voice confirmation instruction may be, for example, "Recommend information according to the first recommendation".

[0049] According to some embodiments, the recommendation information refers to the recommended information corresponding to the first recommendation. The recommendation information does not specifically refer to a certain fixed information. For example, when the determination method of the recommendation information changes, the recommendation information can also change accordingly. For example, when the specific information corresponding to the recommendation information changes, the recommendation information can also change accordingly.

[0050] According to some embodiments, when a confirmation instruction for the first recommendation is received, display the recommendation information corresponding to the first recommendation.

[0051] S103. Determine and display extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information.

[0052] In some embodiments, the first interaction information refers to the interaction information corresponding to the recommendation information after the recommendation information is displayed. The first interaction information does not specifically refer to a certain fixed information. The "first" in the first interaction information is only used to distinguish it from the rest of the interaction information and does not specifically refer to a certain fixed information. The first interaction information includes but is not limited to browsing duration, number of clicks, and clicked content.

[0053] According to some embodiments, the extended recommendation information refers to the information obtained by extending the recommendation information based on the first interaction information. The extended recommendation information does not specifically refer to a certain fixed information. For example, when the determination method of the extended recommendation information changes, the extended recommendation information can also change accordingly. For example, when the extension direction changes, the extended recommendation information can also change accordingly.

[0054] According to some embodiments, when a confirmation instruction for the first recommendation statement is received, display the recommendation information corresponding to the first recommendation statement. The extended recommendation information corresponding to the recommendation information can be determined and displayed according to the first interaction information corresponding to the recommendation information.

[0055] In one or more embodiments of the present disclosure, a large model is used to identify user personality information, generate and display a first recommendation statement corresponding to the user personality information; when a confirmation instruction for the first recommendation statement is received, display the recommendation information corresponding to the first recommendation statement; determine and display the extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information. Therefore, the time for users to discover interesting information can be reduced through the recommendation statement, diverse interest points can be covered, the information display efficiency can be improved, the time for users to obtain recommendation information can be reduced, the transparency and interpretability of the recommendation can be improved, the situation where information recommendation is inconvenient due to recommendation changes can be reduced, the interactivity with users during the information recommendation process can be improved, and the recommendation statement is determined by the large model, and information recommendation and extension of the recommendation information are performed according to the recommendation statement, so as to improve the matching degree between the recommendation information and the user personality information, improve the accuracy of information recommendation, and at the same time improve the satisfaction of information recommendation. In addition, the technical solution of the embodiments of the present disclosure involves multiple processes of recommendation, and can provide full - range intelligent companionship to improve the efficiency of users to obtain recommendation information. Then, in the form of a recommendation statement, the interaction method can be enriched, the diversity of interaction can be improved, and the sense of participation of users can be improved.

[0056] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of the information recommendation method according to the second embodiment of the present disclosure. Specifically,

[0057] S201. Use a large model to identify user personality information, generate and display a first recommendation corresponding to the user personality information.

[0058] The specific process is as described above and will not be elaborated here.

[0059] According to some embodiments, the user personality information includes user historical consumption behavior information, user portrait information, and user interest point information. Using a large model to identify user personality information and generate and display a first recommendation corresponding to the user personality information includes:

[0060] Use a large model to identify user historical consumption behavior information, user portrait information, and user interest point information, and generate and display a first recommendation corresponding to the user personality information. Therefore, using a large model for identification can improve the accuracy of user intention understanding, improve the accuracy of user interest determination, can improve the accuracy of recommendation determination, and improve the accuracy of information recommendation.

[0061] Among them, the user interest point information may include, for example, user long-term interest point information and user short-term interest point information. Among them, the user long-term interest point information may be, for example, the interest points within the first historical time period. Among them, the user short-term interest point information may be, for example, the interest points within the second historical time period. Among them, the first historical time period is greater than the second historical time period.

