Multi-screen interaction content recommendation method, device, equipment and storage medium

Abstract

The application relates to the technical field of multi-screen interaction, and discloses a multi-screen interaction content recommendation method, device, equipment and storage medium, the method comprising the following steps: when a family member at a first permission level initiates a multi-screen interaction request, intercepting a video stream segment corresponding to the multi-screen interaction request and sending the video stream segment to a family member at a second permission level, the second permission level being higher than the first permission level; and in response to a recommendation instruction fed back by the family member at the second permission level, performing content recommendation according to the recommendation instruction; since the application performs content recommendation based on multi-screen interaction and permission levels, the demand of a new interactive network television scene can be fully considered, and better multi-screen interaction experience can be provided for users.

Description

Technical Field

[0001] This invention relates to the field of multi-screen interaction technology, and in particular to a method, apparatus, device and storage medium for multi-screen interactive content recommendation. Background Technology

[0002] Currently, Internet Protocol Television (IPTV) offers high-quality services such as Virtual Reality (VR) videos, VR venues, free viewpoints, multi-viewpoints, and interaction between large and small screens, providing family members with a brand-new experience that is ultra-high-definition, immersive, personalized, and autonomous. Therefore, how to recommend IPTV content in multi-screen interactive scenarios to better meet user needs is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] The main objective of this invention is to provide a method, apparatus, device, and storage medium for recommending multi-screen interactive content, aiming to solve the technical problem of how to recommend IPTV content in multi-screen interactive scenarios and better meet user needs.

[0004] To achieve the above objectives, the present invention provides a multi-screen interactive content recommendation method, the multi-screen interactive content recommendation method comprising:

[0005] When a family member with the first permission level initiates a multi-screen interaction request, a video stream segment corresponding to the multi-screen interaction request is captured and sent to a family member with the second permission level, where the second permission level is higher than the first permission level.

[0006] In response to a recommendation instruction from a family member at the second permission level, content is recommended based on the recommendation instruction.

[0007] Optionally, the step of responding to a recommendation instruction from a family member at the second permission level and recommending content based on the recommendation instruction includes:

[0008] In response to recommended instructions from family members at the second permission level;

[0009] Generate content tags and calculate the content tag weights corresponding to the content tags;

[0010] Generate interest tags for family members and calculate the interest tag weights corresponding to the interest tags;

[0011] Calculate the similarity between family members;

[0012] Calculate the content attribute similarity between content items;

[0013] Content recommendations are made based on the content tag weights, interest tag weights, family member similarity, and content attribute similarity.

[0014] Optionally, generating content tags and calculating the content tag weights corresponding to the content tags includes:

[0015] Generate content tags and count the total number of times the content is recommended and the total number of times the content tags are recommended;

[0016] Detect whether the content category to which the content tag belongs is marked by a family member with the second permission level, and generate a correction factor based on the detection result;

[0017] The content tag weight corresponding to the content tag is calculated based on the total number of times the content is recommended, the total number of times the content tag is recommended, and the correction factor.

[0018] Optionally, generating interest tags for family members and calculating the interest tag weights corresponding to the interest tags includes:

[0019] Collect family member data and generate a set of interest tags for family members based on the family member data;

[0020] Obtain the number of times an interest tag appears in the set of interest tags for family members, and calculate the average number of occurrences;

[0021] Calculate the relationship intimacy level based on the relationships between family members, and calculate the average relationship intimacy level.

[0022] The interest tag weight is calculated based on the frequency of occurrence, the average value, and the average value assigned to the degree of relationship intimacy.

[0023] Optionally, calculating the similarity between family members includes:

[0024] Obtain the evaluation scores of family members for the content, and obtain the average score of family members for the content;

[0025] Obtain the evaluation time of family members regarding the content, and set a decay factor based on changes in family members' interests;

[0026] The similarity between family members is calculated based on the evaluation value, the average score, the evaluation time, and the decay factor.

[0027] Optionally, the calculation of content attribute similarity between contents includes:

[0028] Extract relevant text from the content and convert the relevant text into text vectors;

[0029] The similarity of content attributes between contents is calculated using a text similarity algorithm based on the text vectors.

[0030] Optionally, the multi-screen interactive content recommendation method further includes:

[0031] Acquire multi-screen interaction data of family members;

[0032] The permission levels of the family members are determined based on multi-screen interaction data, and the permission levels include a first permission level and a second permission level.

