Intelligent storage management method and system for set top box data

Abstract

The invention discloses an intelligent storage management method and system for set top box data, which are used for improving the watching experience of a user. The method comprises the following steps: acquiring user data of the set top box, wherein the user data comprises historical watching records and real-time interaction data; performing value evaluation on watching content types in different time periods in the user data based on a preset scoring rule to obtain a content value scoring set; according to the content value scores corresponding to the watching content types in different time periods, carrying out storage space proportion distribution, and determining the watching content types lower than a preset score in the content value score set as to-be-cleaned contents; detecting whether the current load state of the set top box is a reasonable use state or not; and if not, releasing the storage space of the to-be-cleaned content.

Description

Technical Field

[0001] This application relates to the technical field of data storage, and particularly to an intelligent storage management method and system for set-top box data. Background Art

[0002] With the rapid development of intelligent devices and streaming media technology, the set-top box, as the core terminal of home entertainment, the importance of its data storage management has become increasingly prominent. Efficient storage management not only concerns the smoothness of the user experience, but also directly affects the operating efficiency and resource utilization rate of the system, and is a key link in promoting the development of the smart home ecosystem. In the field of set-top boxes, how to reasonably utilize the limited storage space to meet the diverse and personalized viewing needs of users has become a key direction that cannot be ignored in technical research and industrial applications.

[0003] Currently, during the operation of the set-top box, the set-top box usually adopts a static management method of fixed partition storage and cannot make corresponding adjustments according to user needs. When the storage space is insufficient, the space can only be cleared by deleting the whole, which may cause the content that the user is interested in to be accidentally deleted, affecting the user's viewing experience. Summary of the Invention

[0004] This application provides an intelligent storage management method and system for set-top box data to improve the user's viewing experience.

[0005] In the first aspect of this application, an intelligent storage management method for set-top box data is provided, including: Obtain the user data of the set-top box, where the user data includes historical viewing records and real-time interaction data; Based on a preset scoring rule, evaluate the value of the viewing content types in different time periods in the user data to obtain a content value scoring set; Allocate the storage space ratio according to the content value scores corresponding to the viewing content types in different time periods, and determine the viewing content types in the content value scoring set that are lower than the preset score as the content to be cleared; Detect whether the current load state of the set-top box is a reasonable use state; If not, release the storage space of the content to be cleared.

[0006] Optionally, the evaluating the value of the viewing content types in different time periods in the user data based on a preset scoring rule to obtain a content value scoring set includes: Group all the viewing contents in different time periods in the historical viewing records according to the viewing content types; Calculate the viewing frequency ratio of each viewing content type in the corresponding time period; Calculate the proportion of the interaction frequency of the real-time interaction data corresponding to each watched content type in the corresponding time period; Calculate the content value score of each watched content type according to the watched frequency proportion and the interaction frequency proportion, and obtain a content value score set.

[0007] Optionally, the value evaluation of the watched content types in different time periods in the user data based on a preset scoring rule to obtain a content value score set further includes: Calculate the proportion of the actual watched duration of each watched content type in the corresponding time period; Correct the content value score set according to the proportion of the actual watched duration.

[0008] Optionally, the storage space proportion allocation according to the content value scores corresponding to the watched content types in different time periods includes: Sort the watched content types in different time periods in the content value score set in descending order according to the content value scores; Divide the storage space of the set-top box into a high-heat storage area, a medium-heat storage area, and a low-heat storage area, and the storage space proportions of the high-heat storage area, the medium-heat storage area, and the low-heat storage area are sorted in descending order from high to low; Cache the watched content types with sorting rankings higher than the first preset ranking into the high-heat storage area, cache the watched content types with sorting rankings higher than the second preset ranking and lower than the first preset ranking into the medium-heat storage area, and cache the watched content types with sorting rankings lower than the second preset ranking into the low-heat storage area.

[0009] Optionally, determining the watched content types in the content value score set that are lower than the preset score as the content to be cleared includes: Calculate the average value of the content value scores of all watched content types in the low-heat storage area, and determine the average value as the preset score; Determine the watched content types in the low-heat storage area that are lower than the preset score as the content to be cleared. Optionally, after releasing the storage space of the content to be cleared, the method further includes: Extract the time feature and the watched content type feature from the historical watch record; Use the time feature and the watched content type feature as training samples to train a preset model of the watched content type, and the preset model of the watched content type is used to output the watched content type in a preset time period; Reallocate the storage space proportion of the set-top box according to the watched content type in the preset time period.

