Big Data Isn’t Almighty

2018-03-31 04:10
风景园林 2018年8期

Since the earliest knot-tying recording in human history, the collection, analysis and processing of data has become part of the human civilization. Later on, each time a new data recording or processing method was produced in human society, the world would undergo dramatic changes. The use of language enabled information to be better disseminated and utilized, and the creation of scripts allowed information to penetrate further into the social life and capacitated information preservation for long time. The printing instigated information widely distributed to boom data assortment and accumulation which has completely changed people’s perception of the world. In modern times, the rapid development of computer,mobile communication and internet technologies has brought human society into the information era. With digitalization more adopted for recording, analyzing and processing data, the main place for data storage has moved from books to hard disks and cyberspace to make human society facing another revolution.

Nowadays every one of us by daily behaviors leaves plenty information intentionally or unintentionally as the data producer to be linked into an invisible data network. In the era of rapid development of internet technology, layers of data keep aggregating in increased types and on amazing scale that almost to go beyond humans’ grip.

The huge amount of complex data reflects the environmental conditions of the entire world and all aspects of human life. For all professions and research fields one can simply extract the resources interested from the data base for analysis processing with modern technological methods to discover the rules. So in many fields,there has evolved the Big Data + thinking research and working models, that is, to collect relevant massive data, perform analytical processing to make judgments, decisions and conclusions.

For example, city as a complex huge system, the data left by its residents’ daily activities is of great value. These data reflect aspects of urban life in terms of time and space, such as the demographic composition and dynamics of the city, the economic development,traffic condition, space usage, employment status of residents, travel,recreation, shopping and people’s preferences. The data also reflect the state and changes of the urban ecological environment, such as the meteorological, hydrological and biological conditions, addresses and pollutions. Collecting, summarizing and analyzing the data can portrait the complex urban systems more deeply and intuitively from multiple dimensions, which shall enable better urban planning and management, more efficient urban land uses, more ecological-sound urban natural systems, enhancing adjustment and optimization of urban and regional structure and spatial layout, improvement of urban traffic conditions, and improvement of urban infrastructure,which have brought new perspectives and approaches.

Big data as a brand new tool to analyze many other complex systems go beyond cities. If the data is analyzed for use appropriately, all complex systems can be better understood for improved decision-making and optimized operations.

However, big data is by itself a complicated system. There are different categories of data, some of which truly reflect the development of nature and society while others has high uncertainty in terms of truthfulness, completeness and objectiveness.

I once read an article written by a designer about his design works and I also heard a guest scholar introducing his design in a symposium. Both started with plenty of information collected from Weibo and online forums regarding the public audience’s comments on their works to conclude that the works are widely welcomed so are successful. However, each of us may have the experience that though we have our own opinions on certain issues online, but for one reason or other we never or rarely leave comments accordingly. In fact, people who leave messages on the Internet often only cover a small part of the total netizens.They are specific groups less representative of the whole netizen groups by age, social class, and education level. The opinions of a large number of netizens are either hidden inside or expressed beyond the cyberspace thus leaving no trace on the Internet. In other words, the online messages about certain issue may not all authentically reflect the general public viewpoints. Moreover,online data is also very easy to be influenced by human force, even manipulated by internet marketer. Also, I have good reason to suspect that the data online maybe filtered according to somebody’s own likes and dislikes. Therefore, the conclusions drawn from the online message as the basic information, analyzed by the method of big data collection are not that convincing. If one keep using online data to extract the theoretical findings and guide design practices, they are taking false steps far from the right destination.Therefore, the large amount of data obtained from the internet or similar methods especially the subjective and criticized data tend to be biased and over generalized. There is a huge risk if decision is made based on the analysis of such datum.

Big data is not almighty. Even for the large amount of objective data that faithfully reflect the realities, it is hard to say that existence is reasonable. The decision bending to objective reality can be also be risky. For works that requires imagination and innovation, the role big data can play is actually very limited.

China as the world largest internet and smart phone user has the most abundant data resource. Due to the popularity of mobile payment, shared bicycle, WeChat and Weibo, huge amount of data are accumulated on various platforms. To comprehensively utilize these data, scientific research and calculation with new technologies of cloud computing and artificial intelligence can be carried out, which will undoubtedly enable urban planners, designers, builders and managers to make more accurate judgments far superior than those made on basis of theories, experience, and samples. The value and application prospects of big data are immeasurable. However, what issue are reflected from a given datum? What kind of data is needed to solve a given problem? What kind of algorithm can achieve the optimal result? Answers to these questions cannot be obtained without human research and judgment. Therefore, big data, like other data revolutions in human history, has liberated part of human labor but meanwhile put forward more serious challenges for human beings.

Translator: WANG Xiyue

Chief Editor: Prof. Dr-Ing. WANG Xiangrong

August 5th, 2018