Method for intelligently playing recommended programs by tablet computer

Abstract

The invention discloses a method for intelligently playing recommended programs by a tablet computer, and belongs to the technical field of tablet computers. According to the method, head portrait information of a user watching a program is obtained, a program watching habit curve of the user is made according to the head portrait information of the user and information of key watching programs, and an overall preference sequence of the user to the program is made according to the head portrait information of the user and information of all watched programs; and finally, when the user watches the program every time, mapping a user program watching habit curve corresponding to the user and an overall preference sequence of the user to the program through the user head portrait information, and recommending the program to the user through the user program watching habit curve and the overall preference sequence of the user to the program. According to the method, the program watching habit curve of the user and the overall preference sequence of the user for the program are combined to recommend the program for the user, so that the recommended program is more accurate and better conforms to habits and hobbies of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of tablet computers, and in particular to a method for intelligently playing recommended programs on a tablet computer. Background Art

[0002] With the popularization of smart devices, tablet computers, as powerful and convenient terminals, have gradually become one of the main carriers of daily entertainment and information consumption; especially in the field of content consumption, intelligent playback recommendation systems have been widely used. They analyze users' interests and behavior data to recommend personalized programs, videos or music, greatly improving the user experience; however, existing intelligent recommendation systems still have some significant defects.

[0003] First, most existing recommendation systems rely on algorithms based on historical data, such as collaborative filtering and content-based recommendations. Although these methods can provide personalized content recommendations to a certain extent, they also have a "cold start" problem. When a user uses a device for the first time, due to the lack of sufficient historical data, the system often finds it difficult to make accurate recommendations. In addition, most existing recommendation algorithms focus on users' short-term interests and lack recognition of long-term preferences or potential interests, which may cause users to feel that the recommended content is monotonous and boring after long-term use.

[0004] Secondly, the intelligence of most current recommendation systems is limited and they are unable to fully understand the user's true intentions. For example, users' needs change in different situations. For example, they may prefer to obtain news and business information when working, but prefer to watch entertainment programs or movies when relaxing. This diverse demand has not been fully identified and responded to. In addition, the accuracy of personalized recommendations, the real-time nature of content updates, and the integration of cross-platform content in existing intelligent recommendation systems still need to be strengthened. In particular, they are unable to handle the interaction of user behavioral data across multiple platforms and devices well, resulting in large differences in recommended content between different devices.

[0005] Therefore, although intelligent recommendation technology has brought convenience and entertainment to users, it still faces a series of technical bottlenecks in achieving more accurate and diversified personalized recommendations, such as insufficient data, insufficient understanding of demand changes, and limited intelligence. Solving these problems has become an important development direction for future smart tablet content recommendation systems. Summary of the Invention

[0006] The purpose of the present invention is to overcome the above problems existing in the prior art and to greatly improve the technical effect thereof on the basis of the existing technology. To this end, the present invention provides a method for intelligently playing recommended programs on a tablet computer, the method comprising:

[0007] When playing a program through a tablet player, the tablet camera function is synchronously called to obtain the avatar information of the user watching the program; wherein, synchronously calling the tablet camera function refers to: choosing to manually turn on the tablet camera function after opening the tablet player; and choosing to automatically start the tablet camera function when the tablet is turned on through application development.