[0010] Optionally, after releasing the storage space of the content to be cleaned up, the method further includes: Receiving the satisfaction feedback results of the user on the types of viewing content in different time periods; Reallocating the storage space ratio of the set-top box according to the satisfaction feedback results.

[0011] The second aspect of the present application provides an intelligent storage management system for set-top box data, including: An acquisition unit, configured to acquire user data of the set-top box, where the user data includes historical viewing records and real-time interaction data; An evaluation unit, configured to perform value evaluation on the types of viewing content in different time periods in the user data based on a preset scoring rule to obtain a content value scoring set; An allocation unit, configured to allocate the storage space ratio according to the content value scores corresponding to the types of viewing content in different time periods, and determine that the types of viewing content in the content value scoring set that are lower than the preset score are the content to be cleaned up; A detection unit, configured to detect whether the current load state of the set-top box is a reasonable usage state; A release unit, configured to release the storage space of the content to be cleaned up when the current load state of the set-top box is a reasonable usage state.

[0012] Optionally, the evaluation unit is specifically configured to: Group all the viewing content in different time periods in the historical viewing records according to the types of viewing content; Calculate the viewing frequency ratio of each type of viewing content in the corresponding time period; Calculate the interaction frequency ratio of the real-time interaction data corresponding to each type of viewing content in the corresponding time period; Calculate the content value scores of each type of viewing content according to the viewing frequency ratio and the interaction frequency ratio to obtain a content value scoring set.

[0013] Optionally, the evaluation unit is further specifically configured to: Calculate the actual viewing duration ratio of each type of viewing content in the corresponding time period; Correct the content value scoring set according to the actual viewing duration ratio.

[0014] It can be seen from the above technical solutions that the present application has the following effects: First, obtain the user data of the set-top box, which includes historical viewing records and real-time interaction data; then, based on a preset scoring rule, evaluate the value of the viewing content types in different time periods in the user data to obtain a content value scoring set; next, allocate the storage space ratio according to the content value scores corresponding to the viewing content types in different time periods, and determine the viewing content types in the content value scoring set that are lower than the preset score as the content to be cleared; finally, detect whether the current load status of the set-top box is in a reasonable usage state; if not, release the storage space of the content to be cleared. In this way, by evaluating the content value scores, it can reflect in real time the preference degree of users for different viewing content types in different time periods. Then, through this content value score, the storage space ratio is allocated and the content to be cleared is determined, so as to realize the adjustment of the storage space according to the user's needs, reduce the situation of the content that the user is interested in being deleted by mistake, and thus improve the user's viewing experience. Description of the Drawings

[0015] Figure 1 Schematic diagram of an embodiment of an intelligent storage management method for set-top box data in this application; Figure 2-1 and Figure 2-2 Schematic diagram of another embodiment of an intelligent storage management method for set-top box data in this application; Figure 3 Schematic diagram of an embodiment of an intelligent storage management system for set-top box data in this application; Figure 4 Schematic diagram of another embodiment of an intelligent storage management system for set-top box data in this application. Detailed Embodiment

[0016] This application provides an intelligent storage management method and system for set-top box data to improve the user's viewing experience.

[0017] The intelligent storage management method for set-top box data described in this application is implemented by being executed on a system or server. Please refer to Figure 1 As shown, an embodiment of the intelligent storage management method for set-top box data in this application includes: 101. Obtain the user data of the set-top box, which includes historical viewing records and real-time interaction data; In this embodiment, all user data stored in the set-top box memory is obtained. This part of user data includes the user's historical viewing records and real-time interaction data. Among them, due to the limited storage space of the set-top box memory, this part of historical viewing records is usually the content viewed by the user within a certain period of time. For example, the obtained historical viewing records are the viewing records of the user within 30 days or the viewing records of the user within 60 days. In addition, the historical viewing records may include videos, pictures or texts, and the specific file format is not limited here. It should be noted that the historical viewing records also include the corresponding time information. For example, the user watched video A from 19:30 to 19:45.

[0018] It should be noted that the real-time interaction data is the human-computer interaction carried out by the user when viewing relevant content. For example, when the user watched video A, the user followed the publisher of video A and liked video A.