[0033] Furthermore, to achieve the above objectives, the present invention also proposes a multi-screen interactive content recommendation device, the multi-screen interactive content recommendation device comprising:

[0034] The sending module is used to intercept a video stream segment corresponding to the multi-screen interaction request and send it to a family member with a second permission level, where the second permission level is higher than the first permission level, when a family member with a first permission level initiates a multi-screen interaction request.

[0035] The recommendation module is used to respond to recommendation instructions from family members at the second permission level and to recommend content based on the recommendation instructions.

[0036] Furthermore, to achieve the above objectives, the present invention also proposes a multi-screen interactive content recommendation device, which includes a memory, a processor, and a multi-screen interactive content recommendation program stored in the memory and executable on the processor. The multi-screen interactive content recommendation program is configured to implement the multi-screen interactive content recommendation method described above.

[0037] Furthermore, to achieve the above objectives, the present invention also proposes a storage medium storing a multi-screen interactive content recommendation program, which, when executed by a processor, implements the multi-screen interactive content recommendation method as described above.

[0038] This invention discloses a method whereby, when a family member with a first-level access permission initiates a multi-screen interaction request, a video stream segment corresponding to the request is captured and sent to a family member with a second-level access permission. The second-level access permission is higher than the first-level access permission. The invention responds to recommendation instructions from family members with the second-level access permission and recommends content based on these instructions. Because this invention recommends content based on multi-screen interaction and access permission levels, it can fully consider the needs of new interactive network TV scenarios and thus provide users with a better multi-screen interaction experience. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the structure of a multi-screen interactive content recommendation device in the hardware operating environment involved in the embodiments of the present invention;

[0040] Figure 2 This is a flowchart illustrating the first embodiment of the multi-screen interactive content recommendation method of the present invention;

[0041] Figure 3 This is an overall interaction diagram of an embodiment of the multi-screen interactive content recommendation method of the present invention;

[0042] Figure 4 This is a flowchart illustrating the second embodiment of the multi-screen interactive content recommendation method of the present invention;

[0043] Figure 5 This is a flowchart illustrating the third embodiment of the multi-screen interactive content recommendation method of the present invention;

[0044] Figure 6 This is a structural block diagram of the first embodiment of the multi-screen interactive content recommendation device of the present invention.

[0045] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0046] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0047] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of a multi-screen interactive content recommendation device in the hardware operating environment involved in the embodiments of the present invention.

[0048] like Figure 1 As shown, the multi-screen interactive content recommendation device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen, and optionally, it may also include a standard wired interface or a wireless interface. In this invention, the wired interface of the user interface 1003 may be a USB interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or non-volatile memory (NVM), such as a disk storage device. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0049] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on multi-screen interactive content recommendation devices and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0050] like Figure 1 As shown, the memory 1005, which is identified as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a multi-screen interactive content recommendation program.

[0051] exist Figure 1 In the multi-screen interactive content recommendation device shown, the network interface 1004 is mainly used to connect to the backend server and communicate data with the backend server; the user interface 1003 is mainly used to connect to the user device; the multi-screen interactive content recommendation device calls the multi-screen interactive content recommendation program stored in the memory 1005 through the processor 1001 and executes the multi-screen interactive content recommendation method provided in the embodiment of the present invention.

[0052] Based on the above hardware structure, an embodiment of the multi-screen interactive content recommendation method of the present invention is proposed.

[0053] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the multi-screen interactive content recommendation method of the present invention, which presents the first embodiment of the multi-screen interactive content recommendation method of the present invention.

[0054] Step S10: When a family member with the first permission level initiates a multi-screen interaction request, a video stream segment corresponding to the multi-screen interaction request is captured and sent to a family member with the second permission level, where the second permission level is higher than the first permission level.

[0055] It should be understood that the execution subject of this embodiment may be a multi-screen interactive content recommendation device with data processing, network communication and program running functions, such as an IPTV terminal, or other electronic devices that can achieve the same or similar functions. This embodiment does not limit this. Among them, IPTV terminals include, but are not limited to, smart set-top boxes.

[0056] It is understood that the multi-screen interactive content recommendation method of the present invention is applied to multi-screen interactive scenarios, which include, but are not limited to, large and small screen interactive scenarios. For example, each family member uses a small screen terminal such as a mobile phone to interact with a large screen TV through an IPTV terminal such as a smart set-top box.

