The functions and advantages of data journalism

At present, international media that have accumulated experience in big data news production include the Guardian, the New York Times, and the Washington Post, but they are also in the exploratory stage. By studying the big data news practices of representative media at home and abroad, we can summarize the four functions of big data news, namely description, judgment, prediction, and information customization.

On January 5, 2012, the "Guardian" website published a big data news report about the "Arab Spring". The report uses dynamic charts and a timeline as the main line to describe a political movement that took place in 17 Arab countries in the year from December 2010 when a Tunisian man set himself on fire to December 2011. Through this four-dimensional dynamic report, netizens can clearly understand the different manifestations of the Arab Spring in different countries from the macro to the micro. There are push and pull buttons for time at the top of the chart. Netizens can push and pull to the time point they want to watch, and they can clearly see related events that happened in different countries at the same time point. At the bottom of the screen are labels for each country. Netizens can also use the country labels to follow the political evolution of a specific country on the vertical timeline. Different political events are marked with different colors: green for mass protests, light blue for relevant international reactions, yellow for political events, and red for regime change. If netizens want to know the specific content of an event, they can click on the icons of different colors to get links to in-depth reports. This method of news reporting will present the complex "Arab Spring", which involves more than a dozen countries and spans as long as a year, in a clear and dynamic way. It is difficult for pure text reports to achieve such a communication effect.

Big data journalism can also describe invisible short-term processes, such as how rumors spread on social networks. The Guardian tracked and analyzed 2.6 million tweets and used visual dynamic charts to describe the entire process from the beginning of the rumor to the end of its refutation. It also uses time as the axis, using circle sizes and color changes to describe the entire process. The green circles represent tweets that spread rumors, the red circles represent tweets that correct the rumors, the gray circles represent neutral evaluation tweets, and the yellow circles is a tweet skeptical of rumors. The size of the circle represents the degree of influence of the tweet. The larger the circle, the greater the degree of influence. If you want to know the specific content, click on which circle, and information such as the publisher of the tweet represented by the circle, the date of publication, the number of people who retweeted it, etc. will be immediately displayed next to the screen. Through this dynamic evolution process, people can clearly see that social networks are not, as generally imagined, a place for blindly spreading false news. In fact, not long after the fake news appeared, various rumors refuting the rumors appeared on social networks.

It can be seen from these two examples that the reporting method of big data news can provide a clearer and more comprehensive view of an event at a macro level, the complex evolution process of the event, and all aspects of this process. All described intuitively and interestingly. In August 2011, a black Muslim man was stopped by the police while taking a taxi on the streets of London. A shootout broke out and the man died on the street. Two days later, about 300 people gathered at a police station in central London for a protest that turned into days of rioting, with protesters setting cars, shops and buses on fire. That night, similar attacks on police officers, robberies, and arson occurred in other areas of London. Some media comments pointed out that this is related to the gap between rich and poor. In an interview, British Prime Minister Cameron claimed that the riots had nothing to do with the gap between rich and poor.

The reporter of the British "Guardian" used the analysis results of big data to make a series of reports on this incident. One of the topics of the reports was whether there is a connection between the riots and poverty. The reporter used Google Fusion Chart to mark the residence information of the rioters (yellow dots), the actual locations of the riots (gray dots), and the distribution of poor areas (the more red, the poorer) on the map of the London area. According to this map of central London, netizens can expand the map to see the entire Greater London area, or they can focus on specific neighborhoods and zoom in to observe where the flow of people at each marked riot point comes from and where they go. , thus clearly seeing a certain connection between poverty and riots. The expression of this relationship is clearer and more convincing than a simple text report. During the National Day holiday in 2013, a large number of tourists were stranded in Jiuzhaigou and triggered mass incidents. If the news media or tourism authorities could have used China's local big data for predictive reporting before, such mass incidents could have been completely avoided. Because the media can use this big data to report in advance how many people are going to Jiuzhaigou from which places during a specific time period, including how many men, women, elderly, children, etc. there are.

This is just a small example. Big data can predict all aspects of society and people's daily lives. By mining big data, the media can technically produce visual and interactive charts to inform many matters. Microscopic ones, such as the onset of epidemic diseases and traffic congestion; macroscopic ones, such as changes in economic indexes, the coming of some kind of social crisis, etc. Baidu has opened a "Baidu Forecast" webpage, launched with the slogan "Big Data, Know the World". The predicted products include college entrance examination, World Cup, movie box office and so on. The products they plan to launch later have expanded to a wider range of fields, such as financial forecasts, real estate forecasts, etc.

Using the analysis results of big data to meet the personalized information requirements of netizens is the latest attempt by foreign media. For example, the Five Thirty Eight data blog launched a new column for readers' letters "Dear Mona" on May 23, 2014. The purpose explained in the opening sentence of the first issue is: "I started this column to help readers answer some important or serious questions in life, such as whether I am normal, where I am in the world, etc. The purpose It’s not to answer readers’ questions or tell them what they should and shouldn’t do. On the contrary, I provide data to explain and describe your experience.”

Looking at this column, readers have asked a variety of questions. , more serious ones such as: "How many people in the United States have never had a drop of alcohol?" "How many male flight attendants are there in the United States?" There are also more personal ones such as: "How often should I change my socks?" "Will living together before marriage cause Divorce” and so on. The columnist uses big data across the United States to immediately inform the parties of the analysis results, but avoids giving guiding opinions and only informs the analysis results of various data, allowing netizens to deal with the problems they face based on the analysis results. This column is different from the traditional column for letters from readers of print media. It does not provide some chicken soup-style answers through zodiac signs, blood types, birth dates or pretending to be experienced experts. It only uses data to speak.

Such attempts are not uncommon in the media. In 2011, the BBC broadcaster worked with KPMG to create a budget calculator based on the 2012 government budget. Users only need to enter some daily information, such as how much beer to buy, how much gasoline to use, etc., to calculate the new budget. How much tax will you pay, and will your life be better next year?

Providing personalized big data services based on user needs is the future development trend. These reports have a certain nature. The media are committed to focusing on the needs of users, using big data to explain the impact of macro-social phenomena on users, or to answer users' confusing questions. The media can be accurately positioned and, through background calculations, push services to users based on their reception habits, work habits and living habits.