Smart television personalized content recommendation method, system and device and storage medium

Abstract

The invention relates to the technical field of smart televisions, and particularly provides a smart television personalized content recommendation method, system and device and a storage medium, and the method comprises the steps: obtaining user basic data and behavior record data, the user basic data comprising user personal information and device information; extracting browsing content and browsing time from the behavior record data, and setting the browsing time as interestingness of the browsing content; analyzing the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interestingness, counting a core feature tag with the highest interestingness; video resources associated with the core feature tags serve as candidate resources, and priorities are divided for the candidate resources based on user basic data; and arranging the candidate resources from first to second according to the priority, and pushing the candidate resources according to the arrangement sequence. According to the invention, the recommendation accuracy and the user experience are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart televisions, and in particular relates to a method, system, device and storage medium for recommending personalized content for smart televisions. Background Art

[0002] With the popularity of smart TVs, TV content providers, advertisers, and users themselves all have a stronger demand for the personalization of smart TVs. The current market status of smart TVs, whether it is advertising or TV content recommendations, is relatively single-minded and cannot effectively attract user attention, resulting in a poor user experience.

[0003] Some intelligent recommendation algorithms only focus on the user's historical behavior data, and the pushed content is limited to TV programs. Therefore, the accuracy and content of the recommendations are greatly limited. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method, system, device and storage medium for recommending personalized content for a smart TV to solve the above-mentioned technical problems.

[0005] In a first aspect, the present invention provides a method for recommending personalized content for a smart TV, comprising: Obtain user basic data and behavior record data, wherein the user basic data includes user personal information and device information; Extracting browsing content and browsing time from the behavior record data, and setting the browsing time as the interest level of the browsing content; Analyze the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interest level, count the core feature tags with the highest interest level; The video resources associated with the core feature tags are used as candidate resources, and the candidate resources are prioritized based on the user's basic data; Arrange candidate resources from first to last according to priority, and push candidate resources in the order of arrangement.

[0006] In an optional embodiment, the method further comprises: Collect historical order information; Parse the product tags of the transaction products from the historical orders, and filter out the core product tags with the highest interest by counting the number of times the product tags appear; Filter matching advertising resources based on core feature tags and core product tags; Push matching advertising resources during advertising periods.

[0007] In an optional implementation, obtaining user basic data and behavior record data includes: Obtaining registration data of a user account, and extracting user personal information from the registration data, the user personal information including gender and age; Obtain the device information of the user account logged in, the device information includes the device model, device brand and device price.

[0008] In an optional implementation, the feature tag combination of the browsed content is parsed, and based on the feature tag combination of the browsed content and the corresponding interest level, the core feature tag with the highest interest level is counted, including: Parse the type, subject matter and participant information of the browsed content, convert the type, subject matter and participant information into feature tags in JSON format, and summarize the feature tags into feature tag combinations; The interest level of the browsed content is taken as the interest level of all feature tags in the corresponding feature tag combination; The interest levels of the same feature tag in different feature tag combinations are accumulated to obtain the interest levels of all feature tags; Normalize the interest of all feature labels to obtain the interest coefficient of the feature label; The feature labels whose interest coefficients reach the set interest threshold are selected as core feature labels.

[0009] In an optional implementation, video resources associated with core feature tags are used as candidate resources, and the candidate resources are prioritized based on user basic data, including: Generate feature tags for video resources in the resource pool in advance; Calculate the matching degree between the core feature label and the feature label of the video resources in the resource pool; Screening video resources whose matching degree reaches a set matching degree threshold as candidate resources; The correspondence between the user basic data and the feature tags is preset, and the candidate resources whose feature tags have a correspondence with the user basic data are set to a high priority, and the candidate resources whose feature tags do not have a correspondence with the user basic data are set to a low priority.

