Video Pushing Method Based on Search Keyword and Electronic Device
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video recommendation systems face issues where users must exit the internal stream to re-select videos, leading to increased processing time and user dissatisfaction due to irrelevant video recommendations based on ambiguous search keywords, resulting in resource wastage and poor user experience.
Innovation Solution
A method and apparatus for video pushing that transmits search keywords to a server, receives and displays video preview information, and upon user interaction, retrieves and displays relevant video previews on the same interface without requiring a return to the external stream, ensuring more accurate video recommendations based on user intent.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If videos are pushed to users based on search keywords in the external stream, then users can find multiple types of videos matching the keyword, but the user has to exit to re-select videos, increasing processing time and reducing user experience
Solution Approach 1:
The system segments the video recommendation process into two stages: initial broad search in the external stream showing multiple video types, and subsequent targeted recommendation in the internal stream based on user interaction. This allows comprehensive coverage while reducing the time needed for each individual recommendation action.
Solution Approach 2:
The patent introduces an intermediary mechanism where the terminal device acts as a mediator between the external stream search results and internal stream video playback. It captures user interaction data from the external stream and uses this to trigger targeted video pushing in the internal stream, eliminating the need for users to manually re-select videos.
2Adaptability or versatility
If multiple types of videos are recommended based on ambiguous search keywords, then more video options are provided, but search noise increases and user experience deteriorates
Solution Approach 1:
The system dynamically adjusts the video recommendation strategy based on user interaction. Initially, it provides diverse video types matching the search keyword in the external stream. Upon detecting user engagement with a specific video type, it dynamically switches to providing only videos of that type in the internal stream, thereby reducing search noise while maintaining versatility.
Solution Approach 2:
The patent implements a feedback mechanism where user interaction data from the external stream (such as playback initiation) is fed back to the terminal device, which then uses this feedback to refine and adjust the video recommendation strategy for the internal stream. This feedback loop eliminates irrelevant recommendations while preserving relevant video diversity.
3Measurement precision
If users manually re-select videos after watching one, then precise video selection is achieved, but operating complexity increases
Solution Approach 1:
The system performs self-service by automatically analyzing user interaction data from the external stream and autonomously generating targeted video recommendations for the internal stream. This eliminates the need for users to manually re-select videos, maintaining high selection accuracy while significantly simplifying the operating process.
Solution Approach 2:
The terminal device performs preliminary action by capturing and analyzing user interaction data during the external stream viewing phase. This preliminary data collection and analysis enables the system to pre-configure accurate video recommendations for the internal stream before the user needs to re-select, thereby maintaining precision while reducing operational complexity.
Data Source
AI summary
Provided are a video pushing method and apparatus, and an electronic device. The method includes: transmitting a received search keyword to a server; receiving first video preview information, fed back by the server, of at least two first videos matching the search keyword and including at least two types of videos, and displaying the first video preview information on a first display interface; in response to receiving a play-triggering operation on first target preview information in the first video preview information, transmitting, to the server, a video-obtaining request carrying identification information of a first target video corresponding to the first target preview information; and receiving at least two second videos and corresponding second video preview information that are fed back by the server based on the video-obtaining request, and displaying a first predetermined number of pieces of preview information in the second video preview information on a second display interface.