Skip to main content

Week 3

Week 3

Big and small data

Small data is a way of keeping the size of data down. A file such as an image could take a lot of space depending on the size of the image, but by compressing it and losing some of the quality you can massivly reduce ths size. An example of small data could be Meteorological Aviation Report (METAR). The image below shows this off, each section of the code represents a different piece of information and saves entire data entries from being typed out and takes less time to process. 
EGBB - Airfield identifier 161150Z - Time 27011KT- windspeed and direction 4000 - horizontal visibility RA - weather OVC015 - cloudbase and cover 13/11 - air temperature and dewpoint Q1003 - air pressure
https://www.futurelearn.com/courses/big-data-and-the-environment/3/steps/420387

Citizen Science

Is the collection of data about the natural world that has been gathered by the general public, usually as part of a project with other data scientists. This can be used to gain vast amounts of information quickly since many people at once can gather the information as opposed to just one scientist or one sensor. One example is Thames 21, a charity which worked with Citizen Science to create a catalog of all the types of litter which affect the flow of the river.


Visualization 

When putting data into a visual form, it is important that people are able to understand the data. Data should be visualized as useful, easy to read, and reliable. Visualization comes with many advantages and disadvantages such as:
Advantages - 
-Saves time as you don't have to read through large amounts of data
-reduces the amount of info people need to retain
-arguably more engaging as it is more expressive and illustrated

Disadvantages - 
-often only scratches the surface of how much data there actually is
-powerful visualization techniques can require extensive learning to do

Comments

Post a Comment

Popular posts from this blog

7. Limitations of traditional data analysis

7. Limitations of traditional data analysis Security: Security is foremost aspect for every technology. Big data is prone to data breaches. The important information that is provided to some third party may get leaked to customers. Proper encryption must be made in order to protect the data.  Large growth in data : data is growing faster than the processing power. Large volumes of data are being exploded in past years. We need some new machines to work; otherwise we will get over run by data. Large Data centres can solve this problem.  Inconsistencies: Sometimes the tools we use to gather big data sets are imprecise. This will happen when the data is collecting for example, consider Google search Edinburgh College results of the search on one day will be different from other day, this is mainly due to inconsistency in data collection.
Post a Comment
Read more

2. Historical development of Big Data

2. Historical development of Big Data 1980's - networks started to appear across other countries, although possible, connection was still extremely difficult without travelling, the networks had to communicate in the same language as opposed to different ones. The principle link at CERN connected the US and Europe in 1989 1990's - remote access of data in the terabytes was now easily achievable around the world. To share data more easily, the web was created so that the actually location of the data wasn't necessary to know 2000's - Big data was now so large that CERN was unable to store such a vast amount of petabytes themselves and as such had to share it between partners around the world to offload some of it. This lead to the creation of cloud computing 2005 - The term Big Data is used for the first time by Roger Mougalas, a year after the creation of the term "web 2.0" which refers to data too large to handle with traditional business methods 2009 -...
Post a Comment
Read more