[0010] In a second aspect, the present invention provides a smart TV personalized content recommendation system, comprising: A data acquisition module is used to acquire basic user data and behavior record data, wherein the basic user data includes user personal information and device information; A behavior extraction module, used to extract browsing content and browsing time from the behavior record data, and set the browsing time as the interest level of the browsing content; A behavior analysis module is used to analyze the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interest level, count the core feature tags with the highest interest level; A resource screening module is used to select video resources associated with core feature tags as candidate resources and prioritize the candidate resources based on user basic data; The push control module is used to arrange the candidate resources from first to last according to the priority, and push the candidate resources in the arrangement order.

[0011] In an optional embodiment, the system further includes: Order collection module, used to collect historical order information; The order parsing module is used to parse the product tags of transaction products from historical orders and filter out the core product tags with the highest interest by counting the number of occurrences of product tags; Type matching module, used to filter matching advertising resources based on core feature tags and core product tags; The advertisement push module is used to push matching advertisement resources during the advertisement period.

[0012] In an optional embodiment, the data acquisition module includes: A personal information acquisition unit, used to acquire registration data of a user account and extract user personal information from the registration data, wherein the user personal information includes gender and age; The device information acquisition unit is used to acquire the device information of the user account login, and the device information includes the device model, device brand and device price.

[0013] In a third aspect, a device is provided, comprising: A memory, used for storing a personalized content recommendation program for a smart TV; The processor is used to implement the steps of the smart TV personalized content recommendation method provided in the first aspect when executing the smart TV personalized content recommendation program.

[0014] In a fourth aspect, a computer-readable storage medium is provided, on which a smart TV personalized content recommendation program is stored. When the smart TV personalized content recommendation program is executed by a processor, the steps of the smart TV personalized content recommendation method provided in the first aspect are implemented.

[0015] The beneficial effects of the present invention lie in that the method, system, device and storage medium for personalized content recommendation for smart TVs provided by the present invention determine core feature tags through behavior record data, screen out candidate resources based on the core feature tags, and then sort the candidate resources based on user basic data, manage the interface layout based on the sorting of the candidate resources, thereby improving the accuracy of recommendations and user experience, and comprehensively considering shopping orders for advertising push, further improving the user experience.

[0016] In addition, the invention has a reliable design principle, a simple structure and a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention.

[0019] Figure 2 is a schematic block diagram of a system according to an embodiment of the present invention.

[0020] Figure 3 A schematic diagram of the structure of a device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0023] The intelligent television personalized content recommendation method provided by the embodiment of the present invention is executed by a computer device, and accordingly, the intelligent television personalized content recommendation system runs in the computer device.

[0024] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention. Figure 1 The execution subject may be a smart TV personalized content recommendation system. According to different requirements, the order of the steps in the flow chart may be changed, and some may be omitted.

[0025] like Figure 1 As shown, the method includes: S1. Obtain user basic data and behavior record data, wherein the user basic data includes user personal information and device information.

[0026] Obtain user basic data and behavior record data. The user basic data includes but is not limited to the user's personal information (such as name, age, gender, occupation, etc.) and device information (such as device model, operating system version, network status, etc.). This information provides a basic basis for subsequent personalized recommendations.

[0027] S2. Extract browsing content and browsing time from the behavior record data, and set the browsing time as the interest level of the browsing content.

[0028] The user's browsing content (specifically which web pages, videos, articles, etc. were browsed) and browsing time (how long the user stayed on these contents) are carefully extracted from the behavioral record data, and the length of browsing time is used as an indicator to measure the user's interest in the browsing content, that is, the longer the browsing time, the higher the user's interest.

[0029] S3. Analyze the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interest level, count the core feature tags with the highest interest level.

[0030] In-depth analysis of the feature tag combinations of the browsed content, which may include multiple dimensions such as the content's theme, type, style, and emotional color. Then, based on the feature tag combinations of the browsed content and their corresponding interest levels, statistical and machine learning algorithms are used to calculate the core feature tags with the highest user interest levels. These tags best represent the user's preferences and interests.

[0031] S4. Take the video resources associated with the core feature tags as candidate resources and prioritize the candidate resources based on the user basic data.

[0032] Based on the core feature tags, we filter out the video resources related to them as candidate resources in the video resource library. Then, we further prioritize the candidate resources based on the basic user data (such as the user's age, gender, occupation, etc.). For example, young users may prefer to be pushed fashion and entertainment videos, while middle-aged and elderly users may prefer health and wellness content.

