The path analysis functions automate path analysis. They are useful for clickstream analysis of web site traffic and other sequence/path analysis tasks, such as advertisement or referral attribution.
Terms Function Descriptions Use
Term | Definition |
---|---|
Path | Ordered, start-to-finish series of actions, such as the page views of a user from the start to the end of a session. For example, if the user visits page a, page b, and page c, in that order, the path is: a,b,c |
Sequence | Path in this format:^,path The carat (^) indicates that a path follows. For example: ^,a,b,c |
Subsequence or prefix | For a given sequence, a possible subset of steps that start with the initial step. For example, for the path a,b,c these are the subsequences:^,a ^,a,b ^,a,b,c |
Exit subsequence or prefix | Subsequence or prefix that is the same as its sequence, indicated by a final dollar sign ($). For example: ^,a,b,c$ |
Depth | Number of steps in a sequence or subsequence. For example, the immediately preceding subsequences have depths 1, 2, and 3, respectively. |
Node | Single step on a path. For example, one web page that the user visits during the session. |
Parent | Path the user traveled to a given node. For example, the parent of c is: ^,a,b |
Child | Path the user traveled from a given node. For example, these are the children of ^,a:^,a,b ^,a,b,c |
Path Analysis Functions
Function | Description |
---|---|
PathGenerator | Takes a set of paths and outputs the sequence and all possible subsequences. |
PathSummarizer | Takes PathGenerator output and returns, for each prefix in the input table, the parent and children, and number of times each of its subsequences was traveled. |
PathStart | Takes PathGenerator output and returns, for each parent in the input table, the parent and children and the number of times that each of its subsequences was traveled. |
PathAnalyzer | Runs the preceding path analysis functions in the order shown, using the output of PathGenerator as input to PathSummarizer and the output of PathSummarizer as the input to PathStart. |