What is a user-portrait?
It simply can be understood, user portrait is user abstract in real life described it perfectly abstract a user's information the whole picture, and the portrait itself through the label consisting of the whole system of labels constitute the user complete user portrait .
The core part of the user is actually the user portrait label included therein, user tags can be understood as a series of symbols in fact, the user characteristic representation, each tag can be understood as a user's understanding of an angle, the whole of each label assembly dimensions to form a complete user portrait together. Users portrait of providing enough basic information to help businesses quickly find precise user groups and users demand more extensive feedback.
Later in the era of big data, the machine must learn to interpret from the bitstream users to build user portrait becomes especially important, the behavior of the user's data label, we users have an intuitive understanding. At the same time the computer can understand the user, the user's behavior information for personalized recommendation, personalized search, advertising precise delivery and smart marketing and other fields.
Why user portrait?
The core work is the portrait of the user for the user to play tag, play an important purpose of the label is to allow people to understand and convenient for computer processing, such as, the user can do grouping: How many users like baseball? Like baseball crowd, all ages ratio?
We can also use the user to do portraits of data mining: computing association rules, like baseball's what people usually like fast-food brand? Analysis using clustering algorithm, like baseball man sex ratio?
Large data processing, the computer can not do without the operation, the label provides a convenient way so that the computer can be programmed to process information related to the person, even "understood" by the algorithm human model. When the computer has the capability, whether it is a variety of applications search engines, recommendation engines, advertising, etc., we will be able to further enhance the accuracy, improve the efficiency of access to information.
benefit:
Market research to understand the user : change the past, "empiricism" mode of production, forcing the company through market research and we found that user features, understand the real needs of users, more suitable for users design and manufacture of products.
Precision Marketing to meet customer needs : the enterprise to collect information infrastructure attributes of users, social attributes, products, events behavior through data analysis platform, depicts a user's picture, to get a more complete user tag information, the operator can be accurately follow label grouping targeted marketing, in order to facilitate the purchase, improve the user experience and customer loyalty.
Building the user's portrait
Building the user's portrait is divided into the following four parts:
(1) real-time acquisition of user data
(2) the user open up multi-channel information
(3) open up multi-channel product information
(4) user data mining model
We mentioned before, users actually a collection of portraits of user tags, so we build a user-portrait, in fact, in building user tags. As a labeling system, its structure is pyramid-shaped, therefore, how the grading system to the user tag user picture it? We need layers of analytical methods to build user tags.
General principle: step by step subdivided based on a classification of the above categories.
The first category: demographic attributes, asset characteristics, marketing characteristics, hobbies, shopping preferences, demand characteristics
The second category: demographic attributes (basic information, location, character information), characterized assets (held products, trends) ......
The third classification ......
.......
In this way, from top to bottom of the building user portrait basic properties, can be considered as a portrait of the user modeling label.
First, based on statistical analysis of the raw data we collect, we can get the facts label;
Second, modeling analysis, model label;
Third, the model predicts been predicted label.
After the completion of the label model, macro-level summary from the above steps, users will be able to get the portrait data architecture.
to sum up:
With the advent of the era of big data, user portrait essential for long-term development of an enterprise is.
A portrait of the user is constructed to collect the data abstracted from the fact that the model, and then applied in practical applications a simplified model from a closed loop to the fact data. These indicators come from marketing practice, it can reflect the real situation, which also demonstrates the importance and feasibility user portrait in marketing and product operations.