[0057] It should be understood that each family member has a permission level, which can be preset in the IPTV terminal system or determined by acquiring multi-screen interaction data from each family member's terminal. This embodiment does not impose any restrictions on this. The second permission level is higher than the first permission level; for example, the second permission level is a high permission level, and the first permission level is a low permission level.

[0058] Step S20: In response to the recommendation instruction from a family member at the second permission level, and recommend content according to the recommendation instruction.

[0059] For ease of understanding, please refer to Figure 3 This explanation does not limit the scope of this solution. Figure 3 This is an overall interaction diagram of an embodiment of the multi-screen interactive content recommendation method of the present invention. The diagram illustrates the specific steps of the multi-screen interactive content recommendation method as follows:

[0060] 1. When a family member with lower access level interacts with the large and small screens via IPTV, capture a segment of the current video stream and send it to the terminal of a family member with higher access level;

[0061] When a family member initiates IPTV for interaction between large and small screens via a mobile phone or other small-screen terminal, IPTV obtains the family member's identity information and determines their permission level based on the correspondence between the identity information and permission levels. If the family member has a low permission level, the IPTV terminal extracts a video stream segment within a preset time period and sends the video stream segment to the terminal of a family member with a higher permission level.

[0062] This scenario can be understood as follows: a child in a family uses a mobile phone to connect to a large-screen TV via an IPTV terminal to play cloud games, and the IPTV terminal can send game video stream clips to the parent's mobile phone.

[0063] 2. Respond to recommendation commands from family members with higher access levels to recommend content;

[0064] After watching a video stream clip, a family member with high privileges can select the "Recommend" button. In response to this recommendation, the IPTV terminal will launch the recommendation program.

[0065] This scenario can be understood as follows: when a family member with high access levels watches a video stream and deems the current game unsuitable for the family member currently playing it, they can use the recommendation button to have the IPTV terminal initiate a recommendation program to re-recommend suitable games for the current family member for multi-screen interaction.

[0066] Of course, in order to improve the efficiency and accuracy of content recommendation, in this embodiment, the server corresponding to the IPTV terminal can also perform content recommendation, and the IPTV terminal can distribute the recommendation results. This embodiment does not limit this.

[0067] In this embodiment, when a family member with a first permission level initiates a multi-screen interaction request, a video stream segment corresponding to the multi-screen interaction request is captured and sent to a family member with a second permission level. The second permission level is higher than the first permission level. The system responds to the recommendation instructions fed back by the family member with the second permission level and recommends content based on the recommendation instructions. Since this embodiment recommends content based on multi-screen interaction and permission levels, it can fully consider the needs of new interactive network TV scenarios and thus provide users with a better multi-screen interaction experience.

[0068] Reference Figure 4 , Figure 4 This is a flowchart illustrating the second embodiment of the multi-screen interactive content recommendation method of the present invention, based on the above. Figure 2 The first embodiment shown presents a second embodiment of the multi-screen interactive content recommendation method of the present invention.

[0069] In the second embodiment, step S20 includes:

[0070] Step S201: In response to a recommendation instruction from a family member at the second permission level.

[0071] It should be understood that in this embodiment, when making recommendations to users with low access levels, the similarity of family members and the similarity of content attributes are considered, and the similarity is fused by combining the weight of interest tags and the weight of content tags, thereby improving the accuracy of content recommendations.

[0072] Step S202: Generate content tags and calculate the content tag weights corresponding to the content tags.

[0073] Understandably, content tags can be generated for each piece of content. These tags can be built into the content itself, or natural language processing can be used to process the text of the content description to obtain multiple tags for each piece of content, forming a tag set. Taking video content as an example, the tags for video A could be: USA / Drama / Romance / Comedy.

[0074] It should be understood that the content tag weight corresponding to the content tag can be calculated based on a preset content tag weight value algorithm, where the preset content tag weight value algorithm can be set in advance.

[0075] Furthermore, in order to improve the accuracy of content tag weights, step S202 includes:

[0076] Generate content tags and count the total number of times content is recommended and the total number of times the content tags are recommended; detect whether the content category to which the content tag belongs is marked by family members with the second permission level, and generate a correction factor based on the detection result; calculate the content tag weight corresponding to the content tag based on the total number of times content is recommended, the total number of times the content tag is recommended, and the correction factor.