[0008] A user program viewing habit curve is prepared based on the user's avatar information and the information of the programs that are watched most, and a user's overall preference sequence for programs is prepared based on the user's avatar information and the information of all the programs that are watched; the information of the programs that are watched most refers to programs that are watched continuously for more than 5 minutes; wherein, the user program viewing habit curve prepared based on the user's avatar information and the information of the programs that are watched most includes: using the user's avatar information as marking information, the horizontal axis is the viewing time information of the programs, the vertical axis is the classification information of the corresponding viewing programs, and a rectangular wave curve is prepared according to the time sequence of watching programs of different classifications, and the rectangular wave curve is called the user program viewing habit curve; the user's overall preference sequence for programs prepared based on the user's avatar information and the information of all the programs that are watched includes: an offline overall preference sequence and an online overall preference sequence; the offline overall preference sequence refers to the user's overall preference sequence when the tablet computer is not playing programs; the offline overall preference sequence uses the user's avatar information as marking information, and is based on the user's total cumulative preferences for different types of programs. Viewing time: programs are sorted according to viewing time to form a user's offline overall preference sequence, with programs with longer cumulative viewing time being ranked at the top of the sequence; the online overall preference sequence uses user avatar information as marking information, and based on the real-time offline overall preference sequence, the offline overall preference sequence is updated according to the time of continuous viewing of programs of the same type to form an online overall preference sequence; the updating of the offline overall preference sequence includes: first, sorting the programs in the offline overall preference sequence from 1 to n, that is, marking the first program in the offline overall preference sequence as 1 and the last program as n; then, updating the time of the real-time continuously viewed programs; updating the time of the real-time continuously viewed programs by setting weights and the total viewing time of the first program; finally, re-sorting the offline overall preference sequence based on the updated time of the continuously viewed programs to form the user's online overall preference sequence; the formula for updating the real-time continuously viewed program time by setting weights and the total viewing time of the first program is:

[0009] H i =αh1+h i

[0010] Among them, h1 is the total viewing time of the program ranked first in the offline overall preference sequence, h iis the total viewing time of the program ranked in the ith sequence in the offline overall preference sequence, α is the weight coefficient, α∈[0,1], which stipulates that when the continuous viewing time is less than 5 hours, When the continuous viewing time is not less than 5 hours, α=1; H i is the total viewing time corresponding to the program ranked in the i-th sequence in the updated offline overall preference sequence.

[0011] Furthermore, the offline overall preference sequence includes: as the time spent watching programs increases, the total time of each type of program in the offline overall preference sequence is continuously updated, and the updating method is to add the newly added viewing time of the corresponding program to the original classification of the program, that is, to obtain the latest offline overall preference sequence; the online overall preference sequence includes: the formation of the online overall preference sequence is updated based on the latest offline overall preference sequence; when the type of program continuously watched is changed, the updated online overall preference sequence is deleted; and the user's new online overall preference sequence is formed by re-updating the latest offline overall preference sequence.

[0012] Each time a user watches a program, the user's avatar information is obtained through the camera of the tablet computer, and the user's program viewing habit curve corresponding to the user and the user's overall program preference sequence are mapped through the user avatar information, and programs are recommended to the user based on the user's program viewing habit curve and the user's overall program preference sequence; wherein, the recommendation of programs to the user based on the user's program viewing habit curve and the user's overall program preference sequence includes: comparing the programs watched by the user in real time with the user's program viewing habit curve to generate programs recommended to the user based on the user's program viewing habit curve; recommending programs to the user based on the newly formed online overall preference sequence; the comparison of the programs watched by the user in real time with the user's program viewing habit curve includes: forming a corresponding user viewing curve based on the programs watched by the user in real time this time, comparing the formed corresponding user viewing curve with the user's program viewing habit curves of previous times, and calculating the corresponding user viewing habit curves formed according to the order and time of watching programs. The curve of program viewing accounts for a percentage of the user's previous program viewing habit curves, and the user's program viewing habit curve corresponding to the highest percentage is selected, and programs are recommended to the user through the user's program viewing habit curve corresponding to the highest percentage; the said recommending programs to the user through the formed new online overall preference sequence includes: taking the top three program types in the formed new online overall preference sequence as user programs to be recommended; comparing the programs recommended to the user through the user's program viewing habit curve with the user's programs to be recommended generated by the new online overall preference sequence, if the program recommended to the user through the user's program viewing habit curve is one of the user's programs to be recommended, then the program recommended to the user through the user's program viewing habit curve is the recommended program; if the program recommended to the user through the user's program viewing habit curve is not one of the user's programs to be recommended, then it is considered that the user's program viewing habit this time is different from the previous program viewing habit, and the program ranked first among the programs to be recommended is the recommended program.