[0019] 102. Based on a preset scoring rule, evaluate the value of the viewing content types in different time periods of the user data to obtain a content value scoring set; In this embodiment, the time information corresponding to each viewing content is extracted from the historical viewing records, and the proportion of the viewing content corresponding to each time period is calculated according to the time period sequence to determine the time period with higher user activity. Among them, the time periods with relatively low viewing content proportions can be excluded to determine the remaining different time periods for subsequent use. For example, set the preset viewing content proportion for a single time period to 10%. The proportion of the viewing content corresponding to the time period from 7:00 to 10:00 in the morning is 30%, the proportion of the viewing content corresponding to the time period from 17:00 to 19:00 in the afternoon is 20%, and the proportion of the viewing content corresponding to the time period from 22:00 to 23:00 in the evening is 40%. The proportions of the viewing content in the remaining time periods are all less than 5%. Therefore, it can be determined that the time periods from 7:00 to 10:00 in the morning, from 17:00 to 19:00 in the afternoon, and from 22:00 to 23:00 in the evening are the target time periods available for subsequent use.

[0020] It should be noted that the viewing content types in different time periods refer to the content types corresponding to all the viewed content of the user within a single time period. For example, the user watched video A, video B, and video C during the time period from 17:00 to 19:00 in the afternoon. Among them, video A is sports type content, video B is news type content, and video C is program type content. By evaluating the value of the viewing content types in different time periods, the preference degrees for different viewing content types in different time periods can be obtained.

[0021] 103. Allocate the storage space proportion according to the content value scores corresponding to the viewing content types in different time periods, and determine the viewing content types with content value scores lower than the preset score in the content value scoring set as the content to be cleared; In this embodiment, the storage space of the set-top box is divided into regions according to the content value scores of the watched content types in different time periods, and available memory settings are made for each region to improve the effective utilization rate of the internal storage space of the set-top box. For example, in the time period from 17:00 to 19:00 in the afternoon, the sports type content has the highest proportion, so more storage space can be allocated for the sports type content; while the education type content has the lowest proportion, so less storage space can be allocated for the education type content. For the watched content types with content value scores lower than the preset score, they can be listed as content to be cleared, so that when the memory is insufficient, the low-value watched content can be accurately cleared.

[0022] 104. Detect whether the current load status of the set-top box is a reasonable usage status. If so, execute step 105; In this embodiment, first, real-time metrics such as the utilization rate of the central processing unit (CPU), the memory occupancy rate, and the number of read and write operations per second of the disk (IOPS, Input / Output Operations Per Second) are collected through performance probes deployed on the set-top box server node, and then the sliding window algorithm is used to calculate the load average of the obtained real-time metrics. When the load average is less than the threshold, it indicates that the current load status of the set-top box is a reasonable usage status; when the load average is greater than or equal to the threshold, it indicates that the current load status of the set-top box is an unreasonable usage status. In addition, the network traffic can also be analyzed through the NetFlow protocol, and the EWMA is used to process the inbound traffic data per second. When the traffic surges exceed the baseline value threshold and last for a certain period of time, it is determined as an abnormal traffic impact, and the current load status of the set-top box is further confirmed through the determination result of the abnormal traffic impact to improve the detection accuracy of the load status.

[0023] 105. Release the storage space of the content to be cleared.

[0024] When it is determined that the current load status of the set-top box is a reasonable usage status or an unreasonable usage status, the content to be cleared is deleted to release the storage space occupied by the content to be cleared.

[0025] In this embodiment, first, the user data of the set-top box is obtained, which includes historical viewing records and real-time interaction data; then, based on a preset scoring rule, the value of the viewing content types in different time periods in the user data is evaluated to obtain a content value scoring set; then, the storage space ratio is allocated according to the content value scores corresponding to the viewing content types in different time periods, and the viewing content types with content value scores lower than the preset score in the content value scoring set are determined as the content to be cleared; finally, it is detected whether the current load state of the set-top box is a reasonable usage state; if not, the storage space of the content to be cleared is released. In this way, by evaluating the content value scores, the preference degree of the user for different viewing content types in different time periods can be reflected in real time, and then the storage space ratio is allocated and the content to be cleared is determined through the content value scores, so that the storage space can be adjusted according to the user's needs, reducing the situation that the content the user is interested in is accidentally deleted, thereby improving the user's viewing experience.