[0077] In one example, the weight w1 of each content tag is calculated based on the content category and the number of times the content tag is recommended. The content category includes games, videos, and education, etc. The number of times a tag is recommended is the number of times each tag is recommended within a preset time period. Specifically, this is reflected in the content recommendations. For example, if video A is recommended 5 times, then each of the video's four tags—American, drama, romance, and comedy—is recorded 5 times. The total number of times each content tag is recommended within the preset time period is calculated. The specific formula is as follows:

[0078]

[0079] Where S is the total number of content recommendations, S′' is the total number of times the content tag is recommended, and α is a correction factor, the value of which is determined by whether the content category to which the content tag belongs is marked by a high-privilege family member. If the content category is marked as liked, the value of α is greater than 1; if it is marked as disliked, it is less than 1; if it is not marked, it is equal to 1. Overall, the more times the content tag is recommended, the smaller the weight value; conversely, the fewer times the content tag is recommended, the higher the weight value.

[0080] Step S203: Generate interest tags for family members and calculate the interest tag weights corresponding to the interest tags.

[0081] It should be noted that the tags for family members are primarily interest tags. The collected family member data can be from IPTV or from smart home systems. With user authorization, the data can be sent to the IPTV terminal. Alternatively, the smart home system can calculate the interest tags for each family member and only send the tags to the IPTV terminal; this embodiment does not impose any restrictions on this.

[0082] It is understandable that the interest tag weight corresponding to the interest tag can be calculated by using a preset interest tag weight value algorithm, where the preset interest tag weight value algorithm can be set in advance.

[0083] Furthermore, to improve the accuracy of interest tag weights, step S203 includes:

[0084] Collect family member data and generate a set of interest tags for family members based on the family member data; obtain the number of times the interest tags appear in the set of interest tags for family members and calculate the average number of appearances; calculate the relationship intimacy value based on the relationship between family members and calculate the average relationship intimacy value; calculate the interest tag weight corresponding to the interest tag based on the number of appearances, the average value, and the average relationship intimacy value.

[0085] In one example, when an interest tag appears only in the tag set of one family member, the tag weight W2 = 1. When an interest tag appears in the interest tag sets of multiple family members, the weight W2 of each interest tag is calculated based on the relationship between the family members. Specifically, the relationships between family members are typically spouses, father and son, mother and son, grandfather and grandson, siblings, lovers, friends, cohabitants, etc., and can be divided into three levels—high, medium, and low—based on the degree of intimacy, assigned values ​​of 10, 6, and 4 respectively. The specific calculation formula is as follows:

[0086]

[0087] Where 'a' represents the number of times the interest tag appears in the family member's interest tag set, 'a' represents the average number of occurrences, 'C' represents the average value assigned to the relationship intimacy level, and 'C0' is a preset threshold. The average value assigned to the relationship intimacy level, 'C', is calculated based on the relationship intimacy level. For example, if the tag "games" appears in the tag sets of four family members (A, B, C, and D), the number of occurrences is 4, and there are 6 relationships. The average value assigned to the relationship is the average of the 6 values.

[0088] Step S204: Calculate the similarity between family members.

[0089] It is understandable that calculating the similarity between family members can involve obtaining the evaluation values ​​of family members for the content and obtaining the average rating of family members for the content; obtaining the evaluation time of family members for the content and setting a decay factor based on the changes in family members' interests; and calculating the similarity between family members based on the evaluation values, the average rating, the evaluation time, and the decay factor.

[0090] It should be understood that, considering that family members’ ratings of content can change over time, a user’s recent rating of a certain content can reflect their recent interests. Therefore, the impact of time on similarity is taken into account when calculating the similarity of family members.

[0091] In one example, family member similarity is calculated using the following formula:

[0092]

[0093] In the formula, R ui R vi These are the ratings given by users u and v to the content. as well as t represents the average rating of the content by users u and v. u，xi t represents the time it takes for user u to rate a piece of content. v，xi Let δ be the time it takes for user v to rate a piece of content, and δ be the decay coefficient. A larger δ indicates that family members' interests change more rapidly, while a smaller δ indicates that family members' interests do not change much. The similarity among family members obtained by combining the decay coefficient more accurately reflects the similarity between real users.

[0094] Step S205: Calculate the content attribute similarity between content items.

[0095] It should be understood that calculating the content attribute similarity between content can involve extracting relevant text from the content and converting the relevant text into text vectors; then calculating the content attribute similarity between the content using a text similarity algorithm based on the text vectors.