[0062] In some embodiments, the first recommendation may be, for example, "Hello, is your work going well recently? You seem to be interested in topics related to the workplace and self-improvement. It seems that you are a person with high self-requirements. Here are some skills about workplace communication, promotion, and management that can be shared with you. Would you like to take a look?" At this time, the display interface of the electronic device may be as shown in Figure 3(a).

[0063] According to some embodiments, the method further includes:

[0064] Obtain a first recommendation training set, where the first recommendation training set includes at least one labeled second recommendation.

[0065] Use the self-instruct technology to perform prompt expansion on at least one labeled second recommendation to obtain a second recommendation training set.

[0066] Use the second recommendation training set to fine-tune the large model to obtain an adjusted large model. Therefore, performing prompt expansion on at least one labeled second recommendation can expand the training data to fine-tune the large model. By iterating a small number of model parameters, the efficiency of fine-tuning and optimizing the large model using business data can be improved.

[0067] Among them, obtaining the first recommended phrase training set can be, for example, recommended phrases generated based on information such as the user's historical consumption behavior, user portrait, and long- and short-term user interest points, and the set obtained by annotating these recommended phrases. The first recommended phrase training set refers to a collective formed by converging at least one annotated second recommended phrase. The first recommended phrase training set does not specifically refer to a certain fixed set. For example, when any second recommended phrase in the first recommended phrase training set changes, the first recommended phrase training set can also change accordingly.

[0068] Among them, for example, the prompts in the large model can also be designed. For example, the electronic device can perform information extraction and summarization on the information set, design prompt templates in various different language styles and expression forms according to video or graphic content, and apply them to content highlight generation, comment view extraction, and generation of extended recommended topics.

[0069] S202, when receiving a confirmation instruction for the first recommended phrase, display the recommended information corresponding to the first recommended phrase;

[0070] The specific process is as described above and will not be elaborated here.

[0071] In some embodiments, the recommended information includes but is not limited to text information, audio information, and video information. For example, the recommended information can include text information, the recommended information can include video information, and the recommended information can include text information and video information. The embodiments of the present disclosure do not limit this.

[0072] According to some embodiments, the first recommended phrase can be, for example, "Hello, how's work going recently? You seem to be interested in topics related to the workplace and self-improvement. It seems you're a person with high standards for yourself. Here are some tips on workplace communication, career advancement, and management that I can share with you. Would you like to take a look?" When receiving a click confirmation instruction for the first recommended phrase, the recommended information corresponding to the first recommended phrase can be displayed. At this time, the display interface of the electronic device can be, for example, as shown in Figure 3(b).

[0073] According to some embodiments, displaying the recommended information corresponding to the first recommended phrase includes:

[0074] Perform information extraction on the recommended information corresponding to the first recommended phrase to obtain the extracted recommended information; display the extracted recommended information. This can improve the simplicity of the recommended information display, reduce the time taken to obtain the recommended information, and improve the browsing efficiency of the recommended information.

[0075] According to some embodiments, before presenting the recommendation information corresponding to the first recommendation statement, it further includes: obtaining the first keyword corresponding to the first recommendation statement; searching for the recommendation information including the first keyword in the recommendation information set, and using the recommendation information including the first keyword as the recommendation information corresponding to the first recommendation statement. Therefore, by searching according to the keyword, the relevance between the recommendation information and the first recommendation statement can be improved, the situation where the recommendation information is not accurately determined due to the mismatch between the recommendation information and the first recommendation statement can be reduced, and the accuracy of determining the recommendation information can be improved.

[0076] According to some embodiments, wherein, before presenting the recommendation information corresponding to the first recommendation statement, it further includes: obtaining the first vector corresponding to the first recommendation statement;

[0077] obtaining the second vector of any recommendation information in the recommendation information set;

[0078] Determining the recommendation information corresponding to the first recommendation statement in the recommendation information set according to the similarity between the first vector and the second vector. Therefore, by determining the recommendation information according to the similarity between the vectors, the relevance between the recommendation information and the first recommendation statement can be improved, the situation where the recommendation information is not accurately determined due to the mismatch between the recommendation information and the first recommendation statement can be reduced, and the accuracy of determining the recommendation information can be improved.