[0013] The beneficial effects of the present invention are:

[0014] The present invention proposes a method for intelligently playing and recommending programs on a tablet computer. The advantages of the present invention are: a1, a method for generating a user's program viewing habit curve, generating an offline overall preference sequence and an online overall preference sequence, and a method for updating the offline overall preference sequence and the online overall preference sequence are provided; a2, by combining the user's program viewing habit curve with the user's overall preference sequence for programs to recommend programs to the user, the recommended programs will be more accurate and more in line with the user's habits and preferences. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The present invention provides a flow chart of a method for intelligently playing recommended programs on a tablet computer. DETAILED DESCRIPTION

[0016] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments given here are only used to illustrate and explain the present invention and cannot be used to limit the present invention.

[0017] like Figure 1 As shown, it is a flowchart of a method for intelligently playing and recommending programs on a tablet computer according to an embodiment of the present invention, and the flowchart includes: step S1, when playing a program through a tablet computer player, synchronously calling the tablet computer camera function to obtain the user avatar information of the user watching the program; step S2, creating a user program viewing habit curve based on the user avatar information and information about the programs that are watched most, and creating a user's overall preference sequence for programs based on the user avatar information and information about all programs watched; the information about the programs that are watched most refers to programs that have been watched continuously for more than 5 minutes; step S3, each time the user watches a program, acquiring the user avatar information through the camera of the tablet computer, mapping the user program viewing habit curve corresponding to the user and the user's overall preference sequence for programs through the user avatar information, and recommending programs to the user based on the user program viewing habit curve and the user's overall preference sequence for programs.

[0018] Among them, in step S1, the method of synchronously calling the tablet computer camera function is: choosing to manually turn on the tablet computer camera function after turning on the tablet computer player; and choosing to automatically start the tablet computer camera function when turning on the tablet computer through application development.

[0019] Wherein, in step S2, a user program viewing habit curve is generated based on the user's avatar information and the information of the programs that are focused on being watched, including: using the user's avatar information as marking information, the horizontal axis is the program viewing time information, the vertical axis is the classification information of the corresponding programs that are watched, and according to the time sequence of watching different classification programs, a rectangular wave curve is generated, and the rectangular wave curve is called the user program viewing habit curve; based on the user's avatar information and the information of all the programs that are watched, a user's overall preference sequence for programs is generated, including: an offline overall preference sequence and an online overall preference sequence; the offline overall preference sequence refers to the user's overall preference sequence when the tablet computer is not playing a program; the offline overall preference sequence uses the user's avatar information as marking information, and according to the user's total cumulative viewing time of different types of programs, the programs are sorted according to the length of viewing time to form the user's offline overall preference sequence, and the total cumulative viewing time is used. The longest one is placed at the front of the sequence; the online overall preference sequence uses the user's avatar information as marking information, and based on the real-time offline overall preference sequence, updates the offline overall preference sequence according to the time of continuously watching programs of the same type to form an online overall preference sequence; the updating of the offline overall preference sequence includes: first, sorting the programs in the offline overall preference sequence from 1 to n, that is, marking the first one in the offline overall preference sequence as 1 and the last one as n; then, updating the time of the real-time continuously watched program; updating the time of the real-time continuously watched program by setting the weight and the total viewing time of the first program; finally, re-sorting the offline overall preference sequence according to the updated time of the continuously watched program to form the user's online overall preference sequence; the formula for updating the real-time continuously watched program time by setting the weight and the total viewing time of the first program is:

[0020] H i =αh1+h i

[0021] Among them, h1 is the total viewing time of the program ranked first in the offline overall preference sequence, h i is the total viewing time of the program ranked in the ith sequence in the offline overall preference sequence, α is the weight coefficient, α∈[0,1], which stipulates that when the continuous viewing time is less than 5 hours, When the continuous viewing time is not less than 5 hours, α=1; H i is the total viewing time corresponding to the program ranked in the i-th sequence in the updated offline overall preference sequence.

