Business Intelligence systems-corporate business, system execution statistics and analytics

Opening

Commercial intelligence systems, business INTELLIGENCE,BI, data centers, different names, and other responsibilities, the following collectively referred to as the BI system.

BI load is the impact of the company's business, the implementation of the system and the effect of revision. , to carry out comparative objective statistics and analysis. For the senior management of the company operations, for the product Department for the site design and revision or algorithm adjustment before and after the effect, to provide reference and consideration.

The BI system is responsible for statistical analysis related to the real-time data requirements are not high, agree to more than one day of data latency, and detailed business closely related to, or real-time requirements of high statistical analysis. Should not be placed in the BI system, but should be placed in the respective business support system.

A bi system is a data system. A collection of technologies used to help enterprises better utilize data to improve decision quality is the process of drilling information and knowledge from a large amount of data. Simply speaking is the process of application of business, data and data value. Basic BI is statistical analysis, and the advanced level of BI is decision support. From "What happened before, why it happened", to "what's happening now." What will happen in the future. "

noun explanation

BI: Business Intelligence (Intelligence)

ETL: Extraction, conversion, and onboarding (extraction-transformation-loading)

ODS: Operational datastore (Operational data Storage)

DW: Data Warehouse (Warehouse)

DM: Data mart, data Mining

OLAP: Online Analytical processing (online analysis process)

System-Resolved methods

(1) According to the need of statistics, data extraction. From the business System database. Extracting to a BI database

(a), the process of extraction, may be a daily timing, hourly timing, triggering and other ways. Details are to be based on the detailed scenario

(b), the extracted logic, to be able to ensure repeated extraction, for example, one day when the data failure.

。

(c), the process of extraction, may involve some computational logic. For example, the calculation of gross profit margin requires the cost price of the commodity. For the acquisition of cost price, there may be some logic

(d) Methods of data extraction. The ability to introduce third-party frameworks that streamline development workloads, such as Datax

(2) Data collection

(a), data labelling: depending on the needs of the business, some data need to be flagged, analyzed through the log, or recorded in the database, such as the order: source, ref, POS

(b), site traffic: Insert code or analysis log, http://blog.csdn.net/puma_dong/article/details/38943251#t12

(c), log Analysis: Collect user Access path. Dwell time. Exit page, etc.

(3) Results presented

(a), the use of a good JS framework graphical presentation of BI reports, such as the jquery plugin, but the EXTJS framework is not recommended. Too much packaging and the reason for learning costs

(b), SMS reminders, reminders of important results

(4) Formula definition

(a) Bi's product definition, a clear definition of the content of the data to be displayed. For example, gross margin formula, conversion rate.

system Function settings

Present:

(1) Real-time Data report

(2) post-sale related reports

(3) Commodity inventory related report

(4) channel advertising related reports

(5) Site traffic related reports

Data:

(1) Traffic statistics

(2) Data extraction Middleware

Cond

Without a skeleton there is no flesh or blood. The so-called with. Mao, write it down when you think about it. Maybe plus.

。

。

Copyright notice: This article Bo Master original articles, blogs, without consent may not be reproduced.

Business Intelligence systems-corporate business, system execution statistics and analytics