[0079] Among them, the first vector and the second vector can be, for example, hidden vectors, and specifically can be, for example, embedding vectors. The "first" in the first vector is only used to distinguish it from other vectors and does not specifically refer to a certain fixed vector. For example, when the first recommendation statement changes, the first vector can also change accordingly.

[0080] According to some embodiments, before presenting the comment information corresponding to the recommendation information, it further includes: classifying the comment information set corresponding to the recommendation information using a classification model, and obtaining a subset of comment information related to the recommendation information according to the second interaction information corresponding to any comment information in the comment information set;

[0081] Using a large model and a prompt template to extract information from any comment information in the subset of comment information, and obtaining the extracted any comment information. Therefore, the comment information can be screened to improve the matching between the comment information and the recommendation information, and extracting the comment information can improve the efficiency of users obtaining the comment information.

[0082] Among them, the second interaction information includes but is not limited to the number of likes, the number of replies, etc. of any comment information.

[0083] S203, presenting the comment information corresponding to the recommendation information;

[0084] According to some embodiments, when presenting the recommendation information corresponding to the first recommendation, the comment information corresponding to the recommendation information may be presented.

[0085] Among them, when presenting the comment information corresponding to the recommendation information, for example, it may be presented simultaneously with the recommendation information corresponding to the first recommendation, or for example, it may also be presented when a display instruction for the comment information is received.

[0086] According to some embodiments, the first recommendation statement may be, for example, "Hello, how's work going recently? You seem to be interested in topics related to the workplace and self-improvement. It seems that you are a person with high demands on yourself. Here are some tips on workplace communication, promotion, and management that can be shared with you. Would you like to take a look?" When a click and confirmation instruction for the first recommendation statement is obtained, the recommendation information corresponding to the first recommendation statement may be presented. When presenting the recommendation information, the comment information corresponding to the recommendation information may be presented. At this time, the display interface of the electronic device may be, for example, as shown in FIG. 3(c).

[0087] S204. Determine and present the extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information.

[0088] The specific process is as described above and will not be elaborated here.

[0089] According to some embodiments, determining and presenting the extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information includes:

[0090] Determine the extended dimension information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information; generate extended information corresponding to the extended dimension information; obtain at least one extended recommendation information related to the extended information from the recommendation information set according to the extended information, and use the at least one extended recommendation information as the extended recommendation information corresponding to the recommendation information; present the extended recommendation information corresponding to the recommendation information. Therefore, determining the extended recommendation information based on the extended information can improve the accuracy of determining the extended recommendation information and improve the accuracy of information recommendation. Among them, the at least one extended recommendation information may, for example, become a resource collection.

[0091] In some embodiments, the extended dimension information may, for example, refer to the extended dimension of the recommendation information. The extended dimension information may be, for example, flexible. The extended dimension information may include, but is not limited to, strongly related extended dimensions, semi-related extended dimensions, etc. The extended dimension information may also include, for example, themes, keywords, selected information, etc.

[0092] Among them, the first recommended message can be, for example, "Tourist food in Area A". Among them, the extended information determined by the strongly related extended dimension can be, for example, "Tourist specialties in Area A". Among them, the extended information determined by the semi-related extended dimension can be, for example, "Tourism in Area A" or "Specialties".

[0093] According to some embodiments, among them, obtaining at least one extended recommended message related to the extended information in the recommended message set includes:

[0094] Search for the recommended message including the extended information in the recommended message set, and use the recommended message including the extended information as at least one extended recommended message related to the extended information. Therefore, the method of directly searching can be adopted to determine the delayed recommended information, which can improve the matching between the extended recommended information and the extended information, reduce the probability that the extended recommended information is irrelevant to the extended information, reduce the determination duration of the extended recommended information, and improve the accuracy and efficiency of determining the extended recommended information.

