Abstract Since the big data has entered people's sight, it has gradually become the focus of people's attention. Big data is about the science of the PB era. Essentially, the challenge of big data is the challenge to science in the PB era, and it is also the challenge of cognitive science including data mining. that...
Since the big data has entered people's sight, it has gradually become the focus of widespread concern. Big data is about the science of the PB era. Essentially, the challenge of big data is the challenge to science in the PB era, and it is also the challenge of cognitive science including data mining. So, how do data mining in the era of big data? In today's era, people often refer to big data mainly including three sources: The first is the big data of nature, that is, the natural environment on the earth, which is very large. The second is life big data. The third and most important thing is the social big data that people care about. These data are ubiquitous in people's mobile phones, computers and other devices. Today, a report can be known to people all over the world within 3 minutes.
Obama’s inaugural social scene, with so many faces, has a story under each face, and everyone has big data behind it. Faces are important identifiers for data security. How do you make face recognition clear? People think of a lot of ways. Now there are 800,000 cameras in Beijing. We drive and shop every day under the supervision of the camera. We can use the camera for identity authentication, age recognition, emotional computing, kinship discovery, psychological recognition, regional identification, and ethnic recognition. The main form of this kind of streaming media is unstructured, the relationship between features, the accuracy of device algorithms, etc., all seriously restrict the progress of big data face mining. How to extract the required feature attributes from these massive data using the recognition algorithm and clarify the relationship between the features are the problems now facing.
Technology drives computer development
In 1936, the genius mathematician Turing proposed the Turing model. Later, a computer converted the Turing model into a physical computer. There were three major blocks: CPU, operating system, memory and external memory, as well as input and output. In the first 30 years of computer development, we invested the most in CPU, operating system, software, middleware and application software. At that time, people focused on the improvement of computing performance, we call this era the computing era.
Computing has put a lot of effort into the software, especially high-performance computers. We believe that the calculation played a leading role in the first 20 years, and its mark speed is the molar speed. In such an era of computing leadership, what we mainly do is the mining of structured data. The father of relational databases, Edgar, proposed a relational model in 1970, with relational algebra as the core operation, and a two-dimensional table representation of the relationship between entities and entities. For three or four decades, database and data warehousing technologies from all walks of life, as well as data mining from database discovery knowledge, have become a huge information industry.
Relational algebra is a formal theory and constraint of relational databases. It has a top-level design and data structure, and then fills in the cleaned data. The data is rotated around the structure, and the data is rotated around the program. Users do not need to care about the acquisition, storage, analysis, and extraction of data. Through data mining, classification knowledge, associated knowledge, time series knowledge, abnormal knowledge and so on can be found from the database.
With the expansion of the database industry, people are not satisfied with the database, so Databases is said to be big data, which has encountered two unavoidable challenges. The first challenge is that the formal constraints of relational algebra are too harsh. Unable to represent real-world data; the second challenge is that as the amount of data increases, the performance of relational algebraic computing decreases dramatically. At this time, our storage technology has developed rapidly, and humans have entered the search era. Search because the storage is cheap, the storage speed is about doubled every 9 months, so the storage has driven the pace of technology. This search era has evolved over 20 years and led us into a semi-structured data mining era. The representative of this era is Tim Berners-Lee, the father of the World Wide Web. He proposed hypertext ideas and developed the world's first web server, so we can retrieve another server from one server. Content, the server can release fragmented hypermedia information including text, forms, pictures, audio and video with the support of software.
Therefore, the client server structure and the cloud computing structure are flourishing. At this time, there are no strict formal constraints on algebra. The main ones are the norms and standards. All media exist in the form of entities, even software. The link generates a link.
Formal theory is much looser than relational algebra, creating flexible and diverse entities. At this time, data begins to move around the entity, and the entity rotates around the link. In the context of cloud computing, data mining can also be regarded as a search and personalized service in the cloud computing environment. There is no fixed query method, and there is no unique, 100% accurate query result.
Networked big data mining
With the doubling of Internet bandwidth in six months, humans have entered the era of interaction, and interaction has driven the development of computing and storage.
Big data mining in the era of mobile Internet is mainly unstructured data mining in a networked environment. These data forms reflect fresh, fragmented, heterogeneous, and emotional raw data.
The characteristic of unstructured data is that it is often low-value, strong-noise, heterogeneous, redundant, cold-cold data, and there is a lot of data that is not used in memory. The formal constraints of data are becoming more and more relaxed, and are getting closer to Internet culture, window culture and community culture.
The object of concern has also changed a lot. The first thing to focus on is the niche. Only by satisfying the needs of niche mining can we meet the needs of more masses of the masses. Therefore, an important idea is to win from the bottom up. The top-down design from top to bottom emphasizes the authenticity and timeliness of mining data. It is necessary to discover associations, discover anomalies, and discover trends. In short, we must discover value.
Currently, deep learning is also a data adaptive simplicity. If we use deep learning to search for a face pixel search on Baidu, who is this person? The amount of data increases dramatically, various media forms can be arbitrarily fragmented, organizational structure and mining procedures must be around data, procedures To be fragmented, and to be able to reorganize at any time, mining is often the discovery of different communities in the human-computer interaction environment and the group intelligence formed in the community. In unstructured data mining, data cleaning is naturally performed, and semi-structured data is naturally formed. And structured data to improve data usage efficiency.
Group intelligence is a word that has been said a lot lately. We used to do a Turing test on a computer to let the computer distinguish which codes are generated by humans and which are machine-generated. This is proposed by Carnegie Mellon University. When you shop online, log in to the website, or apply for a website, you will encounter an adapter code that is used. I would like to mention the third representative, Louis, who proposed to use this adaptation code.
If cloud computing supports big data mining to discover value, then we believe that cloud computing is originally an Internet-based public participation computing model. Its computing resources are dynamic, shrinkable, virtualized, and provided in a service manner. Produce system upgrades that are free from traditional configuration, more concise, flexible, and personalized. Mobile phones, game consoles, digital cameras, and TV sets are subtly different. More iCloud products appear, and the interface is user-friendly and personalized. Become a terminal for big data mining.
The excavator supports a variety of big data applications. If we have a data collection center, a storage center, a computing center, and a service center, we must have a data mining center. In this way, we can realize the timely application and value of supporting big data. Timely discovery.
Big data marks the arrival of a new era. This era is characterized not only by the pursuit of rich material resources, but also by the ubiquitous Internet to bring convenient and diversified information services, as well as the value mining of data resources that are different from matter. Value conversion, the information value mining of the virtual world leads to more precise control of the material and energy of the physical world, as well as the new phenomenon of spiritual and cultural aspects brought about by big data mining.
SJ736B, MORE STABLE, EASY MAINTENANCE
Shaoxing Sunfong Intelligent Equipment Inc. , https://www.zjsxsunfong.com