[0096] First, extract relevant text from the content. For example, for video content, extract keyframes and use OCR algorithms to recognize the text of the keyframes and extract keywords; or directly extract keywords from the content's introduction, description, and other text to obtain a keyword set for each piece of content. Then, the keyword set of the content can be analyzed using the TF-IDF algorithm and converted into text vectors. Finally, a text similarity algorithm, such as the cosine similarity algorithm, is used to calculate the content attribute similarity sim2 between the content.

[0097] Step S206: Recommend content based on the content tag weight, the interest tag weight, the family member similarity, and the content attribute similarity.

[0098] In one example, the similarity after fusion is calculated using the following formula:

[0099] sim = W2sim1 + W1sim2

[0100] In the formula, sim represents the fused similarity, w2 represents the interest tag weight corresponding to the interest tag, sim1 represents the family member similarity, w1 represents the content tag weight corresponding to the content tag, and sim2 represents the content attribute similarity.

[0101] Subsequently, the nearest neighbors of the target family members are selected based on the fused similarity. The rating values ​​of the target family members for each item are predicted based on the rating values ​​of all users in the nearest neighbors. The top N content is selected as the content recommendation result.

[0102] In this embodiment, considering the similarity of family members and the similarity of content attributes, the fused similarity matrix can make recommendations more accurate. At the same time, during the fusion process, the weights of family member tags and content tags are combined, which further improves the accuracy of content recommendations.

[0103] Reference Figure 5 , Figure 5 This is a flowchart illustrating the third embodiment of the multi-screen interactive content recommendation method of the present invention. Based on the above embodiments, the third embodiment of the multi-screen interactive content recommendation method of the present invention is proposed.

[0104] In the third embodiment, before step S10, the method further includes:

[0105] Step S01: Obtain multi-screen interaction data of family members.

[0106] It should be understood that, in order to improve the reliability of permission levels, in this embodiment, the permission level of family members is determined based on the multi-screen interaction data of family members.

[0107] It should be noted that the multi-screen interaction data can be the data of various family members using IPTV terminals to interact with each other within a preset time period in the past. The preset time period can be set in advance.

[0108] Step S02: Determine the permission level of the family member based on the multi-screen interaction data. The permission level includes a first permission level and a second permission level.

[0109] In one example, data on IPTV multi-screen interactions among family members over a past period can be obtained. Permission levels are determined based on content category and data object. Content categories can include games, audio-visual content, and educational content. Family members with a higher frequency of audio-visual interactions have higher permissions. Similarly, for game-related interactions, permission levels can be determined based on the game's age requirements (i.e., the data object). For instance, in multi-screen interactions within the same game category, if user A's game is 18+ and user B's game is 4+, then user A's permission level is higher than user B's.

[0110] In this embodiment, the permission level of family members is determined based on the multi-screen interaction data of family members, thereby improving the reliability of the permission level.

[0111] In addition, refer to Figure 6 This invention also proposes a multi-screen interactive content recommendation device, which includes:

[0112] The sending module 10 is used to capture a video stream segment corresponding to the multi-screen interaction request and send it to a family member with a second permission level, where the second permission level is higher than the first permission level, when a family member with a first permission level initiates a multi-screen interaction request.

[0113] It is understood that the multi-screen interactive content recommendation device of the present invention is applied to multi-screen interactive scenarios, which include, but are not limited to, large and small screen interactive scenarios. For example, each family member uses a small screen terminal such as a mobile phone to interact with a large screen TV through an IPTV terminal such as a smart set-top box.

[0114] It should be understood that each family member has a permission level, which can be preset in the IPTV terminal system or determined by acquiring multi-screen interaction data from each family member's terminal. This embodiment does not impose any restrictions on this. The second permission level is higher than the first permission level; for example, the second permission level is a high permission level, and the first permission level is a low permission level.

[0115] The recommendation module 20 is used to respond to recommendation instructions from family members at the second permission level and to recommend content based on the recommendation instructions.

