
0

Heather Collis answered on 17 Nov 2020:
I help teach first year university students how to use Python. We teach Python because it has lots of broad applications for a mathematics degree, including data analysis.
If you’re interested in data analysis, and statistical analysis of data a bit more specifically, R is a really good language for this sort of analysis and is also commonly taught during undergraduate degrees.
The number 1 thing I like about Python and R is that they are both open source programming languages. This means they are free to use which means any work you do in these languages can easily be run by someone else on another computer (as long as they have the language installed).

Chris Budd answered on 18 Nov 2020:
There are many tools used to analyse data. The whole field of statistics for example, or the modern methods of machine learning. If you buy something on a website such as Amazon, and it suggests other things that you might like to buy, then it is likely that this prediction has been made by using machine learning to analyse the data of what you have chosen in the past. However the main tools that I use myself are those of ‘signal processing’ and in particular Fourier and Wavelet analysis. Using these tools you see how the data is made up of simpler types of function, for example sines and cosines. That tells you a lot about it. You can use these tools to remove the noise in a signal, deblur a photograph, or to find an object in an image. The workhorse of all of these methods is the Fast Fourier Transform or FFT. Your mobile phone is using the FFT all of the time to deal with the data coming from the phone signal. The FFT and other signal processing tools are all available in Python, R or other languages such as Matlab.

Alan Walker answered on 18 Nov 2020:
As mentioned by others, there are loads of statistical software available in order to help you analyse data from that point of view. I teach my students about Excel, SPSS, and R (although there are loads more such as SAS, Python, Minitab…). Each software has its own quirks which are suited to different people. For example, R is great if you’ve a mind for coding, but if you’re much more of a “clicky button” person – then SPSS is much simpler to use (but less customisable).
That being said, I’m not a statistician. A lot of my work is related to analysing waves. Here, I’ll use mathematical objects known as transforms. These allow me to transform a wave from “timespace” to “frequencyspace” allowing me to view the wave in a better context for me work. This “transforming” of data is incredibly useful in many areas of mathematics – why work in the real world when you can work in an imaginary one!

Eduard CampilloFunollet answered on 18 Nov 2020:
I choose different tools depending on what data to analyse. Just to give two examples, I teach Python to analyse data using machine learning techniques, for instance to look for patterns in the DNA. On the other hand, I teach R to do statistics on experiments results, for instance finding how often an organism mutates.

Nathan Turner answered on 18 Nov 2020:
Interestingly, when I read this question I had a different interpretation of the word ‘tools’ as I use maths in engineering a lot. The ‘tools’ of applying maths in engineering are more conceptual such as:
How to build mathematical models of systems
How to use statistics to analyse data sets (averaging, variance, modelling as distributions etc)
How to plot and visualise dataThese are different to ‘codes’ which are essential for doing complex modelling/mathematics in computing but they don’t teach you how to solve problems independently! That said, I have mostly used MATLAB & Simulink in modelling of gas turbine engines, control systems and data analytics.

Christos Klerides answered on 18 Nov 2020:
There are many tools that you can use to analyse data. They can range from simple formulas in Excel to coding in programming languages, specifically Python and R. R is specifically used for statistics whereas Python has a broad range of applications.

Chris Tognini answered on 19 Nov 2020:
Although not something I use to teach people currently, in my previous job I used Microsoft Excel quite a lot for analysing data. I had basic Excel experience from university, but found that I had to learn how to use some of its more advanced features quite quickly. It can be a pretty powerful tool that is well suited to a lot of aspects of data analysis, particularly for large datasets. You can, quite quickly, import tens of thousands of rows of data into Excel and create all sorts of useful graphs and charts, and apply all sorts of useful formulae, to your data. One project I worked on involved plotting temperature measurements on aircraft engines that were flying all over the world to measure the performance of certain parts of the engine over time.
I use a lot of programming in my current job, and my first real taste of this was learning some Visual Basic for Applications (VBA) code which works really well with Excel. That was really useful for performing routine tasks such as importing data in particular formats, performing particularly complex calculations on data and even automatically creating graphs when the raw data was imported.

Freya Addison answered on 24 Nov 2020:
Depends on what they are doing. Jupyter Notebooks is good for a short introduction to python, but a longer and more solid foundation into coding I would recommend doing a course in a compiler language. In our research group, there are specific packages that we use to view and process our radar data such as PyArt. Pandas have also become a python module favourite along with Bokeh for instant plots. We do try and move away from using spreadsheets and teach people how to read in and write csv and netcdf files. Working with Biology at the moment, they use R so there are specific statistical packages which we would use and teach from there. Image analysis is a whole other realm and what software and tools would depend on the purpose.

Comments