[0026] Please refer to Figure 2-1 and Figure 2-2 As shown, another embodiment of the intelligent storage management method for set-top box data in this application includes: 201. Obtain the user data of the set-top box, which includes historical viewing records and real-time interaction data; Step 201 in this embodiment is the same as step 101 in the embodiment Figure 1 shown above, and will not be elaborated here.

[0027] 202. Group all the viewing content in different time periods in the historical viewing records according to the viewing content type; 203. Calculate the viewing frequency ratio of each viewing content type in the corresponding time period; 204. Calculate the interaction frequency ratio of the real-time interaction data corresponding to each viewing content type in the corresponding time period; 205. Calculate the content value score of each viewing content type according to the viewing frequency ratio and the interaction frequency ratio to obtain a content value scoring set; Optionally, in this embodiment, first, all the viewed contents within different time periods are grouped by content type. For example, a user watched Video A, Video B, Video C, and Video D during the time period from 17:00 to 19:00 in the afternoon. Among them, Video A, Video B, and Video C are grouped into the sports type group, and Video D is grouped into the news type group. Then, the viewing frequency ratio is calculated for the content types of the viewed contents within each different time period. The viewing frequency ratio represents the ratio of the number of views of each content type within the same time period to the total number of views of all content types. For example, a user watched Video A, Video B, Video C, and Video D during the time period from 17:00 to 19:00 in the afternoon. The total number of views of all content types is 4 times. The sports type group contains 3 videos (Video A, Video B, and Video C), and the news type group contains 1 video (Video D). Then, the viewing frequency ratio of the sports type group is three-fourths, and the viewing frequency ratio of the news type group is one-fourth. Next, the interaction frequency ratio is calculated for the real-time interaction data corresponding to the content types of the viewed contents within each different time period. The interaction frequency ratio represents the ratio of the number of interactions of each content type within the same time period to the total number of interactions. For example, a user had 5 interactions during the time period from 17:00 to 19:00 in the afternoon. Among them, the user gave 1 like and 1 comment to Video A in the sports type group, followed 1 video in the sports type group (Video B), and followed and commented on Video D in the news type group once. Then, the interaction frequency ratio of the sports type group is three-fifths, and the interaction frequency ratio of the news type group is two-fifths. Finally, different weights are assigned to the viewing frequency ratio and the interaction frequency ratio respectively, and the weighted viewing frequency ratio and the interaction frequency ratio are summed up to obtain the content value score , and the corresponding formula is: , where represents the content type of the viewed content, represents the viewing frequency ratio, represents the weight of the viewing frequency ratio, represents the interaction frequency ratio, represents the weight of the interaction frequency ratio.

[0028] In another implementable manner, the actual viewing duration ratio of each viewing content type in the corresponding time period can be calculated; the content value scoring set is corrected according to the actual viewing duration ratio. The calculated content value score is corrected by adding a correction factor to improve the credibility of the content value score. Specifically, the actual viewing duration ratio represents the ratio of the actual viewing duration of a single viewing content type in a single time period to the total viewing duration. For example: the total duration of all videos in the sports type group is 3 minutes, and the total duration of all videos in the news type group is 5 minutes. Among them, the actual duration of the user watching all videos in the sports type group is 1 minute, and the actual duration of the user watching all videos in the news type group is 2 minutes. Then the actual viewing duration ratio of the sports type group is one-third, and the actual viewing duration ratio of the news type group is two-fifths. Multiplying the calculated content value score by the corresponding actual viewing duration ratio can obtain the final content value score.

[0029] 206. Sort the viewing content types in different time periods in the content value scoring set in descending order according to the content value score; 207. Divide the storage space of the set-top box into a high-heat storage area, a medium-heat storage area, and a low-heat storage area. The storage space ratios of the high-heat storage area, the medium-heat storage area, and the low-heat storage area are sorted in descending order from high to low; 208. Cache the viewing content types with sorting rankings higher than the first preset ranking into the high-heat storage area, cache the viewing content types with sorting rankings higher than the second preset ranking and lower than the first preset ranking into the medium-heat storage area, and cache the viewing content types with sorting rankings lower than the second preset ranking into the low-heat storage area; Optionally, in this embodiment, first sort the different viewing content types in the same time period in descending order according to the content value score. For example: the content value scores of the sports type group, the news type group, and the program type group are 0.69, 0.72, and 0.55 respectively. Then the sorting order is news type group > sports type group > program type group. Then, divide the storage space of the set-top box into a high-heat storage area, a medium-heat storage area, and a low-heat storage area according to the popularity of different viewing content types, and set the corresponding storage space ratios for each storage area according to a preset ratio. For example: the storage space ratio of the high-heat storage area is 60%, the storage space ratio of the medium-heat storage area is 30%, and the storage space ratio of the low-heat storage area is 10%. Finally, cache the different viewing content types into the corresponding storage areas according to the preset rankings. For example: the top 3 viewing content types are cached into the high-heat storage area, the viewing content types ranked 3-5 are cached into the medium-heat storage area, and the viewing content types ranked after the 5th are cached into the low-heat storage area.

