The international free version of the new crown epidemic situation data analysis APP is officially released!

Posted May 27, 20206 min read

Introduction


In early February of this year, SLS has released an APP for domestic dynamic display analysis of the new coronavirus pneumonia epidemic. Currently, this capability is fully open to governments, communities and third parties The platform and openers are widely used and are completely free to open. Students who have n t followed can learn about the background through the following links:

Recently, with the outbreak of the new coronavirus pneumonia epidemic in the world, SLS has launched an analysis market that tracks the global epidemic situation. Compared with the domestic market, which mainly focuses on domestic epidemic situations(data from CCTV News, People s Daily, and announcements of provincial and municipal health and health committees), the international market for epidemic situations tracks the global epidemic situation, and the data sources are widely cited by the international community Johns Hopkins University Open Source Dataset .

SLS


Alibaba Cloud Log Service(SLS) is a one-stop service for log data. It can quickly complete the collection, consumption, delivery and query analysis of massive log data without development, improving operation and maintenance and operational efficiency. Log service mainly includes real-time collection and consumption, data delivery, query and real-time analysis and other functions, suitable for various development, operation and maintenance, operation and security scenarios from real-time monitoring to data warehouse.

As the log analysis center, the log service provides one-stop data collection, processing, query analysis, AI calculation, visualization, and supports interconnection.

Highlights


1 . Provide regular epidemic situation data and update it regularly and daily


SLS has collected and regulated the epidemic-related data, updated it regularly every day, and formed a visual platform to cover the epidemic information of various countries/regions, provinces/states around the world. You only need to focus on the analysis and display of data, and other tedious details SLS have been dealt with.

2 . Pre-defined rich data disk, and support customization


Built-in multiple copies of data disks and support for customized provision of epidemic situation in various countries/regions, provinces/states around the world. Support interactive query analysis, custom reports, deep drilling and alarms.

Overview of the global epidemic situation

Outbreak details by country

3 . Open data platform and interconnection


The data platform is open, and the interconnection log service is open, and it can be docked with a large number of other environment systems, third-party applications, or open source. Provides easily expandable data analysis, storage, and visualization platform capabilities, such as DataV, Blink, OSS, stream computing, Grafana, SOC, etc.

4 . Completely free


Completely free epidemic service application and related resource data, including dashboard, alarm and other functions are completely free.

data


Import and organize


There may be some students who only focus on the data analysis of the COVID-19 epidemic and want to try it out, but they do not know how to obtain and process data from various data sources, or they are not very skilled in SLS SQL. In order to help these students realize data analysis conveniently and quickly, SLS has collected and regulated the relevant data of the epidemic situation, and it is updated regularly every day. You only need to focus on the analysis and display of data, and other tedious details SLS have been dealt with.

Sample data


type:Country/Region Cases
version:v2020-04-17T11:55:36
Last Update:2020-04-09 01:12:20
Country/Region:China
Country/Region(ch):China
LatLng:35.000074,104.999927
Confirmed:83798
Confirmed Hist:\ [644, 923, 1409, 2079, 2882 ]
Confirmed Trend:{"2020-01-23":644, "2020-01-24":923, "2020-01-25":1409, "2020-01-26":2079, "2020-01-27 ":2882}
New Confirmed Hist:\ [95, 279, 486, 670, 803 ]
New Confirmed Trend:{"2020-01-23":95, "2020-01-24":279, "2020-01-25":486, "2020-01-26":670, "2020-01- 27 ":803}
Deaths:3352
Deaths Hist:\ [18, 26, 42, 56, 82 ]
Deaths Trend:{"2020-01-23":18, "2020-01-24":26, "2020-01-25":42, "2020-01-26":56, "2020-01-27 ":82}
Recovered:78556
Recovered Hist:\ [30, 36, 39, 49, 58 ]
Recovered Trend:{"2020-01-23":30, "2020-01-24":36, "2020-01-25":39, "2020-01-26":49, "2020-01-27 ":58}

Data Format


Various epidemic-related data are placed in a log library ncp, which is distinguished by the field type as type:Global Cases, Country/Region Cases and Province/State Cases. Mark the data version through the version field, and each version will contain complete data, which is used to correct the data. The list of data fields is as follows:

Field name description Sample type Data type Global Cases, Country/Region Cases or Province/State Casesversion data version v2020-01-26T12:30:00Last Update latest source news release time 2020-01-26 18:23Confirmed latest confirmed cases cumulative data 1058Confirmed Hist cumulative data of confirmed cases(from 2020.01.23 to the current historical data array) \ [270, 444, 444, 549, 729, 1058 ]Confirmed Trend cumulative data of confirmed cases(from 2020.01.23 to the current historical trend data Dictionary) {"2020-01-21":1, "2020-01-22":1, "2020-01-23":1, "2020-01-24":2, "2020-01-25" :2, "2020-01-26":3} Recovered cumulative data of the latest cured cases 42Recovered Hist cumulative data of the cured cases(from 2020.01.23 to the current historical data array) \ [0, 28, 28, 31, 32, 42 ]Recovered Trend cumulative data of cured cases(from 2020.01.23 to the current historical trend data dictionary) {"2020-01-21":1, "2020-01-22":1, "2020-01-23":1, "2020-01-24":2, "2020-01-25":2, "2020-01-26":3} Cumulative data of the latest deaths of Deaths 52 Cumulative data of the deaths of Deaths Hist(from 2020.01.23 to Current historical data array) \ [3, 17, 17, 24, 39, 52 ]Deaths Trend cumulative death case data(from 2020.01.23 to the current historical trend data dictionary) {"2020-01-21":1 , "2020-01-22":1, "2020-01-23":1, "2020-01-24":2, "2020-01-25":2, "2020-01-26":3 } New Confirmed Hist existing data of suspected cases(from 2020.01.23 to the current historical data array) \ [11, 0, 41, 0, 56, 127 ]New Confirmed Trend existing data of suspected cases(from 2020.01.23 to Current historical trend data dictionary) {"2020-01 -21 ":1," 2020-01-22 ":1," 2020-01-23 ":1," 2020-01-24 ":2," 2020-01-25 ":2," 2020-01 -26 ":7}

Analysis and display


SLS provides large-scale log real-time query and analysis capability, which has the following advantages:

  • Real-time:Can be analyzed immediately after writing.
  • Fast:Within one second, query(5 conditions) can process 1 billion level data, and analysis(5 dimension aggregation + GroupBy) can aggregate 100 million level data.
  • Flexible:You can change any query and analysis conditions to obtain results in real time.
  • Eco-rich:In addition to the reports, dashboards, quick analysis and other functions provided by the console, it can also seamlessly interface with Grafana, DataV, Jaeger and other products, and supports protocols such as Restful API and JDBC.

As mentioned above, the outbreak data provided by SLS uses the version field to mark the data version. In order to query the latest version of the data, you can use the following SQL:

type:"Province/State Cases" | select .... from log l right join(select max(version) as version from log) r on l.version = r.version

In the predefined epidemic market, each dashboard corresponds to a SQL analysis. Taking Global Cases Trend as an example, in order to analyze the global cumulative diagnosis, death, cure, and existing case development trends, we can use the following SQL query and save the result icon as a dashboard, which is convenient and fast.

type:"Global Cases" | select date \ _format(date \ _parse(la, '%Y-%m-%d'), '%b%e') as "Date", lb as "Confirmed", lb- r2.b-r6.b as "Active Confirmed", r2.b as "Deaths", r6.b as "Recovered" from(select a, b from log l right join(select max(version) as version from log) r on l.version = r.version, unnest(cast(json \ _parse("Confirmed Trend") as map(varchar, bigint))) as t(a, b)) l left join(select a, b from log l right join(select max(version) as version from log) r on l.version = r.version, unnest(cast(json \ _parse("New Confirmed Trend") as map(varchar, bigint))) as t(a, b)) r on la = ra left join(select a, b from log l right join(select max(version) as version from log) r on l.version = r.version, unnest(cast(json \ _parse("Deaths Trend") as map(varchar, bigint))) as t(a, b)) r2 on la = r2.a left join(select a, b from log l right join(select max(version) as version from log) r on l.version = r.version, unnest(cast(json \ _parse("Recovered Trend") as map(varchar, bigin t))) as t(a, b)) r6 on l.a = r6.a order by l.a

Instructions


Log in to Alibaba Cloud Log Service Console, you can see the application of epidemic analysis:

Click to enter for the first configuration(one time, the subsequent data will be automatically synchronized), and then you can directly use the multiple data disks provided by the log service to start the interactive analysis and visualization journey.