Personal Information Security in the Big Data Era Series III: "Friends of Likes"
The Internet is like a road, users will leave footprints when they use it.
Everyone is generating data all the time, and while consuming data, they are also being consumed by data.
Recently, the news that a college graduate stole the school's intranet data and collected the personal privacy information of all students in the school has aroused people's renewed attention to the issue of personal information security in the era of big data. In the era of big data, recommendation algorithms and AIGC pose new challenges to personal information security.
1. Friends with likes
"Friends with likes" refers to the exchange of likes on social media. Some people like to like their favorite information or blog posts on Weibo and Moments. Little do we know that such a small move will lead to our private information being snooped on.
In 2012, Kosinski demonstrated that based on an average of 68 "likes" on Facebook, it was possible to predict a user's skin color (95 percent accuracy), sexual orientation (88 percent accuracy), and political leaning (Democratic or Republican, 85 percent accuracy). Predictable content goes far beyond that and includes intelligence, religious affiliation, and alcohol, cigarette, and drug use. It is even possible to infer whether someone's parents are divorced.
After continuous research and improvement, its model has become more and more perfect:
- Can rate subjects more accurately than their colleagues based on just 10 likes;
- 70 "likes" are enough to know the subject better than the subject's friends;
- 150 likes can understand the subject better than the subject's parents;
- 300 likes allow the subject's partner to know more about the subject.
Based on more likes, Kosinski knew the subject even better than the subject himself.
A very simple method was later developed by Kosinski et al. First, they provided the test subjects with questionnaires in the form of an online quiz. Based on their responses, psychologists calculated the subjects' personal "Big Five" personality scores (OCEAN: Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism). Kosinski's team then compared the results with the subjects' other online data, such as data on what they "liked," shared or posted on Facebook, as well as data on gender, age, where they lived, and more, allowing researchers to establish associations between specific online behaviors and personality traits.
In 2016, Trump and Hillary Clinton’s presidential election staged a "data war". The Trump team signed the "Cambridge Analytica" (Cambridge Analytica, CA) big data company and hired the company's CEO Alexander Nix (Alexander Nix) as the head of digital strategy for the campaign.
Cambridge Analytica used the "This is Your Digital Life" personality test tool developed by itself, distributed questionnaires through Facebook, and obtained 50 million user data. By analyzing the user's age, gender, race, address, phone number, email address, personal preferences, family status, activity range, circle of friends and other personal privacy data, Cambridge Analytica established a psychological model, tapped group preferences, analyzed user personality, and predicted political positions. This trend ultimately affected the US election and helped Trump succeed in the campaign.
2. Like scam
Routine one:
Criminals pretend to be merchants and post “Likes and Win Prizes” information in WeChat groups or Moments, claiming that high-end items will be given away for free when a certain amount is collected. The victim only needs to pay the postage, and asks the participants to send their personal information such as their name, phone number, and address to the WeChat background.
After obtaining the information, they will commit fraud in the form of "handling fee", "notarization fee", "guarantee deposit", or send defective products and collect the goods fee by express delivery. When the victim went back to investigate, the links or official accounts of "Like, Collect Likes" all disappeared.
Routine two:
Criminals use the name of "Like" to induce victims to download APP software containing Trojan horse viruses or click on links, causing the victim's mobile phone to be poisoned, stealing the victim's mobile phone address book to commit fraud, or directly transfer the victim's property.
3. Personal Information Protection Tips
You can't believe these advertisements in Moments! ! !
- Be cautious about downloading links or software from unknown sources or sent in the name of activities, and be careful of being implanted with a Trojan horse program that will cause your account to be lost.
- Carefully identify formal WeChat public platforms and sponsors, and be wary of various types of activities such as likes and votes;
- Be vigilant when filling in personal information, do not involve bank card numbers, account numbers and other information, and try to avoid filling in detailed information.
The key is to remember that there will be no pies in the sky, don't be greedy for petty gains, and live with peace of mind.