[0033] S5. Arrange the candidate resources from first to last according to the priority, and push the candidate resources in the order of arrangement.

[0034] The selected candidate resources are arranged from high to low priority to form a personalized recommendation list. Then, according to this order of arrangement, these candidate resources are pushed to users through their devices or platforms, such as mobile apps and web pages, to ensure that users can first receive the content they are most likely to be interested in, thereby improving user experience and satisfaction.

[0035] In an embodiment of the present invention, based on step S1, a possible embodiment is given below to illustrate its specific implementation scheme in a non-limiting manner.

[0036] In a specific example, the method of obtaining basic data includes: S101. Obtaining registration data of user account User personal information is extracted from the registration data, which is voluntarily provided by the user to ensure the personalization and security of the service. The user personal information includes but is not limited to gender and age. Gender information helps to understand the user's preferences and interests, such as certain content or services may be more popular with users of a certain gender; age information helps to judge the user's maturity, life stage and possible areas of interest.

[0037] S102. Obtaining user basic data and behavior record data On the basis of the original, this step now also includes gender and age information extracted from registration data. This information, together with device information, constitutes a complete view of the user's basic data. User personal information: gender, age (from registration data); device information: device model, operating system version, network status, etc. (collected when the user logs in or uses the service); behavior record data: record all user behaviors on the platform, including browsing, clicking, purchasing, commenting, etc., for subsequent analysis of user interests and preferences.

[0038] In an embodiment of the present invention, based on step S2, a possible embodiment is given below to illustrate its specific implementation scheme in a non-limiting manner.

[0039] In a specific example, the interest degree is calculated as follows: These records are collected through the background log system, including user ID, browsed content URL, start browsing timestamp, end browsing timestamp and other information.

[0040] Remove invalid data: Delete records that are missing key information (such as user ID, browsed content URL, etc.).

[0041] Clean noise data: Correct or delete records with abnormal timestamps (such as the start time is later than the end time).

[0042] Data format conversion: Convert timestamps to a date-time format that is easier to process.

[0043] Parse URLs: Use regular expressions or URL parsing libraries to extract information such as article titles, authors, and publishing times. For web page content, you may need to use web crawler technology to obtain the complete HTML page and extract the required information from it.

[0044] Content classification: Based on the extracted information, the browsing content is classified into different categories (such as movies, TV series, variety shows, news, games, etc.).

[0045] Calculate browsing time: For each record, calculate the difference between the start browsing timestamp and the end browsing timestamp to get the browsing time (in seconds).

[0046] In an embodiment of the present invention, based on step S3, a possible embodiment is given below to illustrate its specific implementation scheme in a non-limiting manner.

[0047] S301. Analyze browsing content and generate feature tag combinations After obtaining the user's browsing content, it is necessary to conduct in-depth analysis of the content to extract key types, themes and participant information. This information will be used as feature tags for subsequent user interest analysis and personalized recommendations.

[0048] Type: The types of content you browse may include news, movies, TV series, variety shows, documentaries, short videos, etc. These types reflect the basic attributes and forms of the content.

[0049] Theme: The theme focuses more on the subject and field involved in the content, such as science fiction, romance, war, history, suspense, etc. The choice of theme often reflects the user's specific interests and preferences.

[0050] Participants: Information about participants includes directors, actors, hosts, authors, etc. These personnel are closely related to the quality and style of the content, and are also important factors affecting user interest.

[0051] In order to express this information in a structured manner, it is converted into feature tags in JSON format. For example: "content_id": "xxx", / / Unique identifier of the browsed content "feature_labels": "type": "Movie", "genre": "Science Fiction", "participants": [ {"name": "Director A", "role": "Director"}, {"name": "Actor B", "role": "Starring"}.

[0052] In this JSON object, content_id is the unique identifier of the browsed content, and feature_labels is an array containing multiple feature labels. Each feature label contains type, subject matter, and participant information.