[0095] According to some embodiments, among them, obtaining at least one extended recommended message related to the extended information in the recommended message set includes:

[0096] Obtain the third vector corresponding to the extended information;

[0097] Obtain the second vector of any recommended message in the recommended message set;

[0098] Determine at least one extended recommended message related to the extended information in the recommended message set according to the similarity between the third vector and the second vector. Therefore, the method of vector similarity can be adopted to determine the delayed recommended information, which can improve the matching between the extended recommended information and the extended information, reduce the probability that the extended recommended information is irrelevant to the extended information, reduce the determination duration of the extended recommended information, and improve the accuracy and efficiency of determining the extended recommended information.

[0099] According to some embodiments, among them, the method further includes:

[0100] Obtain the third interaction information corresponding to at least one extended recommended message;

[0101] Sort at least one extended recommended message according to the third interaction information to obtain at least one sorted extended recommended message;

[0102] Determine at least one extended recommended message for display according to at least one sorted extended recommended message. Therefore, sorting can be performed according to the interaction information of the extended recommended message, which can improve the matching between the extended recommended message and the extended information, reduce the probability that the extended recommended message is irrelevant to the extended information, reduce the determination duration of the extended recommended message, and improve the accuracy and efficiency of determining the extended recommended message.

[0103] Among them, the sorting in the embodiments of the present disclosure includes but is not limited to processes such as rough sorting and fine sorting. Rough sorting and fine sorting can be, for example, processes of sorting at least one extended recommendation information through different metrics.

[0104] According to some embodiments, the first recommendation message can be, for example, "Hello, how's work going recently? You seem to be interested in topics related to the workplace and self-improvement. It seems you're a person with high self-requirements. Here are some tips on workplace communication, career advancement, and management that can be shared with you. Would you like to take a look?" When a click and confirmation instruction for the first recommendation message is obtained, the recommendation information corresponding to the first recommendation message can be displayed. When displaying the recommendation information, the comment information corresponding to the recommendation information can be displayed. Extended recommendation information corresponding to the recommendation information can be obtained according to the theme of the recommendation information. At this time, the display interface of the electronic device can be, for example, as shown in FIG. 3(d).

[0105] Before displaying the extended recommendation information, a prompt message can be sent, for example. When a confirmation instruction for the prompt message is received, the extended recommendation information is displayed. At this time, the display interface of the electronic device can be, for example, as shown in FIG. 3(e).

[0106] Figure 4 It is a schematic flowchart of an information recommendation method according to an embodiment of the present disclosure. As Figure 4 shown, the information recommendation method of the present disclosure can be, for example, a companion-style recommendation method. Before recommendation, for example, a recommendation message can be generated based on user personality information to guide the user to discover more content they like, and content highlights can be displayed according to the content characteristics of the article or video to expand the scope of interest; during recommendation, for example, comments that meet the conditions in the comment area can be mined, and comment viewpoints can be generated according to the comment characteristics, enabling the user to interact with the comment viewpoints in real time to improve the display efficiency of the recommendation information; after recommendation, according to the content characteristics of the recommended information browsed by the user, depth and breadth extensions can be made to enhance the user's immersive consumption experience. Among them, user personality information can include, for example, user historical consumption behavior, user portrait, and user's long and short interest points, and specifically can include, for example, age, region, occupation, hobbies, interest points, etc. Content characteristics include but are not limited to title, body text, classification, interest points, etc. Comment characteristics include but are not limited to resource title, comment content, comment length, comment likes, and comment replies. The content characteristics of the recommendation information include but are not limited to title, body text, classification, interest points, clicks, displays, shares, and collections.

[0107] In one or more embodiments of the present disclosure, the comment information corresponding to the recommendation information can be displayed. Therefore, the comment information corresponding to the recommendation information can be directly displayed when the recommendation information is displayed, which can provide different perspectives for the display of the recommendation information, can improve the richness of the display of the recommendation information, and can improve the information display efficiency.

[0108] The following are embodiments of the disclosed device, which can be used to implement the method embodiments of the present disclosure. For details not disclosed in the device embodiments of the present disclosure, please refer to the method embodiments of the present disclosure.

