eanbit-rt / mini-projects-2022

For the Residential Training Mini-projects
0 stars 2 forks source link

Project 7: Reproducing a Machine Learning paper #7

Open kipkurui opened 2 years ago

brendamuthonikamau commented 2 years ago

I choose this project as my first option because:

  1. I have an interest in machine learning, thus I have taken time to read and watch tutorials on machine learning this project will give me an opportunity to apply the little knowledge and expand my knowledge in machine learning
  2. I would like to gain skills in machine learning and understand how it can be used to solve real life problems and this project offers that.
  3. I am curious to understand how different machine learning models compares when analyzing the same data
  4. Since this project aims at reproducing a paper, it will give me an opportunity to appreciate the concept of reproducible research
fetche-lab commented 2 years ago

MY FIRST OPTION: Reasons to select this as my first option: • Am interested to acquire skills related to machine learning and its applicability in biological data science and research. • I have little exposure to this discipline. I therefore take this as a challenge to extend and diverse my data analytics skills as a growing bioinformatics scientist. • An added advantage to my skill-set and helpful in the next upcoming potential research projects and other related projects that will be entrusted to my expertise.

bonfaceonyango commented 2 years ago

First choice:

Reasons:

  1. I developed interest in machine learning through participation in kaggle,Zindi Africa ML projects for beginners and tutorial courses. Doing this project will therefore help me to enhance and gain new skills in using machine learning to solve Bioinformatics problems, for this case,ancestry inference
  2. I will explore how to conduct a reproducible machine learning project
  3. Interest to understand machine learning algorithms and algorithm selection for Bioinformatics research
  4. Desire to be future machine learning expert. With increasing volume of NGS data, there is need for future Machine learning experts for automated omics data analysis
fredrickkebaso commented 2 years ago

Second choice.

Reasons.

  1. I have had a grasp of various machine learning aspects and their application to human genetics research through various online platforms and pieces of training. This project presents a noble opportunity for me to apply as well as advance these skills.

  2. My curiosity and interest in Machine Learning and Artificial Intelligence present this project as a platform for me to explore more in these cutting-edge fields. Undertaking this mini-project will therefore play a great impact on the main project that I will undertake from September.

VioletChege commented 2 years ago

First Choice

My reasons:

  1. With my interest and background in statistics my niche is data science, therefore upskilling with machine learning will be an opportunity for me to grow in my path.
  2. Machine learning is extremely general, very useful and newly applicable and because of it generality I am curious to see its application in forensics genetics analysis.
  3. I am interested to learn the various machine learning algorithms and see how traditional statistics methodologies have been automated.
sephoh commented 2 years ago

This my second choice: Reasons 1) I have skills in machine learning in R language. I would like to extend my knowledge using python language by inter-operating between the language for reproducible research. 2) Machine learning skills allows for detailed interrogation of data. This project impress me because it will help me make the data speak to tell a statistical story. 3) ML and AI allows for automation and generation of information from data. This would allow me to analyse biological dataset in less laborious workflow

totodingi commented 2 years ago

Third Choice

For this project, It would be interesting to explore the capabilities of the deep learning flavor of machine learning in arriving at the same conclusions as in the publication. I would be keen in benchmarking the performance of neural networks with the classical ML algorithms used in the publication.