[0116] For ease of understanding, please refer to Figure 3 This explanation does not limit the scope of this solution. Figure 3 This is an overall interaction diagram of an embodiment of the multi-screen interactive content recommendation method of the present invention. The diagram illustrates the specific steps of the multi-screen interactive content recommendation method as follows:

[0117] 1. When a family member with lower access level interacts with the large and small screens via IPTV, capture a segment of the current video stream and send it to the terminal of a family member with higher access level;

[0118] When a family member initiates IPTV for interaction between large and small screens via a mobile phone or other small-screen terminal, IPTV obtains the family member's identity information and determines their permission level based on the correspondence between the identity information and permission levels. If the family member has a low permission level, the IPTV terminal extracts a video stream segment within a preset time period and sends the video stream segment to the terminal of a family member with a higher permission level.

[0119] This scenario can be understood as follows: a child in a family uses a mobile phone to connect to a large-screen TV via an IPTV terminal to play cloud games, and the IPTV terminal can send game video stream clips to the parent's mobile phone.

[0120] 2. Respond to recommendation commands from family members with higher access levels to recommend content;

[0121] After watching a video stream clip, a family member with high privileges can select the "Recommend" button. In response to this recommendation, the IPTV terminal will launch the recommendation program.

[0122] This scenario can be understood as follows: when a family member with high access levels watches a video stream and deems the current game unsuitable for the family member currently playing it, they can use the recommendation button to have the IPTV terminal initiate a recommendation program to re-recommend suitable games for the current family member for multi-screen interaction.

[0123] Of course, in order to improve the efficiency and accuracy of content recommendation, in this embodiment, the server corresponding to the IPTV terminal can also perform content recommendation, and the IPTV terminal can distribute the recommendation results. This embodiment does not limit this.

[0124] In this embodiment, when a family member with a first permission level initiates a multi-screen interaction request, a video stream segment corresponding to the multi-screen interaction request is captured and sent to a family member with a second permission level. The second permission level is higher than the first permission level. The system responds to the recommendation instructions fed back by the family member with the second permission level and recommends content based on the recommendation instructions. Since this embodiment recommends content based on multi-screen interaction and permission levels, it can fully consider the needs of new interactive network TV scenarios and thus provide users with a better multi-screen interaction experience.

[0125] In one embodiment, the recommendation module 20 is further configured to respond to recommendation instructions from family members at the second permission level; generate content tags and calculate the content tag weights corresponding to the content tags; generate interest tags for family members and calculate the interest tag weights corresponding to the interest tags; calculate the family member similarity between family members; calculate the content attribute similarity between content; and recommend content based on the content tag weights, the interest tag weights, the family member similarity, and the content attribute similarity.

[0126] In one embodiment, the recommendation module 20 is further configured to generate content tags, and count the total number of times content is recommended and the total number of times the content tags are recommended; detect whether the content category to which the content tag belongs is marked by a family member of the second permission level, and generate a correction factor based on the detection result; and calculate the content tag weight corresponding to the content tag based on the total number of times content is recommended, the total number of times the content tags are recommended, and the correction factor.

[0127] In one embodiment, the recommendation module 20 is further configured to collect family member data and generate a set of interest tags for family members based on the family member data; obtain the number of times the interest tags appear in the set of interest tags for family members and calculate the average number of appearances; calculate the relationship intimacy assignment based on the relationship between family members and calculate the average relationship intimacy assignment; and calculate the interest tag weight corresponding to the interest tag based on the number of appearances, the average value, and the average relationship intimacy assignment.

[0128] In one embodiment, the recommendation module 20 is further configured to obtain the evaluation values ​​of family members for the content, and obtain the average rating of family members for the content; obtain the evaluation time of family members for the content, and set a decay factor according to the changes in the interests of family members; and calculate the similarity between family members based on the evaluation values, the average rating, the evaluation time, and the decay factor.

[0129] In one embodiment, the recommendation module 20 is further configured to extract relevant text of the content and convert the relevant text into text vectors; and calculate the content attribute similarity between the content based on the text vectors using a text similarity algorithm.

[0130] In one embodiment, the multi-screen interactive content recommendation device further includes:

[0131] The permission level module is used to acquire multi-screen interaction data of family members; and to determine the permission level of the family members based on the multi-screen interaction data, wherein the permission level includes a first permission level and a second permission level.

[0132] Other embodiments or specific implementations of the multi-screen interactive content recommendation device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0133] Furthermore, this embodiment of the invention also proposes a storage medium storing a multi-screen interactive content recommendation program, which, when executed by a processor, implements the multi-screen interactive content recommendation method as described above.