[0053] S302. Using the interest of the browsing content as the interest of the feature tag combination After parsing the feature tag combinations, the interest of the browsed content needs to be used as the interest of these feature tag combinations. This can be achieved by counting the user's interactive behaviors such as the length of time they stay on each browsed content, the number of clicks, and the number of comments. Specifically, these interactive behaviors are quantified into interest scores and assigned to the corresponding feature tag combinations.

[0054] S303. Accumulate the interest of the same feature tag and normalize it Since users may browse multiple contents, and these contents may contain the same feature tags (such as the genres are all movies and the themes are all science fiction), it is necessary to accumulate the interest of the same feature tag in different feature tag combinations. The accumulated interest reflects the importance of the feature tag in the user's interest.

[0055] In order to eliminate the differences in interest metrics between different feature labels, the accumulated interest needs to be normalized. There are many methods for normalization, such as minimum-maximum normalization, Z-score normalization, etc. Here, the minimum-maximum normalization method is selected to map the interest to the interval [0,1].

[0056] The normalized interest degree is called the interest coefficient of the feature tag, which reflects the relative importance of the feature tag in the user's interest.

[0057] S304. Filter core feature tags Finally, it is necessary to screen out feature tags whose interest coefficients reach the set interest threshold as core feature tags. These core feature tags best represent the user's preferences and interests and will be used for subsequent personalized recommendations.

[0058] The interest threshold can be adjusted according to actual needs. Generally speaking, a reasonable threshold is set based on the user's historical behavior data and recommendation results. By continuously adjusting and optimizing the threshold, the accuracy of recommendations and user satisfaction can be improved.

[0059] In an embodiment of the present invention, based on step S4, a possible embodiment is given below to illustrate its specific implementation scheme in a non-limiting manner.

[0060] In a specific example, the video resources associated with the core feature tags are used as candidate resources, and the specific process of prioritizing the candidate resources based on the user basic data is as follows: Before building a personalized recommendation system, the video resources in the resource pool are deeply analyzed and labeled. This step is crucial because it is directly related to the accuracy and efficiency of the subsequent recommendation algorithm.

[0061] Specifically, a set of feature labels is generated for each video resource. These labels should be able to fully and accurately describe the content, type, style, emotional color and other dimensions of the video. These feature labels can be generated through manual annotation, automatic classification algorithms or machine learning models.

[0062] For example, for a movie, the following feature labels are generated for it: Content: Science fiction, action, romance Type: Movie Style: thrilling, exciting Emotional color: positive, active These feature labels will serve as important inputs for subsequent recommendation algorithms to measure the degree of match between video resources and user interest.

[0063] Calculate the matching degree between the core feature labels and the feature labels of the video resources in the resource pool After obtaining the core feature tags of the user and the feature tags of the video resources in the resource pool, the matching degree between them is calculated. This step is usually achieved by calculating the similarity between the two tag sets.

[0064] There are many methods for calculating similarity, such as cosine similarity, Jaccard similarity, Pearson correlation coefficient, etc. Here, a similarity calculation method suitable for the application scenario is selected, and the matching degree between each video resource and the user's core feature label is calculated based on this method.

[0065] The calculation results of the matching degree will be used to measure the correlation between the video resources and the user's interest, thereby providing a basis for subsequent screening and sorting.

[0066] Filter video resources whose matching degree reaches the set matching degree threshold as candidate resources After calculating the matching degree, a reasonable matching degree threshold is set to filter video resources that are highly relevant to the user's interest as candidate resources.

[0067] The matching threshold can be adjusted according to actual needs. Generally speaking, a reasonable threshold is set based on factors such as the user's historical behavior data, recommendation effect, and the number of video resources in the resource pool. By continuously adjusting and optimizing the threshold, the accuracy of recommendations and user satisfaction can be improved.

[0068] After the video resources whose matching degree reaches or exceeds the threshold are screened as candidate resources, they will enter the next step of priority setting.

[0069] Pre-set the correspondence between user basic data and feature tags, and set the priority of candidate resources After selecting candidate resources, we further prioritize them based on basic user data (such as user age, gender, occupation, etc.). The purpose of this step is to ensure that users receive the content they are most likely to be interested in first, thereby improving user experience and satisfaction.

