Another big data achievement of “war epidemic” of Baidu map was selected into the Academic Summit KDD 2020

Novel coronavirus pneumonia novel coronavirus pneumonia epidemic has become a new normal in

, and the international epidemic is continuing to spread. Although China is the earliest country affected by the epidemic, the new crown pneumonia epidemic situation is generally controlled. In this war epidemic, the high-tech represented by Baidu AI has continuously contributed its strength. As a new generation of artificial intelligence map, baidu map has successively launched “fever clinic map” and “epidemic area” thematic map and other functional services during the epidemic period, helping to prevent and control the epidemic situation and return to work. < / P > < p > many measures to prevent and control the epidemic situation have achieved remarkable results, but at the same time, they also bring changes to the public travel mode. Recently, baidu maps published a paper, through the analysis of massive travel data during the epidemic period, the new mode of public travel under the epidemic situation was obtained from five different perspectives, including public travel mode and destination type, and provided beneficial suggestions for epidemic prevention and control from the perspective of travel. At the same time, the paper was selected into the top Data Mining Conference KDD 2020 to share the “China experience” with the international community. During the epidemic period, the proportion of public transport, driving and cycling changed significantly. According to the data, in the first four months of 2020, the proportion of public transport will drop to 45%, and in February, when the epidemic situation is serious, the proportion will drop to 40%. However, according to the data of the same period in 2018 and 2019, the average proportion of public transport will be 54%. At the same time, the proportion of using bicycle and driving increased relatively: compared with the average value in the same period of 2018 and 2019, the proportion of bicycle increased by 5.25% in the first four months of 2020, while the proportion of driving increased by 3.38%. Baidu map big data thinks: during the epidemic period, the public can be encouraged to use bicycles more, and some bicycle lanes should be planned temporarily to facilitate public travel, and staggered peak travel should be encouraged. Through the data analysis of different travel destinations, the home isolation policy during the epidemic period makes the densely populated residential areas become the main destinations. The data shows that in the first four months of 2020, the proportion of residents visiting and transportation facilities is significantly different: the proportion of visiting residential areas has increased significantly to 31.25%, while the proportion of transportation facilities as a destination has decreased to 19%. In addition, in the first four months of 2020, the proportion of visits to hospitals and pharmacies has increased to a certain extent compared with the same period of the same period; however, the proportion of visits to restaurants, educational institutions and other places with different degrees of personnel aggregation has significantly decreased. In this regard, baidu map big data proposed that for the industry which is greatly affected by the epidemic situation, we can focus on the economic recovery policy after the epidemic. The weekend is no longer the only travel time choice due to the impact of epidemic shutdown and home office work. According to the statistics of travel time with weekly cycle, it is found that the public prefer to travel on Wednesdays and Thursdays during the epidemic period, instead of “gathering” on weekends. The distribution of travel time is very consistent with that of the public in the four months before 2018. In the first four months of 2020, the public prefer to travel on Wednesdays or Thursdays. The changes of travel peak during the epidemic period can provide reference for the formulation of epidemic prevention and control measures. < / P > < p > travel distance is an important factor affecting population mobility, and the farther the travel distance is, the more likely it is to bring potential risks to epidemic prevention and control. Through the statistics of public travel distance data, it is found that the proportion of long-distance travel increased in the early stage of the outbreak of the epidemic, while with the further implementation of the epidemic prevention and control measures, the long-distance travel decreased significantly. According to the data, in February 2020, compared with the same period in previous years, the proportion of long-distance travel increased significantly, about 2%. Further analysis shows that this is mainly due to the increase in the proportion of trips with a distance of more than 30 km in the month. With the gradual introduction of epidemic prevention and control measures, the proportion of long-distance travel decreased significantly in March and April 2020. The most significant change during the epidemic period is that the departure or destination of the most frequent OTD mode has changed from transportation facilities to residential areas. Combined with the data analysis in 2018 and 2019, educational institutions, hotels and other destinations were seriously affected by the epidemic, and the top-5 model of OTD was dropped out and replaced by supermarkets, markets and workplaces. Based on the above data and prevention and control experience, baidu map big data believes that it is necessary to focus on residential areas, supermarkets, markets, hospitals and other places, and suggests that quarantine measures should be taken at the transportation hubs to these places. < p > < p > through the analysis of massive spatiotemporal big data of Baidu map, this paper reveals the impact of Xinguan epidemic on public travel mode, and concludes that under the influence of the epidemic, public travel is concentrated in the middle of the week, and the destination is concentrated in residential areas and work places. Data mining technology is used to provide data level support for epidemic prevention and control. < / P > < p > this paper has been selected into the Academic Summit KDD 2020 – AI for covid through fierce competition. ACM SIGKDD International Conference on data mining and knowledge discovery is the top academic conference in the field of data mining. It is known as “World Cup” in data mining field. It is one of the most influential and largest international top conferences in AI field. It is reported that the selection rate of this KDD paper is extremely strict, and the reception rate of research track and applied data Science track are only 16.9% and 16.0%. In addition to this paper, Baidu has 9 successful papers selected, covering intelligent transportation, intelligent recommendation, knowledge map, scientific epidemic prevention and other fields. Baidu has also become one of the companies with the largest number of selected papers in this KDD global technology enterprises. As an important technical force of the first echelon in the field of artificial intelligence in the world, Baidu has also made outstanding contributions to China’s AI war epidemic. During the outbreak of the epidemic, baidu map launched a set of anti epidemic “combination fists”, which not only provided residents with information aggregation of the epidemic situation, but also opened a special green channel for the majority of developers to help relevant products such as epidemic statistics, protection and living security online. In the future, Baidu will continue to share the experience of epidemic prevention and control, help more countries and regions find solutions from the technical level, benefit more people, and make AI technology play a greater value. Continue ReadingAmerican companies begin to give up R & D: who should pay for corporate research?