[0134] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0135] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0136] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as a read-only memory image (ROM) / random access memory (RAM), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0137] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A multi-screen interactive content recommendation method, characterized in that, The multi-screen interactive content recommendation method includes: When a family member with the first permission level initiates a multi-screen interaction request, a video stream segment corresponding to the multi-screen interaction request is captured and sent to a family member with the second permission level, where the second permission level is higher than the first permission level. The multi-screen interaction content recommendation method is applied to an IPTV terminal. The application scenario of the multi-screen interaction content recommendation method is that family members use small-screen terminals and interact with a large-screen TV through an IPTV terminal. In response to a recommendation instruction from a family member at the second permission level, and based on the recommendation instruction, content is recommended. The content recommendation involves re-recommending suitable content for multi-screen interaction to family members at the first permission level.

2. The multi-screen interactive content recommendation method as described in claim 1, characterized in that, The response to the recommendation instruction from a family member at the second permission level, and the content recommendation based on the recommendation instruction, includes: In response to recommended instructions from family members at the second permission level; Generate content tags and calculate the content tag weights corresponding to the content tags; Generate interest tags for family members and calculate the interest tag weights corresponding to the interest tags; Calculate the similarity between family members; Calculate the content attribute similarity between content items; Content recommendations are made based on the content tag weights, interest tag weights, family member similarity, and content attribute similarity.

3. The multi-screen interactive content recommendation method as described in claim 2, characterized in that, The process of generating content tags and calculating the content tag weights corresponding to the content tags includes: Generate content tags and count the total number of times the content is recommended and the total number of times the content tags are recommended; Detect whether the content category to which the content tag belongs is marked by a family member with the second permission level, and generate a correction factor based on the detection result; The content tag weight corresponding to the content tag is calculated based on the total number of times the content is recommended, the total number of times the content tag is recommended, and the correction factor.

4. The multi-screen interactive content recommendation method as described in claim 2, characterized in that, The process of generating interest tags for family members and calculating the weights of those interest tags includes: Collect family member data and generate a set of interest tags for family members based on the family member data; Obtain the number of times an interest tag appears in the set of interest tags for family members, and calculate the average number of occurrences; Calculate the relationship intimacy level based on the relationships between family members, and calculate the average relationship intimacy level. The interest tag weight is calculated based on the frequency of occurrence, the average value, and the average value assigned to the degree of relationship intimacy.

5. The multi-screen interactive content recommendation method as described in claim 2, characterized in that, The calculation of family member similarity among family members includes: Obtain the evaluation scores of family members for the content, and obtain the average score of family members for the content; Obtain the evaluation time of family members regarding the content, and set a decay factor based on changes in family members' interests; The similarity between family members is calculated based on the evaluation value, the average score, the evaluation time, and the decay factor.

6. The multi-screen interactive content recommendation method as described in claim 2, characterized in that, The calculation of content attribute similarity between content includes: Extract relevant text from the content and convert the relevant text into text vectors; The similarity of content attributes between contents is calculated using a text similarity algorithm based on the text vectors.

7. The multi-screen interactive content recommendation method as described in any one of claims 1 to 6, characterized in that, The multi-screen interactive content recommendation method also includes: Acquire multi-screen interaction data of family members; The permission levels of the family members are determined based on multi-screen interaction data, and the permission levels include a first permission level and a second permission level.

8. A multi-screen interactive content recommendation device, characterized in that, The multi-screen interactive content recommendation device includes: The sending module is used to intercept the video stream segment corresponding to the multi-screen interaction request and send it to the family member with the second permission level when the family member with the first permission level initiates the multi-screen interaction request. The second permission level is higher than the first permission level. The multi-screen interaction content recommendation device is applied to the IPTV terminal. The application scenario of the multi-screen interaction content recommendation device is that family members use small-screen terminals and interact with large-screen TVs through IPTV terminals. The recommendation module is used to respond to recommendation instructions from family members with the second permission level, and to recommend content based on the recommendation instructions. The content recommendation is to recommend suitable content for multi-screen interaction by re-recommending family members with the first permission level.

9. A multi-screen interactive content recommendation device, characterized in that, The multi-screen interactive content recommendation device includes: a memory, a processor, and a multi-screen interactive content recommendation program stored in the memory and executable on the processor. When the multi-screen interactive content recommendation program is executed by the processor, it implements the multi-screen interactive content recommendation method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a multi-screen interactive content recommendation program, which, when executed by a processor, implements the multi-screen interactive content recommendation method as described in any one of claims 1 to 7.

Images