Difference between revisions of "Step 7"
From Research management course
| Line 33: | Line 33: | ||
# Look through the [https://github.com/intsystems/m1p/tree/main-2024 table of reports] and see [https://docs.google.com/document/d/1aULkuCEMExH4iB0-DIXj6UqetBJFnIXrngfDCCtldUY/edit?tab=t.0#heading=h.p8tjyih99h9i some particular] LinkReview files. | # Look through the [https://github.com/intsystems/m1p/tree/main-2024 table of reports] and see [https://docs.google.com/document/d/1aULkuCEMExH4iB0-DIXj6UqetBJFnIXrngfDCCtldUY/edit?tab=t.0#heading=h.p8tjyih99h9i some particular] LinkReview files. | ||
# Examples of the comparative tables | # Examples of the comparative tables | ||
| − | ## Generation of simple structured IR functions by genetic algorithm | + | ## [https://m1p.org/papers/Kulunchakov2014RankingBySimpleFun.pdf Generation of simple structured IR functions by genetic algorithm without stagnation] |
| − | without stagnation [https://m1p.org/papers/ | + | ## [https://m1p.org/papers/MotrenkoStrijov2017ECoG_HL_2.pdfMulti-way Feature Selection for ECoG-based Brain-Computer Interface] |
| − | |||
| − | Interface | ||
# See parts of Introduction (to put it in the LinkReview as a draft) | # See parts of Introduction (to put it in the LinkReview as a draft) | ||
## [https://m1p.org/papers/Katrutsa2014TestGenerationEn.pdf Related works:] Stresstest procedures for feature selection algorithms | ## [https://m1p.org/papers/Katrutsa2014TestGenerationEn.pdf Related works:] Stresstest procedures for feature selection algorithms | ||
## [https://m1p.org/papers/Katrutsa2016QPFeatureSelection.pdf The main contributions of this paper are:] Comprehensive study of feature selection methods to solve multicollinearity problem according to evaluation criteria | ## [https://m1p.org/papers/Katrutsa2016QPFeatureSelection.pdf The main contributions of this paper are:] Comprehensive study of feature selection methods to solve multicollinearity problem according to evaluation criteria | ||
| + | # For the future materials see [https://sourceforge.net/p/mlalgorithms/code/HEAD/tree/Group074/Kuznetsov2013SSAForecasting/doc/Ivkin2013ProblemStatement.pdf?format=raw a problem statement] (beware the spherical horse in a vacuum) | ||
| + | <!-- https://sourceforge.net/p/mlalgorithms/code/HEAD/tree/Group074/Kuznetsov2013SSAForecasting/ --> | ||
Revision as of 02:16, 16 November 2024
"If I have seen further, it is because I have stood on the shoulders of giants", – Isaac Newton wrote this in a letter in 1675.
We establish a solid foundation for our research to deliver results of better quality and higher impact with less effort. This foundation comprises essentials, overviews, state-of-the-art, and alternatives to our solution. Not to mention the supplementary materials like references to the algorithms and data in your computational experiment.
Contents
The seminar
- [The warm-up 5-minute test]
- The model selection problem statement
- The Bayesian inference
- How to read papers and gather references
- Buidling the comparative table
Resources
Step 7 YouTube []
Homework
Each solution to your problem has its pros and cons. List them all.
- Collect references for your project according to the plan.
- While collecting put your notes to the temporary file LinkReview (see examples below) or to the .bib and .tex
- Put them into the .bib using JabRef or another tool that generates the bibliographic record using doi.
- Analyse each solution's pros and cons according to the project's quality criteria.
- Put the short version of your analysis in the table.
- Fill in the table in your repository.
- Put the link to your analysis here, Step-7.
While completing the table, remember the formula for an engineering project description: "We propose a solution that offers a unique feature, distinguishing it from other solutions." Of course, each part needs an explanation.
Note: The whole section Introduction is not required. (Ask me why)
Examples
- Look through the table of reports and see some particular LinkReview files.
- Examples of the comparative tables
- See parts of Introduction (to put it in the LinkReview as a draft)
- Related works: Stresstest procedures for feature selection algorithms
- The main contributions of this paper are: Comprehensive study of feature selection methods to solve multicollinearity problem according to evaluation criteria
- For the future materials see a problem statement (beware the spherical horse in a vacuum)