Not only reflect the value of data in the enterprise, individuals may also experience the charm of the data, exploratory behavior password using the technology, big data approach allows everyone, welcome straight were concerned about my public number, we can discuss data those funny things.
My public number: livandata
1, the data life cycle analysis is as follows:
2, if there is force grid ask a question:
Generally based on a scene, such as:
In the case of operations is often pv drop, if directly asked: Why is my pv dropped? It estimated that no one can answer, as the operations staff will be very clear understanding of their activity, based on a scenario to a question, that is what our recent on-line activities, but declined pv Why?
At this time, the analyst can analyze the causes data pv falling from user quality, positioning activities.
3, personalized portraits of operations by the user:
1) labeling system approach:
How to tag the user, while reducing customers using the product scene.
Build a user portrait:
After seeing the picture of the user, reducing the user's usage scenario:
Then how to tag the user:
Each click, a user browsing behavior will generate behavior identification, pages visited, the length of time browsing, access to what.
For example: the establishment of successful registration tags:
User behavior has completed all of the data to be registered, the user has completed the process will be marked with the label "registration successful".
Users portrait of four stages:
User portrait significance:
Users portrait of difficulty:
Microcosmic example of a user portrait:
User portrait label Modeling:
Portrait of a user data structure:
How to personalize a user portrait operations:
The first stage is the era of a seller's market, and the second stage is based on the subdivision rules, and the third phase is based on a personal recommendation, that thousands of thousand faces, not the same.
Personalized operations Value:
Fine recommendation is to render the content in the long tail to the right people.
Personalized operations of online applications:
According personalized purchase data operations:
To personalized recommendations based on the search term:
According personalized user complex data operations:
My public number: livandata
