Guido Mascia

Putting things in order is always a good thing. K-Mean #Clustering is one of the many methods available. It is something I found really interesting mostly because it is conceptually easy. Have a look if you are interested! guidomascia.blog/sorting-things…

This approach allows us to quantify how strong is the relationship between each feature and the target outcome. Although easy, it is very helpful, so maybe you want to have a look! guidomascia.blog/unraveling-the…

Most of the methods I found in literature for detecting gait events were excruciatingly convoluted. However, we had the opportunity of developing a stupidly easy method for consistently extracting them from a foot-mounted gyroscope. Have a look here! guidomascia.blog/stupidly-easy-…

A brief explanation on why one should split the #dataset when developing a #MachineLearning model. guidomascia.blog/why-do-we-need…

A brief description on how to easily #normalize a variable, and on which are the implications of not doing it in #MachineLearning. guidomascia.blog/why-normality-…

Can you estimate peak jump height accurately using a #Smarthpone? With the aid of #MachineLearning you can! Check out our new research paper available on @FrontiersIn @BeatriceDeLazz1 @VCamomilla doi.org/10.3389/fspor.…

Testing runners! Here our current project in collaboration with #diadora to evaluate footware comfort perception and running biomechanics #running #ergonomics #biomechanics

I added a little entropy to the scientific production. Really proud of my first paper as main author! It was a long journey, but it was worth it. Check it out at following the doi: 10.1016/j.jbiomech.2022.111202 @BiomechFOROIT @VCamomilla

Really happy for participating to the 7th #BoHNes Colloquium in 🇪🇦 #Valencia 🇪🇦. Sharing knowledge and expertises makes everybody a better person. To the next one! @BiomechFOROIT

Testing Friday 🔍 Developing MLR to predict #jump parameters from #wearable. Moving from lab to the field, but with proper statistical cover-up🏃🏻‍♂️ #sportbiomech #inertialsensor

Proud to be part of an exciting ISBS2021 conference. I presented a work (t.ly/PvsX) stressing out the limitations brought by IMUs and correcting for them, using Multiple Linear Regression models. More content available at GitHub repo (t.ly/odYE).