[0109] Please refer to FIG. 5(a), which shows a schematic structural diagram of an information recommendation device for implementing the information recommendation method of the embodiments of the present disclosure. The information recommendation device can be implemented as all or part of the device through software, hardware, or a combination of both. The information recommendation device 500 includes a recommendation statement generation unit 501, an information display unit 502, and a recommended information determination unit 503, where:

[0110] The recommendation statement generation unit 501 is configured to identify user personality information using a large model, generate and display a first recommendation statement corresponding to the user personality information;

[0111] The information display unit 502 is configured to display recommended information corresponding to the first recommendation statement when a confirmation instruction for the first recommendation statement is received;

[0112] The recommended information determination unit 503 is configured to determine and display extended recommended information corresponding to the recommended information according to the first interaction information corresponding to the recommended information.

[0113] According to some embodiments, please refer to FIG. 5(b), which shows a schematic structural diagram of another information recommendation device for implementing the information recommendation method of the embodiments of the present disclosure. The device 500 further includes:

[0114] A set acquisition unit 504, configured to acquire a first recommendation statement training set, where the first recommendation statement training set includes at least one labeled second recommendation statement;

[0115] The set acquisition unit 504 is further configured to perform prompt expansion on at least one labeled second recommendation statement using the self-instruct technology to obtain a second recommendation statement training set;

[0116] A model adjustment unit 505, configured to fine-tune the large model using the second recommendation statement training set to obtain an adjusted large model.

[0117] According to some embodiments, where the information display unit 502 is further configured to, before displaying the recommended information corresponding to the first recommendation statement:

[0118] Acquire a first keyword corresponding to the first recommendation statement;

[0119] Search for recommended information including the first keyword in the recommended information set, and use the recommended information including the first keyword as the recommended information corresponding to the first recommendation statement.

[0120] According to some embodiments, the information display unit 502 is further configured to, before displaying the recommendation information corresponding to the first recommendation statement:

[0121] Obtain the first vector corresponding to the first recommendation statement;

[0122] Obtain the second vector of any recommendation information in the recommendation information set;

[0123] Determine the recommendation information corresponding to the first recommendation statement in the recommendation information set according to the similarity between the first vector and the second vector.

[0124] According to some embodiments, the information display unit 502 is further configured to:

[0125] Display the comment information corresponding to the recommendation information.

[0126] According to some embodiments, the information display unit 502 is further configured to, before displaying the comment information corresponding to the recommendation information:

[0127] Classify the comment information set corresponding to the recommendation information by using a classification model, and obtain a subset of comment information related to the recommendation information according to the second interaction information corresponding to any comment information in the comment information set;

[0128] Perform information extraction on any comment information in the subset of comment information by using a large model and a prompt template to obtain the extracted any comment information.

[0129] According to some embodiments, when the recommendation information determination unit 503 is configured to determine and display the extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information, it is specifically configured to:

[0130] Determine the extended dimension information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information;

[0131] Generate extended information corresponding to the extended dimension information;

[0132] Obtain at least one extended recommendation information related to the extended information from the recommendation information set according to the extended information, and use the at least one extended recommendation information as the extended recommendation information corresponding to the recommendation information;

[0133] Display the extended recommendation information corresponding to the recommendation information.

[0134] According to some embodiments, when the recommendation information determination unit 503 is configured to obtain at least one extended recommendation information related to the extended information from the recommendation information set, it is specifically configured to:

[0135] Search for recommended information including extended information in the set of recommended information, and use the recommended information including extended information as at least one extended recommended information related to the extended information.

[0136] According to some embodiments, the recommended information determining unit 503, when obtaining at least one extended recommended information related to the extended information in the set of recommended information, is specifically configured to:

[0137] Obtain a third vector corresponding to the extended information;

[0138] Obtain a second vector of any recommended information in the set of recommended information;

[0139] Determine at least one extended recommended information related to the extended information in the set of recommended information according to the similarity between the third vector and the second vector.

