China's data output accounts for 10.5% of the world's data, and the huge amount of information has given rise to "algorithm innovation"
Internet "big data" is not just "big data" ("Integration" view of China)
——One of the series of reports on “Algorithm” in perspective
Reporter Lu Zehua
On various network platforms, "algorithm" has become a widely used technology. Through data analysis and precise recommendations, it has changed the past search model of "people looking for information" and gradually realized "information looking for people."
Opinions vary when it comes to “algorithms.” Some people think that it is more convenient for users and provides more accurate network services; some people think that it is "calculating" users, causing problems such as "information partiality"; and some people think that as a technology, it is not good or bad. , the key lies in how to control it with correct values and norms.
Starting today, this edition launches a series of reports called "Perspective on 'Algorithms'" to interpret the changes and impacts brought about by "algorithms." The first article in a series of reports focuses on the changing habits of netizens in obtaining information in the era of big data.
status quo
Richer content and more diverse platforms
"When I encountered a problem before, my first reaction was to open the browser and search. Now, in addition to searching the web, I will also go to WeChat or short video platforms to find answers." Wang Ying, a Beijing citizen, showed a short video to reporters: "Look, before After two days of cooking, I wanted to make a fish-flavored shredded pork dish, and when I looked for it on Douyin, an instructional video came out.”
Wang Ying is 36 years old and has been exposed to the Internet since around 2000. In the first 10 years or so, she mainly obtained information in two ways, one was by searching on search engines, and the other was by browsing portal websites. But in recent years, this habit has begun to change. “There is so much information online now, it’s dazzling, and it takes a lot of effort to sort through it. Sometimes when you need specific information, it’s easier to find the results in mobile software.” She introduced while opening the mobile phone interface: “For example, if you want to find shopping information, just Go shopping on Taobao; if you want to watch movies, go to Maoyan; if you want to know about celebrity news, go to Weibo; if you want to gain professional knowledge, go to Zhihu Search..."
In Wang Ying's view, this change was formed unconsciously. We use all kinds of mobile apps frequently every day, so it doesn’t take much effort to “search for them”. Over time, I developed the habit of using multiple platforms to search for information. The most important thing is that after "searching", the mobile application will remember your needs and actively recommend information of interest, saving time and effort.
Wang Ying’s experience in using the Internet is somewhat representative. The reporter conducted a small survey on the search behavior of 120 users. The results showed that everyone has the habit of using search engines, but only 15% use it as their only choice. Most people said that they search for information in various mobile applications and often browse the information recommended by various platforms. For example, to obtain life experience information, many people will choose to watch short videos, "When you are in doubt, learn from videos."
Industry data also confirms this. As of June 2023, the number of search engine users in China has reached 841 million. However, nearly 70% of users also complete searches through short videos, and WeChat’s “Souyisou” monthly active users have exceeded 800 million, almost equal to the total number of search engine users. Another survey also showed that users search for information on an average of 3.8 platforms every day.
challenge
Miscellaneous information and “platform barriers”
Although the amount of information on the Internet is greater and there are more channels to obtain information, many users still say that "the user experience is not good."
Information confusion is one of the reasons. "There is a lot of online marketing information now. Although most of it is marked with the word 'advertising', it still needs to be screened, which affects the efficiency of use." Wang Hui is a financial analyst who often needs to retrieve information online and has high requirements for information quality. . But he found that it is very difficult to find a "pure web page" now, which has slowed down his work pace to a certain extent.
Wang Hui’s situation is not unique. The reporter's investigation found that when retrieving information online, many of the displayed results are advertising links. For example, when searching for the word "training", the top two results are advertisements for commercial training institutions.
It is understood that major search websites have also noticed this problem, but how to change the long-established business model is still a proposition that has yet to be solved. "Advertising is the main source of profit for websites. At present, the industry can only try to optimize advertising information and improve user experience. However, it will take some time to fully transform." A sales manager of an Internet company told reporters.
In addition to miscellaneous information, there are also “platform barriers” to varying degrees in the Internet world.
It is understood that in the past, comprehensive search websites were the main entrance for various network platforms to obtain traffic. Through a search box and a keyword, users could search for information on various platforms across the entire network. With the advent of the mobile Internet era, the number of users of vertical platforms such as food, home furnishing, travel, and short videos is getting larger and larger. Each platform has formed its own unique "big database". This "big database" not only contains a large number of high-quality Information also includes data such as user usage habits. Online platforms with hundreds of millions of users no longer need other platforms to "drain traffic". These data are regarded as exclusive resources and are no longer open to the outside world. This makes it difficult for users to find all the information they need through comprehensive platforms.
trend
Develop towards intelligence and precision
Today, China has become one of the countries with the largest amount of data and the richest types of data. Information recently released by the National Data Administration shows that my country’s data production accounts for 10.5% of the world’s total data. Some research institutions predict that by 2025, China's total data may account for nearly 30% of the world's data.
“As a new factor of production, data has been rapidly integrated into all aspects of production, distribution, circulation, consumption, and social service management. However, in the current development process, there are also problems such as insufficient release of data application potential, low quality of data supply, and poor circulation mechanisms. problem." said Shen Zhulin, deputy director of the National Data Administration.
In this context, "big data" alone can no longer meet social needs. "Data precision" and "data accuracy" must be further achieved, thereby effectively reducing the cost of users obtaining information and providing more personalized and refined content services. Driven by this demand, “algorithm” technology emerged.
"Algorithm" is the connection node between artificial intelligence technology and big data technology. Pan Jiaofeng, president of the Institute of Science and Technology Strategy Consulting of the Chinese Academy of Sciences, believes that artificial intelligence is changing the way people communicate in many scenarios such as scientific research, trade, transportation, and medical care. Its characteristics of a "living dictionary" and a mobile "encyclopedia", coupled with language understanding and production capabilities that are getting closer and closer to human levels, make its "responses" both common sense and integrated innovation.
Therefore, based on the massive data on the Internet, "algorithms" can use artificial intelligence technology to analyze and summarize user interests, preferences and needs, draw user portraits, and "tag" characteristics, and then use this as a basis to carry out precision and personalization. customized recommendations.
With the deployment of "algorithm" technology, users can receive information push based on their personal needs every time they open the mobile app. At the same time, by continuously collecting information, the platform will become more and more "understanding" of users - when you open a shopping website, the products you want to buy recently have been recommended to the homepage; when you enter the short video platform, what you like comes one after another. video content; browsing travel software, the places you want to go to are listed one by one, and even the links to purchase air tickets are placed... all of this is achieved with the increasingly mature algorithm technology.
However, some people worry that while "algorithms" accurately allocate network information and improve service levels in the era of big data, they also bring about hidden dangers such as "information partiality", Internet addiction, data leakage, and technology abuse. To what extent do these problems exist? What other solutions are there? We will explain it in the second article of the "Perspective on 'Algorithms'" series of reports, so please continue to pay attention. (People's Daily Overseas Edition)