[0022] In the above embodiment, specifically, the offline overall preference sequence also includes: as the time spent watching programs increases, the total time of each type of program in the offline overall preference sequence is continuously updated, and the updating method is to add the newly added viewing time of the corresponding program to the original classification of the program, that is, to obtain the latest offline overall preference sequence; the online overall preference sequence also includes: the formation of the online overall preference sequence is updated based on the latest offline overall preference sequence; when the type of program continuously watched is changed, the updated online overall preference sequence is deleted; and the user's new online overall preference sequence is formed by re-updating the latest offline overall preference sequence.

[0023] Wherein, in step S3, recommending programs to the user based on the user program viewing habit curve and the user's overall program preference sequence includes: comparing the programs watched by the user in real time with the user program viewing habit curve to generate programs recommended to the user by the user program viewing habit curve; recommending programs to the user based on the formed new online overall preference sequence; comparing the programs watched by the user in real time with the user program viewing habit curve includes: forming a corresponding user program viewing curve based on the programs watched by the user in real time this time, comparing the formed corresponding user program viewing curve with the user program viewing habit curves of previous times, and calculating the percentage of the formed corresponding user program viewing curve to the user program viewing habit curves of previous times according to the order and time of watching the programs; for example: the corresponding user program viewing curve is formed for watching variety shows for 15 minutes, watching anime for 30 minutes and watching movies for 1 hour; and the program viewing habit curve of a certain user is watching variety shows for 20 minutes, watching TV series for 40 minutes and watching movies for 1 hour; then the percentage is: (15+60 )÷(20+40+60)=63%; because of the mismatch between watching anime and watching TV series, the time spent watching anime this time cannot be counted into the percentage calculation; then, the user program viewing habit curve corresponding to the highest percentage value is selected, and programs are recommended to the user based on the user program viewing habit curve corresponding to the highest percentage value; recommending programs to the user based on the formed new online overall preference sequence includes: taking the top three program types in the formed new online overall preference sequence as user programs to be recommended; comparing the programs recommended to the user based on the user program viewing habit curve with the user programs to be recommended generated by the new online overall preference sequence, if the program recommended to the user based on the user program viewing habit curve is one of the user programs to be recommended, then the program recommended to the user by the user program viewing habit curve is the recommended program; if the program recommended to the user based on the user program viewing habit curve is not one of the user programs to be recommended, it is considered that the user's program viewing habit this time is different from the previous program viewing habit, and the program ranked first among the programs to be recommended is the recommended program.

Claims

1. A method for intelligently playing recommended programs on a tablet computer, characterized in that: The method comprises the following steps: When playing a program through a tablet player, the tablet camera function is synchronously called to obtain the user's avatar information of the user watching the program; A user viewing habit curve is generated based on the user's profile picture information and information about the programs they have watched most. A user's overall program preference sequence is generated based on the user's profile picture information and information about all programs they have watched. The information about the programs they have watched most often refers to programs that have been watched continuously for more than 5 minutes. Every time a user watches a program, the user's avatar information is obtained through the camera of the tablet computer. The user's program viewing habit curve corresponding to the user and the user's overall preference sequence for programs are mapped through the user's avatar information, and programs are recommended to the user based on the user's program viewing habit curve and the user's overall preference sequence for programs.

2. The method for intelligently playing recommended programs on a tablet computer according to claim 1, characterized in that: The synchronous calling of the tablet computer camera function includes: selecting to manually open the tablet computer camera function after opening the tablet computer player; and selecting to automatically start the tablet computer camera function when the tablet computer is opened through application development.