[0140] According to some embodiments, the recommended information determining unit 503 is further configured to:

[0141] Obtain third interaction information corresponding to at least one extended recommended information;

[0142] Sort at least one extended recommended information according to the third interaction information to obtain at least one sorted extended recommended information;

[0143] Determine at least one extended recommended information for display according to at least one sorted extended recommended information.

[0144] According to some embodiments, the user personality information includes user historical consumption behavior information, user portrait information, and user interest point information; when the recommendation statement generating unit 501 uses a large model to identify the user personality information and generate and display a first recommendation statement corresponding to the user personality information, it is specifically configured to:

[0145] Use a large model to identify the user historical consumption behavior information, user portrait information, and user interest point information, and generate and display a first recommendation statement corresponding to the user personality information.

[0146] According to some embodiments, the information display unit 502, when displaying the recommended information corresponding to the first recommendation statement, is specifically configured to:

[0147] Extract information from the recommended information corresponding to the first recommendation statement to obtain the extracted recommended information;

[0148] Display the extracted recommended information.

[0149] It should be noted that when the information recommendation device provided in the above embodiments executes the information recommendation method, only the division of the above functional modules is used for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the information recommendation device provided in the above embodiments and the embodiments of the information recommendation method belong to the same concept. The implementation process is detailed in the method embodiments and will not be repeated here.

[0150] The serial numbers of the above embodiments of the present disclosure are only for description and do not represent the advantages or disadvantages of the embodiments.

[0151] In summary, the device provided in the embodiments of the present disclosure includes a recommendation statement generation unit, which is used to identify user personality information using a large model, generate and display a first recommendation statement corresponding to the user personality information; an information display unit, which is used to display recommendation information corresponding to the first recommendation statement when a confirmation instruction for the first recommendation statement is received; and a recommendation information determination unit, which is used to determine and display extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information. Therefore, the time for users to discover interesting information can be reduced through the recommendation statement, the information display efficiency can be improved, the transparency and interpretability of the recommendation can be improved, the situation where the information recommendation is inconvenient due to the change of the recommendation can be reduced, the interactivity with users in the information recommendation process can be improved, and by determining the recommendation statement using a large model and performing information recommendation and extension of the recommendation information based on the recommendation statement, the accuracy of the information recommendation can be improved while the satisfaction of the information recommendation can be improved.

[0152] In the technical solution of the present disclosure, the acquisition, storage, and application of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0153] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0154] Figure 6 FIG. shows a schematic block diagram of an exemplary electronic device 600 that can be used to implement the embodiments of the present disclosure. Among them, the components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0155] As Figure 6As shown, the electronic device includes a computing unit 601, which can perform various appropriate actions and processes according to computer programs stored in a read-only memory (ROM) 602 or computer programs loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0156] Multiple components in the electronic device are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, an optical disc, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the electronic device to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0157] The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 601 executes the various methods and processes described above, such as the information recommendation method. For example, in some embodiments, the information recommendation method can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the information recommendation method described above can be executed. Alternatively, in other embodiments, the computing unit 601 can be configured to execute the information recommendation method by any other appropriate means (e.g., by means of firmware).

[0158] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0159] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine, or entirely on the remote machine or electronic device.

[0160] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0161] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0162] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data electronic device), or a computing system including middleware components (e.g., an application electronic device), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), the Internet, and blockchain networks.

[0163] A computer system can include a client and an electronic device. The client and the electronic device are generally far from each other and usually interact through a communication network. The relationship between the client and the electronic device is generated by computer programs running on respective computers and having a client - electronic device relationship with each other. The electronic device can be a cloud electronic device, also known as a cloud computing electronic device or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services ("Virtual Private Server", or simply "VPS"). The electronic device can also be an electronic device of a distributed system, or an electronic device combined with blockchain.

[0164] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and no limitation is imposed herein.

