Search with “big data analysis” as the keyword, you can see the concept of “data analysis platform” and “visualization” in most entries.
Humans love and are good at using and making tools when development is limited.
Humans are far more efficient at taking in graphic information than they are at using words and Numbers.
Therefore, with the explosive growth of both the amount of information flow and the speed of flow, the visualization big data analysis platform is naturally created.
Data visualization aims to convey and communicate information clearly and effectively by means of graphics. Rich charts can be indicated by visual cues such as position, length, Angle, direction, shape, surface base, volume, saturation, and hue to achieve the effect of focusing the line of sight to express the meaning of ICONS.
In the context of enterprise operation and maintenance decision making, the collected enterprise data are visualized with appropriate data model and statistical analysis method, and the visualization results are translated into clear and accurate conclusion reports, so as to assist the enterprise to understand the operation status, find the development sticking points and assist the decision-making.
The following figure shows the visual types corresponding to the different types of analysis requirements.
According to Gartner’s recent industry reports and at the conference, some trends in the future of data visualization can be seen.
1. Data map
“The ability to create highly interactive dashboards and content through visual exploration and embedded geospatial analysis.”
Gartner said “geographical spatial analysis”, is the “data map”, dedicated to the presentation and analysis of these large data related to the map, this is indeed a very good function, is used to analyze business data on geographic level has considerable value, especially to the now more than ever on the geographical distribution of a wide range of industries, can accurately locate the problem country city or even a operating point. It will not only be more intuitive than the simple table, but also have the effectiveness of information communication and professional image. In terms of type, the data map can be divided into regional map, combined map, marker map, single-layer map, customized picture map, flow map, thermal map, etc.
2. Interactive exploration
“Data exploration through a series of visualization options that are not only basic data graphs like pie charts, bar charts and line charts, but also heat and tree charts, data maps, scatter charts, and other special-purpose charts that better meet the needs of all industries. These tools enable users to analyze and manipulate data through direct visual interaction with the data.
The ability of interactive experience of data visualization is mainly evaluated through its ease of learning, ease of use, ui-friendliness and efficiency of use. Its manifestation is mainly reflected in the operation of directly graphing the data through visual charts, such as up-rolling, down-rolling, slice-rotating, and so on, as well as the important function of data linkage, which provides users with comprehensive data planning and in-depth methods and angles of data mining.
3. Data story
“Data visualization should be a story, and the data should be visualized to meet the needs of the reporters, so as to help them tell a complete data story and make the viewers better understand the theme.”
The results of data visualization can be related to each other to form a story-oriented kanban. The data can be interpreted from a more complete and diversified perspective, showing more details and drawing viewers into the story.
4. Intelligent explorations based on search
The search-based and visualization-based data exploration analysis feature will be integrated into the next generation of data analysis products as a component of the new BI and analytics platform.” ”
By 2021, the number of users of new BI and visual analysis platforms with intelligent data exploration and analysis capabilities will be double that of products and platforms without such capabilities, and will create twice as much business value” ”
Most of the visual data analysis products on the market now adopt the drag-and-drop exploration method, which is simple and quick, but has considerable limitations. When analyzing more and more complex dimensions, drag-and-drop operation will bring a variety of inconvenience to users and require users to have a deep understanding of the data structure to form an ideal visual result. The search exploration method similar to the search engine exploration result method not only contains the advantages of simple and quick drag-and-drop, but further frees the analysts from the premise of having to understand the data structure, and directly enters the business problems into the analysis platform to form visual results.
5. Natural language
“By 2020, 50 percent of analytical queries will use search, natural language processing or speech generation, or will be generated automatically.”
Way to explore the natural language generation and search type, the combination of perfect fit the demand for data visualization analysis of business personnel, become future vision, data visualization platform can understand users using natural language to describe the business logic of the query on demand, into program can accurately understand the query, regeneration results into visual feedback to the user.