[0070] Specifically, the correspondence between user basic data and feature labels is pre-set, such as: Young users: They prefer to be pushed videos about fashion and entertainment (corresponding feature tags: fashion, entertainment) Middle-aged and elderly users: prefer health and wellness content (corresponding feature tags: health, wellness) Then, the candidate resources whose characteristic tags have a corresponding relationship with the user basic data are set to high priority, and the candidate resources whose characteristic tags do not have a corresponding relationship with the user basic data are set to low priority. In this way, in the subsequent recommendation list, the high-priority resources will be ranked first and displayed to the user first.

[0071] Through the above expansion content, a series of steps are implemented, including feature labeling of video resources in the resource pool, calculating the matching degree between core feature labels and video resource feature labels, screening candidate resources, and setting the priority of candidate resources. These steps together constitute an important part of the personalized recommendation system and provide a solid foundation for subsequent recommendation algorithms. On the basis of this framework, we further combine specific business scenarios and user needs to provide users with more accurate and personalized recommendation services.

[0072] In an embodiment of the present invention, based on step S5, a possible embodiment is given below to illustrate its specific implementation in a non-limiting manner.

[0073] Push candidate resources to users in the order of the list. This usually involves the following steps: Resource preparation: Ensure that all candidate resources are usable and properly formatted and optimized for different devices and platforms.

[0074] Push strategy formulation: formulate push strategies based on user preferences and platform characteristics. For example, for mobile device users, you can choose to push resources during the user's active time period; for web users, you can push through pop-ups, sidebars, etc.

[0075] Push fulfillment: Pushing resources to users through their devices or platforms in the order of the priority list. This can be achieved through API calls, messaging systems, or third-party services.

[0076] Effect monitoring and feedback: After push notifications are sent, user behavior data (such as click-through rate, dwell time, conversion rate, etc.) needs to be monitored to evaluate push notification effects. At the same time, user feedback is collected to optimize subsequent recommendation algorithms and push notification strategies.

[0077] On the basis of the above embodiments, in order to further improve the user experience, in one embodiment, push management is performed on advertising resources. The specific management method includes: In order to more fully understand the interests and needs of users, it is necessary to collect historical order information of users. This information includes the types, quantities, prices, purchase times, etc. of products purchased by users. Historical order data can reflect users' consumption habits, purchasing power and preferences, and is an important basis for subsequent analysis and recommendations.

[0078] In actual operation, the order information can be obtained through the user's account system, e-commerce platform or payment platform, etc. At the same time, it is necessary to ensure the accuracy and completeness of the data and comply with relevant privacy policies and laws and regulations.

[0079] After collecting historical order information, each transaction product needs to be labeled. Product labels are an abstraction and summary of product attributes, functions, categories and other information, which can reflect the core features and selling points of the product.

[0080] In order to filter out the core product tags with the highest interest, we can count the number of times each product tag appears in historical orders. The more tags appear, the higher the user's interest in the product. Therefore, we can use these high-frequency tags as core product tags for subsequent advertising resource screening and push.

[0081] After obtaining the core feature tags and core product tags, you can filter matching advertising resources based on these tags. The selection of advertising resources should follow the following principles: Relevance: Advertising content should be highly relevant to core feature tags and core product tags to ensure that the advertisement can accurately reach users' interests.

[0082] Diversity: While maintaining relevance, advertising content should have a certain degree of diversity to prevent users from falling into an information cocoon.

[0083] Timeliness: Advertising content should keep pace with market trends and changes in user needs to ensure the timeliness and attractiveness of the advertisements.

[0084] After determining the matching advertising resources, you need to select the appropriate advertising time slot for delivery. The selection of advertising time slots should take into account the following factors: User active time: Based on the user's historical behavior data, determine the time period when the user is most active, such as evenings and weekends.

[0085] Advertising effectiveness evaluation: Evaluate the effectiveness of different advertising time periods through A / B testing and other methods, and select the time period with the best effect for delivery.