[0030] 209. Calculate the average value of the content value scores of all watched content types in the low-heat storage area, and determine the average value as the preset score; 210. Determine that the watched content types in the low-heat storage area that are lower than the preset score are the content to be cleared; Optionally, in this embodiment, after the storage space of the set-top box is allocated, obtain the content value scores of all watched content types in the low-heat storage area, then calculate the average value of all content value scores, and determine the average value as the preset score. When the content value score of any watched content type in the low-heat storage area is lower than the preset score, it means that this watched content type is low-value content to be cleared. When the memory of the set-top box is insufficient, this part of the content to be cleared is preferentially cleared to reasonably release the storage space.

[0031] 211. Detect whether the current load status of the set-top box is a reasonable usage status. If so, execute step 212; 212. Release the storage space of the content to be cleared; Steps 211 and 212 in this embodiment are similar to steps 104 and 105 in the Figure 1 embodiment shown above, and will not be elaborated here.

[0032] 213. Extract the time feature and the watched content type feature from the historical viewing record; 214. Use the time feature and the watched content type feature as training samples to train the watched content type preset model, and the watched content type preset model is used to output the watched content type in the preset time period; 215. Reallocate the storage space ratio of the set-top box according to the watched content type in the preset time period; Optionally, in this embodiment, first, time features of different time periods in the historical viewing records and viewing content type features corresponding to the corresponding time periods are extracted. Moreover, the time periods in the time features are continuous time periods. For example: the time period from 7:00 to 8:00, the time period from 9:00 to 10:00, and the time period from 10:00 to 11:00. Then, using the time features of the previous time period and the viewing content type features corresponding to the previous time period as input features, and the time features of the next time period and the viewing content type features corresponding to the next time period as predicted values to train the initial model, and determining the initial model after training is completed as the viewing content type preset model. The viewing content type preset model can output the viewing content type of the next time period. For example: when the current time is 18:00 in the afternoon, the viewing content type preset model predicts that the user may watch sports and news content at 19:00 in the afternoon. Finally, adjust the storage space ratio of the corresponding viewing content type features according to the output value of the viewing content type preset model. For example: when the viewing content type preset model predicts that the user can watch sports content in the next time period, the storage space ratio of sports content can be increased to achieve the advanced dynamic adjustment of the storage space ratio of the set-top box and improve the flexibility of the storage space.

[0033] 216. Receive the satisfaction feedback results of the user on the viewing content types of different time periods; 217. Reallocate the storage space ratio of the set-top box according to the satisfaction feedback results.

[0034] Optionally, in this embodiment, a satisfaction survey link of the current viewing content type can be sent to the user, and the user can feedback relevant opinions in the satisfaction survey link. For example: when the user finishes watching video A, a survey link of "Are you satisfied with the current type of video?" pops up, and the user can choose "Yes" or "No". After obtaining the satisfaction feedback results fed back by the user, determine the viewing content types that need to adjust the storage space according to the satisfaction feedback results, and then reallocate the storage space ratio of the corresponding viewing content types to achieve the dynamic adjustment of the storage space ratio allocation of the set-top box according to the real-time needs of the user and further improve the viewing experience of the user.

[0035] Please refer to Figure 3 As shown, an embodiment of the intelligent storage management system for set-top box data in this application includes: An acquisition unit 301, configured to acquire user data of the set-top box, where the user data includes historical viewing records and real-time interaction data; An evaluation unit 302, configured to perform value evaluation on the viewing content types of different time periods in the user data based on a preset scoring rule to obtain a content value scoring set; An allocation unit 303 is configured to allocate the storage space ratio according to the content value scores corresponding to the viewing content types in different time periods, and determine the viewing content types with content value scores lower than the preset score in the content value score set as the content to be cleared; A detection unit 304 is configured to detect whether the current load state of the set-top box is a reasonable usage state; A release unit 305 is configured to release the storage space of the content to be cleared when the current load state of the set-top box is a reasonable usage state.