[0165] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. An information recommendation method, comprising: Using a large model to identify user personality information, generating and displaying a first recommendation statement corresponding to the user personality information; The user personality information includes user historical consumption behavior information, user portrait information, and user interest point information, and the first recommendation statement is a recommendation statement corresponding to the user's personalized information; When a confirmation instruction for the first recommendation statement is received, displaying recommendation information corresponding to the first recommendation statement; Determining and displaying extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information; Wherein, the method further includes: Obtaining a first recommendation statement training set, wherein the first recommendation statement training set includes at least one labeled second recommendation statement; Using the self-instruct technology to perform prompt expansion on the at least one labeled second recommendation statement to obtain a second recommendation statement training set; Using the second recommendation statement training set to fine-tune the large model to obtain an adjusted large model; The method further includes: Using a classification model to classify the comment information set corresponding to the recommendation information, and obtaining a subset of comment information related to the recommendation information according to the second interaction information corresponding to any comment information in the comment information set; Using the large model and a prompt template to perform information extraction on any comment information in the subset of comment information to obtain the extracted any comment information; Displaying the comment information corresponding to the recommendation information.

2. The method according to claim 1, wherein Before displaying the recommendation information corresponding to the first recommendation statement, it further includes: Obtaining a first keyword corresponding to the first recommendation statement; Searching for recommendation information including the first keyword in the recommendation information set, and using the recommendation information including the first keyword as the recommendation information corresponding to the first recommendation statement.

3. The method according to claim 1, wherein Before displaying the recommendation information corresponding to the first recommendation statement, it further includes: Obtaining a first vector corresponding to the first recommendation statement; Obtaining a second vector of any recommendation information in the recommendation information set; Determining the recommendation information corresponding to the first recommendation statement in the recommendation information set according to the similarity between the first vector and the second vector.

4. The method according to claim 1, wherein, The determining and displaying the extended recommendation information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information includes: Determining extended dimension information corresponding to the recommendation information according to the first interaction information corresponding to the recommendation information; Generating extended information corresponding to the extended dimension information; According to the extended information, obtaining at least one extended recommendation information related to the extended information in the recommendation information set, and using the at least one extended recommendation information as the extended recommendation information corresponding to the recommendation information; Displaying the extended recommendation information corresponding to the recommendation information.

5. The method according to claim 4, wherein, The obtaining at least one extended recommendation information related to the extended information in the recommendation information set includes: Search for the recommended information including the extended information in the set of recommended information, and use the recommended information including the extended information as at least one extended recommended information related to the extended information.

6. The method according to claim 4, wherein The obtaining of at least one extended recommended information related to the extended information from the set of recommended information includes: Obtaining a third vector corresponding to the extended information; Obtaining a second vector of any recommended information in the set of recommended information; Determining at least one extended recommended information related to the extended information in the set of recommended information according to the similarity between the third vector and the second vector.

7. The method according to any one of claims 4 to 6, wherein The method further includes: Obtaining third interaction information corresponding to the at least one extended recommended information; Sorting the at least one extended recommended information according to the third interaction information to obtain the sorted at least one extended recommended information; Determining at least one extended recommended information for display according to the sorted at least one extended recommended information.

8. The method according to claim 1, wherein the identifying of the user personality information by using a large model, generating and displaying a first recommendation statement corresponding to the user personality information includes: Using a large model to identify user historical consumption behavior information, user portrait information, and user interest point information, and generating and displaying a first recommendation statement corresponding to the user personality information.

9. The method according to claim 1, wherein The displaying of the recommended information corresponding to the first recommendation statement includes: Performing information extraction on the recommended information corresponding to the first recommendation statement to obtain the extracted recommended information; Displaying the extracted recommended information.

