Real-Time Audience Forecasting via Parallel Data Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advertisers and web publishers face challenges in predicting the real-time size of their target audience for advertising campaigns due to limited access to user behavior data across various websites, making it difficult to estimate audience volume effectively.
Innovation Solution
A system that analyzes historical audience data to provide real-time audience volume predictions, allowing users to specify queries based on various categories and tune them iteratively, using parallel processing by a cluster of servers to deliver predictions quickly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If historical audience data is analyzed using traditional processing methods, then audience volume predictions can be generated, but the prediction speed is too slow to meet real-time requirements
Solution Approach 1:
The system divides the historical audience data into multiple segments or partitions that can be processed independently and in parallel. This segmentation allows the cluster of servers to process different data subsets simultaneously, dramatically increasing prediction speed while maintaining accuracy through aggregated results from all segments.
Solution Approach 2:
The system performs preliminary processing and preparation of historical audience data before the actual prediction is needed. This includes pre-aggregating data, creating intermediate results, and preparing computational resources in advance, so that when a prediction is requested, the system can quickly complete the final calculation without processing raw data from scratch.
2Loss of time
If real-time audience forecasting is implemented, then advertisers can optimize campaigns timely, but the system complexity increases significantly
Solution Approach 1:
The system is designed as a universal platform that handles multiple advertising campaigns, different audience types, and various prediction scenarios through a single unified architecture. This multi-functionality reduces overall system complexity by avoiding separate specialized systems for each campaign type, while still providing real-time forecasting capabilities across diverse use cases.
Solution Approach 2:
The system introduces intermediary components such as data preprocessing layers, standardized interfaces, and abstraction layers between the raw data and the prediction algorithms. These intermediaries simplify the core prediction logic by handling data validation, transformation, and coordination, thereby reducing the complexity of the overall system while enabling real-time processing.
3Measurement precision
If user-specified queries are allowed for audience segmentation, then prediction accuracy improves, but the data processing time increases
Solution Approach 1:
The system dynamically adjusts the level of query complexity and data processing depth based on the specific prediction request. For simple queries, it uses pre-computed results and faster algorithms, while for complex user-specified queries, it selectively processes only the necessary data subsets. This dynamic approach maintains high accuracy for customized queries while minimizing unnecessary processing time.
Data Source
AI summary
A system, method, apparatus and processor readable media are described for real-time prediction of an advertising audience volume through analysis of historical audience data, and tuning of the predicted audience volume. Embodiments enable a user to specify a query for audience volume prediction. Such a query may be a Boolean combination of various audience categories. A time range may be determined that indicates the amount of historical data that is to be analyzed to make the audience volume prediction in real time. Employing the user-specified query, an audience volume prediction may be provided for a future time period, based on an analysis of retrieved historical audience data for the time range. Embodiments may also enable a user to tune the predicted audience volume through modification of the query through one or more iterations.