[0036] In this embodiment, the acquisition unit 301 acquires the user data of the set-top box, and the user data includes historical viewing records and real-time interaction data; the evaluation unit 302 performs value evaluation on the viewing content types in different time periods in the user data based on a preset scoring rule to obtain a content value score set; the allocation unit allocates the storage space ratio according to the content value scores corresponding to the viewing content types in different time periods, and determines the viewing content types with content value scores lower than the preset score in the content value score set as the content to be cleared; the detection unit 304 detects whether the current load state of the set-top box is a reasonable usage state; when the current load state of the set-top box is a reasonable usage state, the release unit 305 releases the storage space of the content to be cleared. In this way, by evaluating the content value scores, the preference degree of the user for different viewing content types in different time periods can be reflected in real time, and then the storage space ratio is allocated and the content to be cleared is determined through the content value scores, so that the storage space can be adjusted according to the user's needs, reducing the situation that the content the user is interested in is deleted by mistake, thereby improving the user's viewing experience.

[0037] Please refer to Figure 4 As shown in the figure, another embodiment of the intelligent storage management system for set-top box data in the present application includes: An acquisition unit 401 is configured to acquire the user data of the set-top box, and the user data includes historical viewing records and real-time interaction data; An evaluation unit 402 is specifically configured to group all the viewing contents in different time periods in the historical viewing records according to the viewing content types; calculate the viewing frequency ratios of each viewing content type in the corresponding time periods; calculate the interaction frequency ratios of the real-time interaction data corresponding to each viewing content type in the corresponding time periods; calculate the content value scores of each viewing content type according to the viewing frequency ratios and the interaction frequency ratios to obtain a content value score set; and is also specifically configured to calculate the actual viewing duration ratios of each viewing content type in the corresponding time periods; and correct the content value score set according to the actual viewing duration ratios.

[0038] The allocation unit 403 is specifically configured to sort the viewing content types in different time periods in the content value score set in descending order according to the content value score; divide the storage space of the set-top box into a high popularity storage area, a medium popularity storage area, and a low popularity storage area, and sort the storage space ratios of the high popularity storage area, the medium popularity storage area, and the low popularity storage area in descending order; cache the viewing content types with sorting rankings higher than the first preset ranking into the high popularity storage area, cache the viewing content types with sorting rankings higher than the second preset ranking and lower than the first preset ranking into the medium popularity storage area, cache the viewing content types with sorting rankings lower than the second preset ranking into the low popularity storage area, calculate the average value of the content value scores of all viewing content types in the low popularity storage area, and determine the average value as the preset score; determine the viewing content types in the low popularity storage area that are lower than the preset score as the content to be cleared; The detection unit 404 is configured to detect whether the current load state of the set-top box is a reasonable usage state; The release unit 405 is configured to release the storage space of the content to be cleared when the current load state of the set-top box is a reasonable usage state.

[0039] The extraction unit 406 is configured to extract the time feature and the viewing content type feature from the historical viewing record; The training unit 407 is configured to use the time feature and the viewing content type feature as training samples to train a preset model of the viewing content type, and the preset model of the viewing content type is used to output the viewing content type in a preset time period; The first reallocation unit 408 is configured to reallocate the storage space ratio of the set-top box according to the viewing content type in the preset time period.

[0040] The receiving unit 409 is configured to receive the satisfaction feedback result of the user on the viewing content type in different time periods; The second reallocation unit 410 is configured to reallocate the storage space ratio of the set-top box according to the satisfaction feedback result.

[0041] In this embodiment, the functions of each unit are similar to the functions of steps 201 to 217 in the foregoing Figure 2-1 and Figure 2-2 shown in the embodiment, and will not be elaborated here.

[0042] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, device, and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated here.

[0043] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of devices or units can be in electrical, mechanical, or other forms.

[0044] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0045] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0046] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

Claims

1. An intelligent storage management method for set-top box data, characterized in that: include: Acquire user data of a set-top box, wherein the user data includes historical viewing records and real-time interaction data; Based on a preset scoring rule, the types of content viewed in different time periods in the user data are evaluated for value, to obtain a content value scoring set; Allocate storage space proportions according to content value scores corresponding to viewing content types in different time periods, and determine viewing content types with content value scores lower than preset scores as content to be cleared; Detecting whether the current load state of the set-top box is a reasonable usage state; If not, the storage space of the content to be cleared is released.

