Streaming Content Prefetch for Fully Populated Ad Pods
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods struggle to accurately pre-identify and pre-fetch secondary content, such as advertisements, to be displayed with primary streaming content due to challenges like transcoding time, frequency caps, and insufficient content availability, leading to empty or partially filled ad pods.
Innovation Solution
Utilizing machine learning models, particularly neural networks, to predict viewer behavior based on historical patterns and identify periodic viewing habits, allowing for timely request and securement of secondary content to ensure fully populated ad pods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If secondary content requests are made near the time the secondary content pod will be streamed, then the content can be timely delivered, but there is insufficient time to handle transcoding and additional requests, resulting in empty or partially filled ad pods
Solution Approach 1:
The system performs preliminary actions by predicting future content requests using machine learning models before the actual viewing occurs. This allows secondary content to be requested, secured, and transcoded in advance, ensuring ad pods are fully populated when the time comes to stream, thus resolving the time availability vs. completeness contradiction
2Reliability
If machine learning models are used to predict viewer behavior and pre-fetch secondary content, then ad pods can be fully populated, but the system complexity increases
Solution Approach 1:
The patent introduces machine learning models as intermediary components that act as mediators between viewer behavior patterns and content delivery decisions. These models analyze historical data and predict future requests, enabling automated pre-fetching and content selection without requiring complex manual intervention, thus managing system complexity while improving reliability
3Adaptability or versatility
If transcoding is performed for secondary content, then content can be adapted to streaming standards, but the process takes longer than available time
Solution Approach 1:
The system performs transcoding as a preliminary action by completing the format adaptation process in advance of when the secondary content is needed. Machine learning predictions enable the system to identify and transcode content beforehand, ensuring it meets streaming standards and is ready for immediate delivery, thus resolving the adaptability vs. time contradiction
Data Source
AI summary
An aspect of the disclosure related to methods and systems configured to identify a periodic viewing pattern for a first user and/or first user device using spectrum data obtained from time series data using a Fast Fourier Transform. A trained learning model configured to predict content requests is accessed and used to predict content requests for a first time period for the first user and/or first user device. The predicted requests are used to cause content to be provided to the first user device during the first time period.