10. An information recommendation device, comprising: A recommendation statement generation unit, configured to use a large model to identify user personality information, and generate and display a first recommendation statement corresponding to the user personality information; The user personality information includes user historical consumption behavior information, user portrait information, and user interest point information, and the first recommendation statement is a recommendation statement corresponding to the user personalized information; An information display unit, configured to display the recommended information corresponding to the first recommendation statement when receiving a confirmation instruction for the first recommendation statement; A recommended information determination unit, configured to determine and display extended recommended information corresponding to the recommended information according to the first interaction information corresponding to the recommended information; The device further includes: A set obtaining unit, configured to obtain a first recommendation statement training set, where the first recommendation statement training set includes at least one labeled second recommendation statement; The set obtaining unit is further configured to use the self-instruct technology to perform prompt word expansion on the at least one labeled second recommendation statement to obtain a second recommendation statement training set; A model adjustment unit, configured to fine-tune the large model by using the second recommendation statement training set to obtain an adjusted large model; The information display unit is further configured to: Use a classification model to classify the set of comment information corresponding to the recommended information, and obtain a subset of comment information related to the recommended information according to the second interaction information corresponding to any comment information in the set of comment information; Using the large model and the prompt template, perform information extraction on any piece of comment information in the subset of comment information to obtain the extracted comment information; Display the comment information corresponding to the recommended information.

11. The apparatus according to claim 10, wherein, The information display unit is further configured to, before displaying the recommended information corresponding to the first recommendation statement: Obtain the first keyword corresponding to the first recommendation statement; Search for the recommended information including the first keyword in the set of recommended information, and use the recommended information including the first keyword as the recommended information corresponding to the first recommendation statement.

12. The apparatus according to claim 10, wherein, The information display unit is further configured to, before displaying the recommended information corresponding to the first recommendation statement: Obtain the first vector corresponding to the first recommendation statement; Obtain the second vector of any recommended information in the set of recommended information; Determine the recommended information corresponding to the first recommendation statement in the set of recommended information according to the similarity between the first vector and the second vector.

13. The apparatus according to claim 10, wherein, When the recommended information determination unit is configured to determine and display the extended recommended information corresponding to the recommended information according to the first interaction information corresponding to the recommended information, it is specifically configured to: Determine the extended dimension information corresponding to the recommended information according to the first interaction information corresponding to the recommended information; Generate extended information corresponding to the extended dimension information; According to the extended information, obtain at least one extended recommended information related to the extended information in the set of recommended information, and use the at least one extended recommended information as the extended recommended information corresponding to the recommended information; Display the extended recommended information corresponding to the recommended information.

14. The apparatus according to claim 13, wherein, When the recommended information determination unit is configured to obtain at least one extended recommended information related to the extended information in the set of recommended information, it is specifically configured to: Search for the recommended information including the extended information in the set of recommended information, and use the recommended information including the extended information as at least one extended recommended information related to the extended information.

15. The apparatus according to claim 13, wherein, When the recommended information determination unit is configured to obtain at least one extended recommended information related to the extended information in the set of recommended information, it is specifically configured to: Obtain the third vector corresponding to the extended information; Obtain the second vector of any recommended information in the set of recommended information; Determine at least one extended recommended information related to the extended information in the set of recommended information according to the similarity between the third vector and the second vector.

16. The apparatus according to any one of claims 13 to 15, wherein The recommended information determination unit is further configured to: Obtain the third interaction information corresponding to the at least one extended recommended information; Sort the at least one extended recommended information according to the third interaction information to obtain the sorted at least one extended recommended information; Determine at least one extended recommended information for display according to the sorted at least one extended recommended information.

17. The apparatus according to claim 10, wherein When the recommendation statement generation unit is configured to identify the user's personality information using a large model and generate and display the first recommendation statement corresponding to the user's personality information, it is specifically configured to: Using a large model to identify user historical consumption behavior information, user portrait information, and user interest point information, and generating and displaying a first recommendation statement corresponding to the user personality information.

18. The device according to claim 10, wherein, When the information display unit is used to display the recommendation information corresponding to the first recommendation statement, it is specifically used for: Performing information extraction on the recommendation information corresponding to the first recommendation statement to obtain the extracted recommendation information; Displaying the extracted recommendation information.

19. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to any one of claims 1-9.

20. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-9.

21. A computer program product, comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-9.

Images