3. The method for intelligently playing recommended programs on a tablet computer according to claim 1, characterized in that: The method of producing a user program viewing habit curve based on user avatar information and information about focused programs includes: using user avatar information as marking information, the horizontal axis as viewing program time information, the vertical axis as classification information of corresponding viewing programs, and formulating a rectangular wave curve according to the time sequence of viewing different classification programs, and the rectangular wave curve is called a user program viewing habit curve; the method of producing a user's overall preference sequence for programs based on user avatar information and information about all viewed programs includes: an offline overall preference sequence and an online overall preference sequence; the offline overall preference sequence refers to the user's overall preference sequence when the tablet computer is not playing programs; the offline overall preference sequence uses user avatar information as marking information, and sorts programs according to the total cumulative viewing time of different types of programs by the user according to the length of viewing time, thereby forming the user's offline overall preference sequence, and the sequence with the longest total cumulative viewing time is ranked by the length of the viewing time. The online overall preference sequence uses user avatar information as marking information, and based on the real-time offline overall preference sequence, updates the offline overall preference sequence according to the time of continuously watching programs of the same type to form an online overall preference sequence; the updating of the offline overall preference sequence includes: first, sorting the programs in the offline overall preference sequence from 1 to n, that is, marking the first one in the offline overall preference sequence as 1 and the last one as n; then, updating the time of the real-time continuously watched programs; updating the time of the real-time continuously watched programs by setting weights and the total viewing time of the first-place program; finally, re-sorting the offline overall preference sequence according to the updated time of the continuously watched programs to form the user's online overall preference sequence; the formula for updating the time of the real-time continuously watched programs by setting weights and the total viewing time of the first-place program is: H i =αh1+h i Among them, h1 is the total viewing time of the program ranked first in the offline overall preference sequence, h i is the total viewing time of the program ranked in the ith sequence in the offline overall preference sequence, α is the weight coefficient, α∈[0,1], which stipulates that when the continuous viewing time is less than 5 hours, When the continuous viewing time is not less than 5 hours, α=1; H i is the total viewing time corresponding to the program ranked in the i-th sequence in the updated offline overall preference sequence.

4. The method for intelligently playing recommended programs on a tablet computer according to claim 3, characterized in that: The offline overall preference sequence includes: as the time of watching programs increases, the total time of each type of program in the offline overall preference sequence is continuously updated, and the updating method is to add the newly added viewing time of the corresponding program to the original classification of the program, that is, to obtain the latest offline overall preference sequence; the online overall preference sequence includes: the formation of the online overall preference sequence is updated based on the latest offline overall preference sequence; when the type of program continuously watched is changed, the updated online overall preference sequence is deleted; and the user's new online overall preference sequence is formed by re-updating the latest offline overall preference sequence.

5. The method for intelligently playing recommended programs on a tablet computer according to claim 1, characterized in that: The method of recommending programs to users based on the user's program viewing habit curve and the user's overall program preference sequence includes: comparing the programs watched by the user in real time with the user's program viewing habit curve to generate programs recommended to the user by the user's program viewing habit curve; recommending programs to the user based on the newly formed online overall preference sequence; the method of comparing the programs watched by the user in real time with the user's program viewing habit curve includes: forming a corresponding user program viewing curve based on the programs watched by the user in real time this time, comparing the formed corresponding user program viewing curve with the user's program viewing habit curves of previous times, calculating the percentage of the formed corresponding user program viewing curve to the user's program viewing habit curves of previous times according to the order and time of watching the programs, selecting the user program viewing habit curve corresponding to the highest percentage, and The user program viewing habit curve corresponding to the highest value of the ratio is used to recommend programs to the user; the program recommendation for the user through the new online overall preference sequence formed includes: taking the top three program types in the new online overall preference sequence as the user's to-be-recommended programs; comparing the programs recommended to the user through the user program viewing habit curve with the user's to-be-recommended programs generated by the new online overall preference sequence; if the program recommended to the user through the user program viewing habit curve is one of the user's to-be-recommended programs, then the program recommended to the user through the user program viewing habit curve is used as the recommended program; if the program recommended to the user through the user program viewing habit curve does not belong to one of the user's to-be-recommended programs, it is considered that the user's current program viewing habits are different from the previous program viewing habits, and the program ranked first among the to-be-recommended programs is used as the recommended program.

Images