Session Count Determination via Interval Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining unique session counts in web analytics face challenges such as the count-distinct problem, which often result in either approximate counts or resource-intensive data storage and analysis, leading to inefficiencies in tracking active sessions over time.
Innovation Solution
The approach involves determining unique new and carry-forward session counts for predefined time intervals, summing these counts to accurately calculate session counts for specific time periods, using a system architecture that includes session recorders, real-time analytics, and databases like Cassandra for efficient storage and retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If HyperLogLog technique is used to determine approximate session count, then the count-distinct problem is solved, but the measurement precision deteriorates because only approximate count values are obtained
Solution Approach 1:
The patent divides the time period into multiple intervals and segments the session data accordingly. For each interval, it calculates the number of new sessions and carry-forward sessions separately, then combines these segmented counts to get the total unique session count for the requested time period. This segmentation allows precise counting without requiring storage of all session data.
2Measurement precision
If entire session data is stored in big database for later analysis, then accurate session count can be determined, but the device complexity and resource consumption increase significantly
Solution Approach 1:
The patent extracts only the necessary information from the session data - specifically, the session start times and their corresponding intervals. Instead of storing and analyzing entire session records, it stores only the interval assignments and counts, dramatically reducing the data volume while maintaining counting accuracy.
Solution Approach 2:
The system performs preliminary action by assigning sessions to time intervals in advance and pre-calculating the new session and carry-forward session counts for each interval. This preliminary processing allows accurate session count queries to be answered quickly without needing to store and process the complete session history.
3Measurement precision
If session data is stored and analyzed for specific time periods, then accurate session count for requested period can be obtained, but the loss of time increases due to data storage and analysis process
Solution Approach 1:
The system performs preliminary action by pre-assigning sessions to time intervals and pre-calculating counts for each interval. When a session count query is requested, the system can quickly retrieve and combine pre-computed values from the relevant intervals without needing to process the entire session history in real-time, thus reducing query response time.
Data Source
AI summary
A method and system for determining a session count is described. At a user interface a request is received to determine a session count for a time period. Based on the received request, the session count is determined based on unique new session counts corresponding to one or more time intervals included in the time period and carry forward session count corresponding to an initial time interval included in the time period. Finally, the determined session count is displayed at the user interface.