[0086] Competitive environment: Consider competitors’ advertising strategies and market environment, and avoid pushing ads during periods of intensive advertising to reduce competitive pressure.

[0087] When pushing ads, you also need to pay attention to the frequency and form of ads to avoid excessively disturbing users or causing them to be disgusted. At the same time, you should provide convenient options to close or skip ads to respect the user's wishes and choices.

[0088] Through the above content, we can collect and process historical order information, as well as filter and push advertising resources based on core feature tags and core product tags. These steps together constitute an important part of the personalized advertising recommendation system, providing a solid foundation for subsequent advertising strategies. Based on this framework, we can further combine specific business scenarios and user needs to provide users with more accurate and personalized advertising services.

[0089] In some embodiments, the smart TV personalized content recommendation system may include multiple functional modules composed of computer program segments. The computer program of each program segment in the smart TV personalized content recommendation system may be stored in the memory of a computer device and executed by at least one processor to perform (see Figure 1 Description) Personalized content recommendation function for smart TVs.

[0090] In this embodiment, the smart TV personalized content recommendation system can be divided into multiple functional modules according to the functions it performs, such as Figure 2 As shown. The functional modules of the system 200 may include: a data acquisition module 210, a behavior extraction module 220, a behavior analysis module 230, a resource screening module 240 and a push control module 250. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0091] A data acquisition module is used to acquire basic user data and behavior record data, wherein the basic user data includes user personal information and device information; A behavior extraction module, used to extract browsing content and browsing time from the behavior record data, and set the browsing time as the interest level of the browsing content; A behavior analysis module is used to analyze the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interest level, count the core feature tags with the highest interest level; A resource screening module is used to select video resources associated with core feature tags as candidate resources and prioritize the candidate resources based on user basic data; The push control module is used to arrange the candidate resources from first to last according to the priority, and push the candidate resources in the arrangement order.

[0092] Optionally, as an embodiment of the present invention, the system further includes: Order collection module, used to collect historical order information; The order parsing module is used to parse the product tags of transaction products from historical orders and filter out the core product tags with the highest interest by counting the number of occurrences of product tags; Type matching module, used to filter matching advertising resources based on core feature tags and core product tags; The advertisement push module is used to push matching advertisement resources during the advertisement period.

[0093] Optionally, as an embodiment of the present invention, the data acquisition module includes: A personal information acquisition unit, used to acquire registration data of a user account and extract user personal information from the registration data, wherein the user personal information includes gender and age; The device information acquisition unit is used to acquire the device information of the user account login, and the device information includes the device model, device brand and device price.

[0094] Figure 3The intelligent TV personalized content recommendation method provided for the embodiment of the present application can be applied to the device. It will be appreciated by those skilled in the art that the device structure involved in the embodiment of the present invention does not constitute a limitation on the device, and the device may include more or less components than shown, or combine certain components, or arrange different components. In an embodiment of the present invention, the device includes but is not limited to a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0095] The device 300 may include: a processor 310, a memory 320 and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention, and it may be a bus structure or a star structure, and may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently.

[0096] The memory 320 may be used to store the execution instructions of the processor 310, and the memory 320 may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the device 300 is enabled to perform some or all of the steps in the following method embodiments.

[0097] The processor 310 is the control center of the storage device, and uses various interfaces and lines to connect various parts of the entire electronic device. It runs or executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic device and / or process data. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of a plurality of packaged ICs with the same or different functions. For example, the processor 310 can include only a central processing unit (CPU). In an embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.

[0098] The communication unit 330 is used to establish a communication channel so that the storage device can communicate with other devices, receive user data sent by other devices or send user data to other devices.

[0099] The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, and when the program is executed, the program may include some or all of the steps in each embodiment provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0100] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes, including several instructions for enabling a computer device (which can be a personal computer, a server, or a second device, a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention.

[0101] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.

[0102] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or modules, which can be electrical, mechanical or other forms.

[0103] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0104] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0105] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention.

