With the development trend of big data technology and data analysis, data-rich analysis-driven enterprises came into being. Let's take a concrete look at the trends and innovations of big data technology and data analysis. This paper also uses some application cases of IBM in helping customers find innovative big data solutions.

1. Data-driven innovation

Nowadays, data has become the cornerstone of the competitive advantage of enterprises. Enterprises that use data and complex data analysis turn their attention to "innovation" to create efficient business processes, help them make strategic decisions, and surpass their competitors in many frontier fields.

2. Rich media data analysis needs advanced technology.

Without reasonable analysis, most of the data are useless. And what opportunities will big data and data analysis bring? International Data Corporation (IDC) predicts that in 20 15 years, the analysis of rich media (video, audio and images) will at least triple, and become a key driving force for investment in big data and analysis technology. Rich media data analysis needs advanced analysis tools, which provides huge market opportunities for enterprises. Take the picture search of e-commerce data as an example. The analysis of image search results should be accurate without manual intervention, which requires powerful intelligent analysis. In the future, with the continuous improvement of the level of intelligent analysis, enterprises will get more opportunities.

3. Predictive analysis is very important.

At present, the application with prediction function is developing rapidly. Predictive analysis improves the overall value by improving efficiency, evaluating the application itself, amplifying the value of data scientists and maintaining a dynamic and adaptive infrastructure. Therefore, the predictive analysis function is becoming a necessary part of analysis tools.

4. Mixed deployment is the future trend.

IDC predicts that in the next five years, the expenditure of cloud-based big data solutions will be four times that of local deployment solutions, and hybrid deployment will be essential. IDC also said that the enterprise metabase will be used to correlate data in the cloud with data outside the cloud. Enterprises should evaluate the products provided by public cloud service providers, which will help them overcome the difficulties in big data management:

Security and privacy policies and regulations affect deployment choices;

Data transmission and integration need a hybrid cloud environment;

In order to avoid too much data, it is necessary to build a business glossary and manage the mapping data.

Build a cloud metadata repository (including business terms, IT assets, data definitions and logical data models).

5. Cognitive computing has opened a new world.

Cognitive computing is a technology that changes the rules of the game. It uses natural language processing and machine learning to help realize natural human-computer interaction, thus expanding human knowledge. In the future, personalized applications using cognitive computing technology can help consumers buy clothes, choose wine and even create new recipes. Watson, IBM's latest computer system, took the lead in using cognitive computing.

6. Big data creates more profits and value.

More and more enterprises make profits by directly selling their data or providing value-added content. IDC survey shows that 70% of large companies have started to buy external data. By 20 19, this number will reach 100%. Therefore, enterprises must understand what their potential customers value, be proficient in packaging data and value-added content products, and try to develop "appropriate" data combinations, and combine content analysis with structured data to help customers who need data analysis services create value.

7. The Internet of Things has promoted the development of real-time analysis.

It is estimated that the compound growth rate of the Internet of Things will reach 30% in the next five years. It will guide enterprises to take the first step of using flow analysis as business driving force. The data explosion caused by the Internet of Things will promote the development of real-time analysis and stream analysis, which requires data scientists and subject matter experts to screen data and find repeatable patterns that can be developed into event processing models. Then, the event processing model can process incoming events, associate them with related models, and monitor real-time situations that require responses. In addition, the event processing is uninterrupted, so the response time is required to be as close as possible to the actual time. Therefore, event handling has become an indispensable module in big data systems and applications.

8. Competition for talents of compound data analysis

Many enterprises want to combine business knowledge with business analysis, but it is difficult to find compound data analysis talents. In particular, large enterprises are deeply touched by this. With the increasing use of technology by enterprises, the demand for compound skills is becoming more and more obvious. The combination of business knowledge and analytical skills is very important for speed-driven enterprises, which helps enterprises to deeply understand business drivers and related data, so as to turn business insight into action more quickly.