2. The intelligent storage management method for set-top box data according to claim 1, characterized in that: The method of evaluating the value of the content types viewed in different time periods in the user data based on a preset scoring rule to obtain a content value scoring set includes: Grouping all viewing contents in different time periods in the historical viewing records according to viewing content types; Calculate the viewing frequency ratio of each viewing content type in the corresponding time period; Calculate the interaction frequency ratio of the real-time interaction data corresponding to each viewing content type in the corresponding time period; The content value score of each viewing content type is calculated according to the viewing frequency ratio and the interaction frequency ratio to obtain a content value score set.

3. The intelligent storage management method for set-top box data according to claim 2, characterized in that: The method of evaluating the value of the types of content viewed in different time periods in the user data based on a preset scoring rule to obtain a content value scoring set also includes: Calculate the actual viewing time ratio of each viewing content type in the corresponding time period; The content value rating set is modified according to the actual viewing time ratio.

4. The intelligent storage management method for set-top box data according to claim 1, characterized in that: The storage space ratio allocation according to the content value scores corresponding to the viewing content types in different time periods includes: Sorting the viewing content types in different time periods in the content value score set in descending order according to the content value scores; Dividing the storage space of the set-top box into a high-heat storage area, a medium-heat storage area, and a low-heat storage area, and sorting the storage space proportions of the high-heat storage area, the medium-heat storage area, and the low-heat storage area in descending order from high to low; The viewing content types with a ranking higher than the first preset ranking are cached in the high-heat storage area, the viewing content types with a ranking higher than the second preset ranking and lower than the first preset ranking are cached in the medium-heat storage area, and the viewing content types with a ranking lower than the second preset ranking are cached in the low-heat storage area.

5. The intelligent storage management method for set-top box data according to claim 4, characterized in that: The step of determining the viewing content types in the content value score set that are lower than a preset score as the content to be cleared includes: Calculating an average value of content value scores of all viewing content types in the low-heat storage area, and determining the average value as a preset score; Determine the viewing content types in the low-heat storage area that are lower than a preset score as content to be cleared.

6. The intelligent storage management method for set-top box data according to claim 1, characterized in that: After releasing the storage space of the content to be cleared, the method further includes: Extracting time features and viewing content type features from the historical viewing records; Using the time feature and the viewing content type feature as training samples to train a viewing content type preset model, the viewing content type preset model is used to output the viewing content type in a preset time period; The storage space ratio of the set-top box is reallocated according to the type of content viewed in the preset time period.

7. The intelligent storage management method for set-top box data according to claim 1, characterized in that: After releasing the storage space of the content to be cleared, the method further includes: Receive user satisfaction feedback on the types of content viewed in different time periods; The storage space ratio of the set-top box is reallocated according to the satisfaction feedback result.

8. An intelligent storage management system for set-top box data, characterized in that: include: An acquisition unit, used to acquire user data of a set-top box, wherein the user data includes historical viewing records and real-time interaction data; An evaluation unit, configured to evaluate the value of the viewing content types in different time periods in the user data based on a preset scoring rule to obtain a content value scoring set; an allocation unit, configured to allocate storage space proportions according to content value scores corresponding to viewing content types in different time periods, and determine viewing content types with content value scores lower than a preset score as content to be cleared; A detection unit, used to detect whether the current load state of the set-top box is a reasonable use state; The release unit is used to release the storage space of the content to be cleared when the current load state of the set-top box is a reasonable use state.

9. The intelligent storage management system for set-top box data according to claim 8, characterized in that: The evaluation unit is specifically used for: Grouping all viewing contents in different time periods in the historical viewing records according to viewing content types; Calculate the viewing frequency ratio of each viewing content type in the corresponding time period; Calculate the interaction frequency ratio of the real-time interaction data corresponding to each viewing content type in the corresponding time period; The content value score of each viewing content type is calculated according to the viewing frequency ratio and the interaction frequency ratio to obtain a content value score set.

10. The intelligent storage management system for set-top box data according to claim 9, characterized in that: The evaluation unit is also specifically used for: Calculate the actual viewing time ratio of each viewing content type in the corresponding time period; The content value rating set is modified according to the actual viewing time ratio.

Images