Butterfly Brand Knife,Real Butterfly Knifes,Practice Butterfly Knifes,Butterfly Tool Knife YangJiang Junxiong Trading & Industry Company Co.,Ltd , https://www.cn-sundo.com
The smart card public transportation ticketing system allows people to ride freely between public transport, subway, tram, train and ship - even if these are operated by different companies - without buying tickets. This is very convenient, but it also has adverse consequences. One of them is that anyone with access can accurately research where people are going by tracking passengers (or their transportation cards).
Companies use this information to optimize services. Of course, this is also a good thing. But there are many other things that can be done using smart card data, some of which can be very frustrating for people who love freedom. A few days ago, the Knowledge Discovery and Data Mining Conference (KDD) held in San Francisco outlined one of these things, which on the surface seemed completely harmless. This is the use of this data to grab a hand.
The idea was thought by Bear Hui of Rutgers University, Du Bowen of Beijing University of Aeronautics and Astronautics, and Hou Zhenshan. They studied the movement of passengers on buses, trains and subways in Beijing. As expected, most passengers move from A to B quickly—the least amount of time or the least number of transfers, and the same itinerary every day. However, there are a small number of people who do not have any regularity in their route or suddenly change their riding mode.
Many of these anomalies can be explained simply: forgetting to take a briefcase or being unfamiliar with some routes. But sometimes, the reason may be malicious: pickpockets carry out plagiarism on the transport network, and may use stolen transport cards.
Fortunately, there are few handcuffs. However, it also made it difficult for them to be found. Dr. Xiong Hui uses a system that includes two steps. First, a computer categorization program looks at the 6 million trips between April and June 2014 in Beijing and then separates outliers from ordinary passengers. The second categorizer pre-fetches hotspot information collected from police station reports and social media messages, and then attempts to identify pickpockets from among these outliers.
In this regard, the study was successful. It identified 93% of known knockouts (ie, pickpockets caught by the police during this period). However, the second goal is to identify the suspects as accurately as possible. This time, its performance is equivocal. Only one out of every 14 suspects is known to be a pickpocket. On the other hand, the data also includes some unknown players.
Although the miscarriage rate is so high, Dr. Xiong Hui believes that he has developed a powerful tool. Using a closed-circuit camera to monitor suspicious few people is far worse than tracking millions of passengers. He said that this technology will soon be piloted in Beijing and subsequently used in other cities in China.
Not all experts believe this technology. Shashi Verma is the CTO of Transport for London and is responsible for the smooth running of the Oyster card system throughout the city. He said that according to his records, millions of ordinary people will make a variety of "weird, wonderful and complicated" behaviors. It is not as easy as it sounds to find a handful of criminals from among innocent passengers. However, Dr. Xiong Hui strongly believes in his team's approach. They also recommend investigating other "anti-social groups" such as the "mobile drinkers, drug addicts, tramps, and drug traffickers" in the public transportation network. Grabbing a pickpocket is one thing. However, using artificial intelligence to track those who are on the edge of society is another matter. Technology does not know the difference between the two, but people need to know.
(Original title: Beijing will use bus card data to catch thieves with an accuracy rate of 93%)
KDD2016, which is just over the scientific and technological developments , Xiong Hui of Rutgers University, Du Bowen and Hou Zhenshan of Beijing University of Aeronautics and Astronautics published their papers. They studied the movement of passengers on buses, trains, and subways in Beijing, analyzed smart bus card data, combined police information published on social networks, and found out pickpockets. The accuracy rate in identifying criminals was 93%. The technology will be piloted in Beijing, followed by other cities.