Claims

1. A method for recommending personalized content for a smart TV, characterized in that: include: Obtain user basic data and behavior record data, wherein the user basic data includes user personal information and device information; Extracting browsing content and browsing time from the behavior record data, and setting the browsing time as the interest level of the browsing content; Analyze the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interest level, count the core feature tags with the highest interest level; The video resources associated with the core feature tags are used as candidate resources, and the candidate resources are prioritized based on the user's basic data; Arrange candidate resources from first to last according to priority, and push candidate resources in the order of arrangement.

2. The method according to claim 1, characterized in that The method further comprises: Collect historical order information; Parse the product tags of the transaction products from the historical orders, and filter out the core product tags with the highest interest by counting the number of times the product tags appear; Filter matching advertising resources based on core feature tags and core product tags; Push matching advertising resources during advertising periods.

3. The method according to claim 1, characterized in that Obtain user basic data and behavior record data, including: Obtaining registration data of a user account, and extracting user personal information from the registration data, the user personal information including gender and age; Obtain the device information of the user account logged in, the device information includes the device model, device brand and device price.

4. The method according to claim 1, characterized in that: Analyze the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interest level, count the core feature tags with the highest interest level, including: Parse the type, subject matter and participant information of the browsed content, convert the type, subject matter and participant information into feature tags in JSON format, and summarize the feature tags into feature tag combinations; The interest level of the browsed content is taken as the interest level of all feature tags in the corresponding feature tag combination; The interest levels of the same feature tag in different feature tag combinations are accumulated to obtain the interest levels of all feature tags; Normalize the interest of all feature labels to obtain the interest coefficient of the feature label; The feature labels whose interest coefficients reach the set interest threshold are selected as core feature labels.

5. The method according to claim 4, characterized in that The video resources associated with the core feature tags are used as candidate resources, and the candidate resources are prioritized based on the user basic data, including: Generate feature tags for video resources in the resource pool in advance; Calculate the matching degree between the core feature label and the feature label of the video resources in the resource pool; Screening video resources whose matching degree reaches a set matching degree threshold as candidate resources; The correspondence between the user basic data and the feature tags is preset, and the candidate resources whose feature tags have a correspondence with the user basic data are set to a high priority, and the candidate resources whose feature tags do not have a correspondence with the user basic data are set to a low priority.

6. A smart TV personalized content recommendation system, characterized in that: include: A data acquisition module is used to acquire basic user data and behavior record data, wherein the basic user data includes user personal information and device information; A behavior extraction module, used to extract browsing content and browsing time from the behavior record data, and set the browsing time as the interest level of the browsing content; A behavior analysis module is used to analyze the feature tag combination of the browsed content, and based on the feature tag combination of the browsed content and the corresponding interest level, count the core feature tags with the highest interest level; A resource screening module is used to select video resources associated with core feature tags as candidate resources and prioritize the candidate resources based on user basic data; The push control module is used to arrange the candidate resources from first to last according to the priority, and push the candidate resources in the arrangement order.

7. The system according to claim 6, characterized in that The system further comprises: Order collection module, used to collect historical order information; The order parsing module is used to parse the product tags of transaction products from historical orders and filter out the core product tags with the highest interest by counting the number of occurrences of product tags; Type matching module, used to filter matching advertising resources based on core feature tags and core product tags; The advertisement push module is used to push matching advertisement resources during the advertisement period.

8. The system according to claim 6, characterized in that The data acquisition module comprises: A personal information acquisition unit, used to acquire registration data of a user account and extract user personal information from the registration data, wherein the user personal information includes gender and age; The device information acquisition unit is used to acquire the device information of the user account login, and the device information includes the device model, device brand and device price.

9. A device, characterized in that: include: A memory, used for storing a personalized content recommendation program for a smart TV; A processor, configured to implement the steps of the smart TV personalized content recommendation method as described in any one of claims 1 to 5 when executing the smart TV personalized content recommendation program.

10. A computer-readable storage medium storing a computer program, characterized in that: The readable storage medium stores a smart TV personalized content recommendation program, and when the smart TV personalized content recommendation program is executed by the processor, the steps of the smart TV personalized content recommendation method according to any one of claims 1 to 5 are implemented.

Images